The Passions in Play PDF

This page intentionally left blank
T H E PA S S I O N S I N P L AY
Thyestes and the Dynamics of Senecan Drama
This is the first monograph in English devoted to the most important of Senecas tragedies, Thyestes, which has had a notable influence
on Western drama from Shakespeare to Antonin Artaud. Thyestes
emerges as the mastertext of Silver Latin poetry, and as an original
reflection on the nature of theatre comparable to Euripides Bacchae.
The book analyses the complex structure of the play, its main themes,
the relationship between Senecas vibrant style and his obsession with
dark issues of revenge and regression. Substantial discussion of other
plays especially Trojan Women, Oedipus and Medea permits a
comprehensive re-evaluation of Senecas poetics and its pivotal role in
post-Virgilian literature. Topics explored include the relationship between Senecas plays and his theory of the emotions, the connection
between poetic inspiration and the underworld, and Senecas treatment of time, which, in a perspective informed by psychoanalysis, is
seen as a central preoccupation of Senecan tragedy.
a l e s s a n d ro s c h i e s a ro is Professor of Latin Language and
Literature at Kings College in the University of London, having
previously taught at Princeton University and the University of
WisconsinMadison. He has published widely on Latin literature,
including Simulacrum et imago (1990) and co-editing, with Phillip
Mitsis and Jenny Strauss Clay, Mega nepios: il destinatario nellepos
didascalico (1993) and, with Thomas Habinek, The Roman Cultural
Revolution (1997).
THE PASSIONS IN PLAY

Thyestes and the Dynamics of Senecan Drama
ALESSANDRO SCHIESARO

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, So Paulo
Cambridge University Press
The Edinburgh Building, Cambridge , United Kingdom
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521818018
Alessandro Schiesaro 2003
This book is in copyright. Subject to statutory exception and to the provision of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.
First published in print format 2003
-
isbn-13 978-0-511-06125-7 eBook (NetLibrary)
-
isbn-10 0-511-06125-0 eBook (NetLibrary)
-
isbn-13 978-0-521-81801-8 hardback
-
hardback
isbn-10 0-521-81801-X
Cambridge University Press has no responsibility for the persistence or accuracy of

s for external or third-party internet websites referred to in this book, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
per mia madre

e in memoria di mio padre
Flectere si nequeo superos, Acheronta movebo.

S. Freud, The Interpretation of Dreams
Contents
Acknowledgements
Note on translations
page ix
xi
Introduction
Poetry, passions and knowledge
Staging Thyestes
26
The poetics of furor

Tantalus tongue
Framing Thyestes
Tragedy, terminable and interminable
A craftier Tereus
70
Thracium nefas
Crime, ritual and poetry
The logic of crime
Perfection, of a kind
26
36
45
61
70
85
98
117
Atreus rex
139
Non quis, sed uter

De clementia
139
151
Fata se vertunt retro
177
The poetics of passions
221
Intertextuality and its discontents

Passions and hermeneutics: the audience
Allegories of spectatorship
The challenge of epos
Epilogue
221
228
235
243
252
Bibliography
Index of passages cited
General index
256
269
281
vii
Acknowledgements
This book has been, alas, very long in the making and has also accumulated
a large number of debts, both personal and institutional. It was begun in
the idyllic surroundings of the Classics Department at Princeton, where
it was fostered by much material support, but especially by the stimulating friendship of Josh Ober, Froma Zeitlin and Richard Martin, and,
not far from East Pyne, that of Glen Bowersock, Adrienne Mayor and
Daniel Mendelsohn. I also remember with gratitude the brilliant students
of my graduate seminars, and the help I received on several occasions from
excellent research assistants.
In London, help at crucial junctures has come from John Henderson
and Victoria Rimell. My colleagues Carlotta Dionisotti, Ingo Gildenhard,
Roland Mayer and Michael Silk have been a great source of learning and
friendship. The anonymous readers for Cambridge University Press have
offered much appreciated criticism and advice. Michael Sharp has been a
very supportive and patient editor. My thanks to them all.
Sadly, Don Fowler can only be thanked in memoriam for all he has done
for this book and its author.
The book incorporates, in a revised form, material that has previously appeared in Vergilius 38 (1992); Materiali e Discussioni 39 (1997); J. Elsner
and J. Masters (eds.), Reflections of Nero (London, 1994); C. Gill and
S. Braund (eds.), The Passions in Roman Thought and Literature (Cambridge, 1996); P. Parroni (ed.), Seneca e il suo tempo (Rome, 2000).
The quotation from Eugène Ionesco, La cantatrice chauve, scene 8, at the
beginning of the Introduction is reproduced by permission of the publisher
from Emmanuel Jacquart (ed.), Thea tre Complet (Collection Bibliothèque
de la Pleiade, no. 372) (Gallimard, Paris, 1990).
ix
Acknowledgements
The quotation from W. H. Auden, Epitaph on a Tyrant (1939) on

page 117 is reproduced from Collected Poems by permission of Faber and
Faber Ltd, London and Random House Inc., New York.
Citations from Senecas tragedies are from Zwierleins OCT edition (1986).
Translations of some authors are taken, by kind permission, from the editions listed in the Note on translations. Other translations are my own.
Abbreviations of classical works correspond to those used in the Oxford
Classical Dictionary and, when not available there, in the Oxford Latin
Dictionary.
The following published translations have been used in this work:

a e s c h y lu s : vo l . i i : ag a m e m n o n . l i b at i o n - b e a re r s . e u m e n i d e s . f r ag m e n ts , Loeb Classical Library Volume 146, translated by
h . w. s m y t h , Cambridge, MA: Harvard University Press, 1926. The
Loeb Classical Library is a registered trademark of the President and
Fellows of Harvard College; Lucan: Civil War, translated with introduction and notes by S. H. Braund (Oxford Worlds Classics) (1992). Reprinted
by permission of Oxford University Press; Ovid: Metamorphoses, edited by
E. J. Kenney, translated by A. D. Melville (Oxford Worlds Classics) (1998).
Reprinted by permission of Oxford University Press; p l ato : vo l s . v/vi:
t h e re p u b l i c, Loeb Classical Library Volumes 237/276, translated by
pau l s h o rey , Cambridge, MA: Harvard University Press, 1936/1937.
The Loeb Classical Library is a registered trademark of the President and
Fellows of Harvard College; p lu ta rc h : vo l . i : m o r a l i a , Loeb Classical Library Volume 197, translated by f r a n k c . b a b b i t t , Cambridge,
MA: Harvard University Press, 1927. The Loeb Classical Library is a registered trademark of the President and Fellows of Harvard College; Seneca:
Medea. With an Introduction, Text, Translation and Commentary, by H. M.
Hine. Published by Aris & Phillips, Warminster, 2000; Fantham, Elaine:
Senecas Troades. A Literary Introduction with Text, Translation, and Comc 1982 by PUP. Reprinted by permission of Princeton
mentary; Copyright
University Press. s e n e c a : vo l s . i/ii: m o r a l e s s ay s, Loeb Classical
Library Volumes 214/254, translated by j o h n w. b a s o re , Cambridge,
MA: Harvard University Press, 1928/1932. The Loeb Classical Library
is a registered trademark of the President and Fellows of Harvard College;
s e n e c a : vo l s . v / vi : e p i s t l e s, Loeb Classical Library Volume 76,
translated by r. m . g u m m e re , Cambridge, MA: Harvard University
Press, 1917/1920. The Loeb Classical Library is a registered trademark
of the President and Fellows of Harvard College; s e n e c a : vo l . viii:
xi
xii
t r ag e d i e s, Loeb Classical Library Volume 62, edited and translated by

j o h n g . f i tc h, Cambridge, MA: Harvard University Press, 2002 (for
Hercules furens). The Loeb Classical Library is a registered trademark
of the President and Fellows of Harvard College; Virgil: Aeneid , translated by David West (Penguin Classics). Published by Penguin Books Ltd,
Harmondsworth, 1991.
Introduction
Mme Martin: Quelle est la morale?

Le Pompier: Cest a` vous de la trouver.
(E. Ionesco, La cantatrice chauve)
Thyestes embodies a tragic conflict, and an even more tragic contradiction,

between a desire to speak and the need to remain silent. Or, for us, between
the desire to watch and the repulsiveness of what is on display. The sheer
dramatic force of this tragedy Senecas best1 springs from casting Atreus
horrific violence as the creative drive behind poetic fiction. Thyestes stands
out among the other plays by Seneca precisely because it mobilizes in
novel, engaging fashion the archetypical connection between tragedy and
violence, power, sacrifice. In this play we witness in its most engaging form
a sustained reflection on the power and limits of poetry, a reflection which
on the one hand appears to sum up almost a century of Latin literature
and on the other codifies Silver poetics at its expressive (and, in a way,
theoretical) peak.
Thyestes foregrounds the complexities inherent in creating poetry as well
as in reading or watching it. Atreus dominates the stage as a gifted poet,
mired in the tension between order and chaos, passion and reason, enthousiasmos and craft. Inspiration, role-playing, deception and recognition are
not only staged, but metadramatically analysed and questioned,2 and force
the audience to reflect on whether enjoyment of this type of poetry is not
also a form of collusion with it.
1
It has attracted Richard Tarrants remarkable commentary (Tarrant (1985)), to which my work is
much indebted. Giancotti (198889, vol. i ) is also often useful. Among critical works specifically
devoted to Thyestes Picone (1984) is especially important; Mantovanelli (1984) offers a stimulating
reading of several aspects of the play; Guastella (2001) deals at length with Thyestes in the context of
a wide-ranging analysis of revenge as a tragic theme in Seneca and his successors. In English, there
is much of value in Littlewood (1997), with whom I occasionally, if independently, concur.
Like sex, self-reflexivity was not invented in the sixties: the scholiast to Il. 3.1267 already remarked
that as she weaves a cloth portraying the contexts between Greeks and Trojans Helen is a worthy
archetype of [the poets] own poetic art. See Bergren (1983) 79.
The Passions in Play
As the prologue shows, the poetic word, qua poetic word, can voice
realities which would otherwise tend to be repressed, and the act of creation
embodied in that word is inevitably an act of rebellion against logic and
order. The complex framework of the prologue also renders the balance of
moral responsibilities in the play difficult to determine and at every point
pressures the audience to distinguish good from evil, illusion from reality,
and hypocrisy from sincerity in the midst of conflicting, often contradictory,
signals.
The conflict between different forms of logic and different attitudes to
the passions can be most readily observed in the opposition between the
rationality of Thyestes and the chorus vis-à-vis the idiosyncratic unpredictability of Atreus. Atreus is not irrational, nor is he mad. He operates
according to different logical protocols, closer to those of the unconscious
than those normally adopted in waking life. Therein lies, I argue, a great
part of his irresistible appeal.
The primary aim of this book is to subject Seneca to the same kind of
sustained literary analysis as is now taken for granted for other major Latin
authors. I do not propose to offer a systematic psychoanalytic reading
of Thyestes,3 although Freudian and post-Freudian theories of literature
have shaped my approach to both literature in general and this tragedy in
particular. Thyestes invites from the very beginning an engagement with
concepts masterfully explored by Freud, since its prologue stages a conflict
between the Furys order to unleash the tragedy and Tantalus desire to
repress it. As the Fury succeeds, the words of the tragedy emerge as the
product of a violent creative urge rooted in the underworld of the Furies
and their passions: Thyestes, like the sixth book of Lucans Bellum Civile,
is a harrowing exploration of the kinship between prophecy, death and
poetry. The underworld and its passions, alluring and disconcerting alike,
are always lurking beneath the surface of the text, just as in the most
sacred part of the Roman forum a small opening, the mundus, permitted a
ritualized and strictly controlled contact with the realm below: ianua patet,
the door is open, Varro informs us, to the gods below, whose presence is
controlled but not denied, regulated but not destroyed.4 In Senecas poetry,
too, for the people of Dis a way is given to those living on earth (Oed.
573: iter . . . populis Ditis ad superos datur).
3
4
Such as, for instance, Segals (1986) monograph on Phaedra, or Janans (1994) on Catullus.
Varro ap. Macrob. Sat. 1.16.18 ( = fr. 66 Salvadore): mundus cum patet, deorum tristium atque inferum
quasi ianua patet, with Coarelli (1983) 20725.
Introduction
Freuds theories about time and temporality and the post-Freudian focus
on the unconscious as an alternative set of logical protocols5 offer valuable
guidance as we approach a play that is marked by temporal discontinuities
and leaps of logic, and approach a character such as Atreus, who is unable
to overcome the past and to set clear boundaries between himself and his
brother-doppelganger. Atreus will be able to punish his brother precisely
because he is able to trust his instincts and to manipulate words in unpredictable and duplicitous ways, to appropriate, even, distinctly feminine
characteristics as they suit him: Thyestes logic, one-sided literalness, will
be no match.
In turn, these insights help us understand the role and function of
Senecas intertextuality, a pervasive and intense feature of his writing, as
indeed of other key first-century authors such as Ovid and Lucan.
Ancient tragedy does not exist, of course, in a vacuum, and each play elicits from the audience a preferred set of emotional identifications. Thyestes
could (and may even aspire to) invite the audience (an admittedly vague
term) to identify with the emotional suffering of the eponymous hero, as
he is disgraced, betrayed and horribly punished. We would fear with him
as his brothers deception unfolds, and suffer with him as his children are
slaughtered. But the specific dramatic construction of Senecas Thyestes radically modifies this expectation: its elaborate metadramatic structure offers
a detailed knowledge of the Furys and especially of Atreus machinations
and makes us party to the superior level of knowledge the latter enjoys over
his brother. We side with the creator of fear and horror, not with his victim:
on the whole we do not fear with Thyestes, we plot with Atreus, even if the
hallmark of any successful work of art, as Freud was the first to admit, resides
precisely in its ability to fragment the audiences identification. Specifically,
since Atreus is endowed with all the characteristics of a successful creator
of poetry, we inevitably pay tribute to his inventiveness as we revel in the
aesthetic rewards of the tragedy and tacitly admit the possibility that powerful poetry may well be at odds with moral propriety. We do not have
to posit a radically modern notion of consciousness to accept that Thyestes
challenges the Stoic prescription that poetry should have an educational
value.
Among the modes of representation which have a particular impact
on the interpretation of the play, I single out, especially in chapter two,
the technique of framing. Frames, to be sure, have attracted considerable
interest from literary theorists, and deservedly so. Yet I will stress their
5
I refer especially to the works by Matte Blanco and Orlando listed in the bibliography.
unusually emphatic role in the emotional dynamics outlined above. As

Freud himself recognized, the very theatricality of the play acts as a frame
by positing a distance between the audience and the events on the stage.
In Thyestes, however, this general, external frame is supplemented by a
very specific and elaborate set of internal frames which articulate different
layers of dramatic action. These frames offer the audience an ordered and
apparently reassuring context which acts to lower their intellectual defences
and to pave the way for the emergence of violent, repressed contents. By the
time these contents emerge fully in Atreus extraordinary mise en scène, the
audience is engulfed in the emotional violence of the tragedy. Indeed,
the same interplay of order and violence constitutes a defining feature of
Atreus personality throughout, as can be observed, I will argue, in the eerily
ordered procedure he follows in slaughtering his nephews.
Once we accept that the very structure of Thyestes maps out a profound
conflict, we must of course investigate the nature of the repressed emotional
truth that we are invited to experience alongside Atreus, a truth simultaneously hidden and revealed by the play. This search is more awkward if
we focus predominantly on Atreus cruelty or we privilege the political
dimension of the conflict foregrounded by the chorus and by Thyestes
that is, Thyestes expulsion from his fathers kingdom, his exile, and Atreus
subsequent willingness to welcome him back as a partner in power. The
harrowing emotional background of the play can be glimpsed, rather, in
Atreus brief but uncontradicted references to Thyestes past behaviour.
Atreus revenge is not primarily motivated by issues of power, even if eliminating his nephews strengthens the dynastic position of his own offspring.
The deep-seated causes of Atreus anger and violence are Thyestes incestuous relationship with Aerope and the consequent uncertainty about the true
paternity of Agamemnon and Menelaus. By privileging a political reading
of the play and heeding Thyestes reflections on the nature and limits of
power, the chorus actually distracts our attention from the primal emotions which motivate Atreus and inevitably cast Thyestes in a less flattering
light. Despite the fact that he has come to laudable conclusions about the
relative merits of power and powerlessness, Thyestes nevertheless remains
an incestuous adulterer. The political subplot of the play is to a significant
extent an enabling device for the emergence of darker instincts and issues
which could not immediately command centre stage. In turn, the truth
which seeps out in the confrontation between the two brothers makes the
moderate political outlook of Thyestes and the chorus, their insistence that
passions can be tamed and conflicts amicably resolved, look very dubious
indeed.
Introduction
Atreus anxiety about his paternity helps to explain why he plans and
executes his revenge in such a way as to ascertain in the process his sons
instinctive allegiance: in the end they do not inform their uncle of the
impending slaughter, and by their deeds they show to Atreus that he is, in
fact, their father, as he finally realizes with joy: now I am convinced that
my children are my own; now I believe that I can trust again the purity of
my marriage-bed (10989: liberos nasci mihi | nunc credo, castis nunc fidem
reddi toris).
A predominantly political reading of the play opens up the possibility of
a moralistic reading, but tames its deeper emotional power. Atreus anger at
the incestuous betrayal and his horror at the thought that the children are
not his own are emotions readily shared by (at least) a Roman audience, and
his revenge fulfils a profound if repressed truth that in a similar situation
they too would want to exact a similarly gruesome retribution. If, as Freud
famously argues about Oedipus Rex and Hamlet, successful tragedies focus
on basic emotions and impulses of the human condition (a concept which
retains full heuristic value even as we modulate it in a historical perspective),
then we can understand why Thyestes is still considered the best of Senecas
tragedies, and why its emotional impact is comparable to that of other great
masterpieces of theatrical literature. As a play which goes to the heart of
the connection between poetry, power and incest, Thyestes can rightfully
aspire to a hallowed position in the canon.
The temptation to read Atreus as a larger-than-life Nero a trend that
might have started very soon after the play was written6 is still strong.
As the pre-eminent literary work of (probably) the fifties,7 this tragedy is
inevitably linked in our historical perception with the image of the emperor.
The association may well be inevitable, but we should resist the temptation
to see the tragedy as a document of sorts for the decadence of Neronian
Rome, or for that matter as a manifesto for moral resistance to that
decadence. All this would be predicated, obviously, even if not explicitly, on
several dubious assumptions: that, for instance, (this) tragedy reflects the
social situation in which it was produced; or that Nero was in fact the cruel
and rather quirky tyrant who sang while Rome burned. Both assumptions,
if proved, could most probably help our understanding of the play. Indeed,
6
7
Calder (1983) neatly shows that the character of Nero in the pseudo-Senecan tragedy Octavia follows
in Atreus footsteps.
There are no certainties about the dating of the tragedies, but metrical data point to a late date for
Thyestes (Fitch (1981); cf. Nisbet (1990)). Calder ((1976) 2830, (1983) 184) argues that Agamemnon is
likely to follow Thyestes. For a more sceptical position on the dating of Ag. see Tarrant (1976) 56.
it is perfectly plausible that a play such as Thyestes could have political

overtones. Augustus, for instance, paid good money for Varius Thyestes,
which was staged after the victory at Actium and hinted at a connection
between Atreus and Antony. Centuries later, Shakespeares Titus Andronicus,
which relies on both the Thyestes and its Ovidian model,8 will offer a
coded but perceptible critique of contemporary royal power. But if I refrain
from casting Thyestes in the dubiously honorific role of prime witness for a
reconstruction of Neronian Rome, it is because we know little about the
circulation of the plays (the longstanding quarrel about their performability
having all but displaced such a crucial issue),9 and thus we are ill at ease when
it comes to evaluating the relationship between the text and its possible
audience: the emperor?10 dissident aristocrats? family members? nobody at
all? (Conversely, we do not know to what extent the archetypical sadist
Nero transmitted to us by generations of awed and scandalized critics is
a product of historical accuracy or the crystallization of anthropological
horrors in an appealing if repulsive set of rhetorical topoi).11 Atreus
winning combination of wit and violence would have looked very different
if staged in front of Nero not long after Britannicus was conveniently
dispatched, or in the secrecy of Senecas home as the Pisonian conspiracy
took shape. As long as we lack for Senecas Thyestes the details we possess
about the circumstances in which Varius Thyestes and Shakespeares Titus
Andronicus were represented, it is much better to focus on Thyestes as a
reflection on power, creativity, perversion and desire which need not be
explained in terms of a specific political background.
A tragedy such as Thyestes must also have been a considerable challenge
for its self-professed Stoic author: Atreus elaborate revenge plot is crowned
with success, and Thyestes less than compelling gestures towards restraint
and morality are met with defeat (compare the very different ending of
Titus). Thus we would probably do well, on the one hand, to dispense with
a political reading (especially a` clef ), and, on the other, to relinquish the
desire to reunite the whole Senecan corpus under the reassuring, conclusive
8
9
10
11
The relative importance is debated; see p. 70 n. 1.

I have little doubt that the tragedies, whether or not they were actually staged, were written as
performable theatre plays. In practice, it is plausible that they were performed in small, private
theatres, in the Hellenistic tradition; Calder (197677), (1984); Marshall (2000). Other theories:
(i) Lesedrama, purely for recitation, as advocated most extensively by Zwierlein (1966) cf. the
review by Lefèvre (1968); (ii) recital with several voices, but no costumes and no stage setting, rather
like operas in concerto-form see Fantham (1982) 3449; (iii) full staging, actual or potential
see Walker (1969); Herington (1982); Sutton (1986).
Calder (1976).
See Elsner and Masters (1994) for discussions of Neronian culture, and specifically of the myth of
Nero, from a variety of methodological points of view.
Introduction
sign of Stoic orthodoxy, or even only of Stoicizing morality. We must give

up the illusion of a Seneca morale, who structures his literary production
along the constant axis of philosophic doctrine, and welcome in its stead the
nuanced image of an author who is at times enigmatic, often contradictory
and always challenging.
Yet precisely because it should discourage a specifically Neronian reading, Thyestes can safely be considered the mastertext of Silver poetics. The
play pushes to breaking-point a debate about the role and function of the
poetic word which lies at the heart of works such as Ovids Metamorphoses
or Lucans Bellum Civile. Its exploration of passion, hatred and horror is
more concentrated and sustained than in Lucan or Statius; its lumping
together of the personal and the political amplifies a line of thought which
is central to post-Virgilian literature, as are its preferred forms of expression self-reflexive, highly charged, bordering on the illogical. I hope that
this book will also shed some light on that peculiar poetics and its main
representatives.
I have framed the treatment of Thyestes with more general reflections on
the nature of tragic poetry gleaned both from other Senecan tragedies
and from his prosework. My goal was not to superimpose on the play a
normative explanation that would forcedly orient interpretation, but, on
the contrary, to claim that the tragedies own self-reflexive statements on the
nature of poetry afford readers considerable latitude in their own exegetical
explorations. Thus, in chapters one and six I argue from within Senecas
own corpus for the legitimacy of an open reading of Thyestes. In chapter
two I disengage the metadramatic aspects of the prologue and reflect on
the methodological implications of this self-reflexive aspect of the play.
Chapters three and four are devoted to the analysis of the main characters,
Atreus, Thyestes and the chorus. There I privilege what I consider to be the
aspects of their textual existence that impact most extensively on the play as
a whole: Atreus role as master of ceremonies in the sacrificial slaughter of his
nephews and his unchallenged epistemic prowess; Thyestes contradictory
and ineffectual penchant for moralization; and the choruss detachment
from events and its incapacity to understand and affect them significantly.
In chapter five I expand my analysis to a number of other plays in order
to come to terms with two interconnected and fundamental aspects of
Thyestes and other tragedies: their obsessive dealings with the past at the
level of subject matter, and (most explicitly) their privileging of intertextual
connections.
chapter 1
iterque populis Ditis ad superos datur

(Seneca, Oedipus 573)
negat enim sine furore Democritus

quemquam poetam magnum esse
posse, quod idem dicit Plato
(Cicero, De divinatione 1.80)
i
At the core of Senecas Oedipus stands Creons stunning narrative of his
search for a truth that has so far escaped his fellow-citizens, even that
cunning antonomastic observer, the king of Thebes. Overcoming a deep
reluctance to speak, on account of Oedipus threats, Creon retells his experience in all its gory detail (509658). Suitably enough, the setting for
his account is grim and terrifying, remote and obscure: there is, far from
the city, a wood dark with ilex-trees near the well-watered vale of Dirces
fount (est procul ab urbe lucus ilicibus niger | Dircaea circa vallis inriguae
loca, 5301).1 It is in this extraordinary location, whose wilderness is the
usual environment for magical contacts with the divine, that the sacerdos
(548), soon referred to as a vates (552), begins his portentous rites. The
prophet, who is possessed by divine powers, intones a magic song: he
unfolds a magic song, and, with frenzied lips, he chants a charm which
appeases or stirs the evanescent ghosts (5613: carmenque magicum volvit et
rabido minax | decantat ore quidquid aut placat leves | aut cogit umbras), then
sings again, and looking at the ground, summons the shades with a deeper,
stunned voice (5678: canitque rursus ac terram intuens | graviore manes
voce et attonita citat). Thus he succeeds in evoking the ghosts of the dead:
I am heard, says the priest; I have uttered prevailing words; blind Chaos
1
For a comparable setting in Thy. 64190 and its interpretation see ch. 4.
is burst open, and for the people of Dis a way is given to those living on
earth (5713: audior vates ait, | rata verba fudi: rumpitur caecum chaos |
iterque populis Ditis ad superos datur ).
The prophets invocation has horrific consequences: trembling (horror,
576) shakes the grove, the earth splits open (5826), and a triumphal procession of infernal creatures abandons its chthonic dens: then grim Erinys
sounded, and blind Fury and Horror, and all the forms that eternal darkness
creates and hides (5902: tum torva Erinys sonuit et caecus Furor | Horrorque
et una quidquid aeternae creant | celantque tenebrae). There follow (5924)
Grief (Luctus), Disease (Morbus), Old Age (Senectus), Fear (Metus) and
Pestilence (Pestis).2 The prophet is not disturbed by this, unlike Manto
(5956). Then other ghosts appear: Zethus, Amphion, Niobe, Agave with
the Bacchants, and Pentheus, a catalogue of tragic figures.3 Last, apart from
the crowd, Laius shows his face, and speaking in a rabid voice (ore rabido,
626), reveals the cause and nature of the plague.
This scene powerfully enacts what poetry and poets do. The traditional
connection between the magic and prophetic power of poets and seers,
crystallized in the multifaceted use of the words vates and carmen, finds
here a contextual motivation.4 The vates, who through his song, that is,
through carefully chosen words endowed with active power, rata verba,
can bring to life the underworlds demonic creatures, is analogous to the
poet, whose inspiration vivifies the characters of tragedy. The regenerative
powers of the vates and the poet intersect in the parade of tragic characters
described at 61118: both the vates5 and the poet can access a domain open
2
Violenta fata, horridus Morbi tremor, Macies, Pestis and Dolor close off the tragedy in Jocastas final
invocation at 105960. A different list of personifications appears in Her. F. 968 and 6906 (see Fitch
(1987) 150 and 300), recalling Virg. Aen. 6.27381. At Ovid, Met. 4.4845 Luctus, Pavor, Terror and
Insania escort Tisiphone back on earth as instruments of Junos rage. See also the metaliterary cortège
of Fama at Met. 12.5961 (Credulitas, Error, Laetitia, Timores, Seditio, Susurri), with Zumwalt (1977)
and Feeney (1991) 2479.
Zethus and Amphion (whose mother Odysseus meets in the underworld: Od . 11.2605) figure in
Euripides lost, but once very famous, Antiopa (177227 Nauck2 ), and in Pacuvius tragedy by the
same name (120a Ribbeck2 ). They build the walls of Thebes by playing on the lyre (Hes. fr. 182
Merkelbach and West). Niobe, Tantalus daughter and Amphions wife, gives her name to tragedies
by Aeschylus and Sophocles. The subject was popular for plays and mimes well into imperial times, if
Nero is reported to have sung that very role (Suet. Nero 21). Agave and Pentheus appear in Euripides
and Accius Bacchae, and in Pacuvius Pentheus (for which we have only Servius argument, ad Aen.
4.469). All these characters are Theban; on Thebes as a privileged locus of tragedy see Zeitlin (1990),
a classic paper.
On vates see Newman (1967). The intersection of meanings between vates and sacerdos in passages
such as Hor. Carm. 3.1.23 or Prop. 3.1.3 is also relevant (see the use of sacerdos at Oed . 548). On the
importance of the concept of vates in Lucan see OHiggins (1988) and Masters (1992). On carmen
see Sharrock (1994) 634.
The term vates does not appear to refer directly to dramatic poets, nor would it be possible for this
particular self-reflexive narrative, which is necessarily linked with a mythic plot, to stage anything like
10
only to a non-rational, horrific form of Dionysiac inspiration,6 and both

testify to the limits of a rigid faith in rational forms of explanation.7
ii
When the play opened, we saw Thebes being slowly destroyed by the
plague, and Oedipus paralysed by fear, after the Delphic oracle predicted
the monstrous deeds he has in fact already accomplished. We are told that
he fears unspeakable things (infanda timeo, 15), yet his reaction is portrayed
as excessive: such a situation should be confronted with reasoned poise, but
Oedipus is completely engulfed by passions, as he declares at 257:
cum magna horreas,
quod posse fieri non putes metuas tamen:
cuncta expavesco meque non credo mihi.
When you dread some great calamity, you must fear also events which you think
cannot happen. I dread every thing, and I do not trust even myself.
Jocastas exhortation at 826 confirms that we are to consider Oedipus

emotions excessive, if not altogether unjustified:
regium hoc ipsum reor:
adversa capere, quoque sit dubius magis
status et cadentis imperi moles labet,
hoc stare certo pressius fortem gradu:
haud est virile terga Fortunae dare.
This very thing, I believe, is regal: to contain adversity and, the more dubious your
station and the more the greatness of power wavers, the more to stand firm, brave,
with unfaltering foot. It is not a mans part to turn the back to Fortune.
This overwhelming fear is the real motor of the tragedy (not so, famously,
in Sophocles). It is this that spurs Oedipus to engage in his painful search
for truth through a tortuous path. His first chance to discover the truth
is in fact vitiated by a residual trust in reason. In the scene beginning at
6
7
a properly named poeta. In Horaces Letter to Augustus (Epist. 2.1.21113), the tragic poet is equated
with a magus who with inanities wrings my heart, inflames, soothes, fills it with false terrors like a
magician, and sets me down now at Thebes, now at Athens.
Further observations below, ch. 3, passim.
By stressing the irrational passions at work in the tragedies I do not want to deny the importance
of the rational elements of artistry and craftsmanship which play an extremely prominent part in
these texts. On the contrary, it is precisely thanks to the elaborate forms of its mannerist rhetoric
that irrational and disruptive contents find their expression: the figure is the perpetual tribute
paid and how willingly it is paid by the language of the conscious ego to the unconscious
(Orlando (1978) 169).
11
line 202, Creon brings the intricate and convoluted vaticinium of the Pythia
(211; 21314), but the king replies that he can easily handle the task, since
this is precisely his prerogative: to read riddles to Oedipus alone is given
(216: ambigua soli noscere Oedipodae datur). Oedipus trust in his rational
faculties outlasts even Creons second and much more explicit description
of a magical rite, as Oedipus, while trying to deny the impact of what he
has just heard, boasts that he knows himself better than the gods do: and
yet my soul, conscious of innocence and known to itself better than to
the gods, makes denial (7667: sed animus contra innocens | sibique melius
quam deis notus negat). In the face of this new and powerful challenge he
will have to delegate his responsibilities more than once, and will confess
to the impotence of his vaunted rational skills. The real breakthrough in
learning the truth occurs only because of the elaborate magic rite organized
by Tiresias and Manto and reported to Oedipus by Creon. I want now to
consider this compelling scene which lies at the structural and emotional
centre of the play.
When Creon and Oedipus meet, the king asks his brother-in-law to reveal
the results of his consultation with the inhabitants of the underworld. The
stichomythic dialogue (50929) leading up to Creons long speech (530658)
is best read alongside a similar exchange between the Fury and Tantalus
in the prologue of Thyestes, where Tantalus tries to resist the Furys order
to bring to earth the crimes (scelera) that actually constitute the play. His
refusal to provoke scelera is a refusal to produce the words that recreate
that scelus in the play. In Oedipus, Creon begs for the right to be silent,
and Oedipus, like the Fury, must persuade him with force. Just as the
words of Tantalus come into existence only because violence quashes his
intransigence, Creons revelation is similarly marked as a forced confession
of truths which he claims are best left unsaid.
Further details concerning the relationship between the various characters should be taken into account. Oedipus, the vates and Laius are structurally linked. Oedipus consults Creon, who turns to the priest, who is
then able to interrogate Laius. As is fitting in a mise en abyme,8 the inset scene is a microcosm of the larger framework, and this makes reflection perceptible. It is significant that Laius speaks with the same rabid
voice (ore rabido, 626) with which he had been summoned by the vates
(rabido . . . | . . . ore, 5612). With different degrees of power and knowledge,
these three characters all embody a desperate search for truth, the very
search that motivates the tragedy from its inception. Oedipus opens the
8
The standard treatment of mise en abyme is still Dallenbach (1977).
12
play with his investigation, and the answer to his questions will come only
from tragedy. The search for truth thus becomes a search for poetry.9
The uncontrolled fear that pushes Senecas Oedipus to search for explanations (unlike his Sophoclean counterpart)10 eventually leads him to
discover in the song of Laius the truth he was afraid to know. Passion leads
to poetry, and poetry is the revelation of truths carefully hidden from the
upper world of reason and power. As the vates literally finds a way for the
creatures of Acheron to come back to earth (573: iter . . . populis Ditis ad
superos datur), so Laius allows a terrible and suppressed truth to be voiced
and heard. Poetry evokes Erinys, the new Muse of this poetry, but also the
sources of a deeper knowledge, one which Oedipus proud rationality had
failed to grasp.
Knowledge can be found in a poetry which is profoundly passionate
in its origins and inevitably chthonic in its appearance. It is a knowledge
which exists and acts in lieu of reason and against it. At the end of the play,
Oedipus, the cunning thinker and observer, the man who boasts his ability
to interpret traces (vestigium, 768) in his search for truth, ultimately destroys the instrument and symbol of his reason. Vision had already proved
to be an unreliable source of knowledge. In the scene starting at line 303,
Manto describes every phase of a sacrifice to Tiresias, who tries to understand why the plague is destroying Thebes. In spite of Mantos accurate
report, however, Tiresias admits finally that the truth cannot be found in this
way, and alia temptanda est via (392). (It is worth noting, again, that great
emphasis is placed on the medium of analysis: Tiresias blindness, which
requires Mantos description, emphasizes the problematic status of vision
more than an eye-witness account would have done.) The analysis of signs
through the eyes, the ultimate rational pursuit which recalls Oedipus pride
in his rational faculties, is doomed to failure. As Tiresias explains, the usual
signs cannot express the name of the culprit (3904).11 As he gouges out
his eyes Oedipus becomes a second Tiresias, thereby implicitly recognizing
the blind seers superior cognitive power (971),12 a power deeply rooted in
the chthonic realm of blood and passions.13
9
10
12
13
This relationship is foregrounded by the fact that Oedipus had failed to draw useful conclusions
from Mantos prophecy at 2338, because at that stage he was still proudly relying on his rational
abilities. He failed to understand the oracle himself, but he took reasonable and obvious steps
towards solving the enigma. In parallel fashion, the text did not emphasize at that point the poetic
character of the prophetesss song, who was nonetheless called vates (230) and acted near the fons
Castalia.
11 On alia temptanda est via see below, pp. 2267.
Cf. Bartezzaghi (1988).
See night, nox (977); darkness, tenebrae (999).
See the string of adjectives at 5515: funesto (551), lugubris (553), squalente (554), mortifera (555).
13
iii
Senecas tragedies offer repeated and complex descriptions of the passions in
action and of the effects of passions on both agents and victims. My intention in this chapter, however, is not to analyse the passions which animate
the characters themselves Medeas and Phaedras destructive love, for instance, Atreus thirst for revenge, or Thyestes own quivering determination
to resist passion. Rather, I aim to examine the way in which these characters establish a connection between passions and poetic creation, and thus
problematize the relationship between passions and aesthetic pleasure. The
basic assumption of my enquiry, which I have already put to work in my
reading of the central rhesis in Oedipus is that, at several critical junctures,
the actions of certain characters embody a reflection of the text on itself
and offer important insights into its poetics. Senecan tragedy is a highly
metadramatic form of theatre; that is, one highly self-conscious in its reflection on the nature and modes of its existence. In this respect, Senecas
metapoetic concerns are clearly on a par with those that animate works
such as Ovids Metamorphoses or Lucans Bellum Civile.14 After Virgil, poetry appears increasingly unable to resist the compulsion to mirror in its
own body the processes of composition and the narrative mechanisms that
make it possible. Succumbing to this temptation can produce the pleasing,
if slightly dizzying, effect of the mirror reflecting its image onto another
mirror, endlessly complicating the modalities of reference. But it can also
produce a sense of enclosure bordering on anguish. For such a mistake,
after all, Narcissus dies.
Such a line of enquiry forces the critic to confront similar dangers. It
does offer, however, considerable strategic advantages over approaching the
tragedies armed prevalently or exclusively with references to Senecas prose
works, as if they could be considered a theoretical, systematic explanation of
the convoluted, dense universe of the tragedies. The prose works, too, can
be shown to oscillate between points of view, and to display self-repressive
tendencies.15 If we are to link the tragedies to the prose works, we should
at least be ready to dispose of any rash assumptions of hierarchy, and to
see the connection going both ways. There is no reason to believe that the
explicit statements of the prose works should have a higher claim to truth
than the tragedies, and thus be used to muffle the potential disruptiveness
of the tragedies.
14
15
For an analysis of the role played by metapoetic elements in Ovid see Rosati (1983) and Hinds (1987);
for Lucan see Masters (1992). On Senecan tragedy Boyle (1997) 193207 is especially good.
See pp. 201 later in this chapter.
14
While I cannot review in detail here the extensive critical debate on

metadrama, some elements should be clearly established.16 Metadrama
and metatheatre, terms which play a prominent role in the modern theorization of the theatrical experience, encompass a variety of phenomena. It is useful, therefore, to retain Abels original term metatheatre in
order to designate only that most elaborate and (from a structural point of
view) least ambiguous of meta-phenomena, the theatre-in-the-theatre
a phenomenon which is as structurally constrained as it is historically
circumscribed.17 Metadrama, on the other hand, embraces more varied
and often less intrusive peripheral forms18 which bear important semiotic implications. As the equivalent in theatrical terms to metanarrative,
metadrama can usefully indicate moments when the play, through a variety
of devices, reflects on itself and its functioning.
There are no proper metatheatrical elements in Senecan tragedy, no
techniques that fracture the fourth wall, and the dramatic illusion with it.
Even in Thyestes the fictional illusion is never directly challenged and broken, and, although I will often refer to Atreus performance, the tragedy
hosts no formalized, Shakespearean play-within-the-play, no Plautine slave
ready to step aside and address the audience outside the boundaries of
fictionality.19 Yet it would be difficult to play down the structural elements which, I will argue, make the audience aware of the constructedness of the performance by distinguishing between different dramatic
levels. A pivotal role in such a complex structure is played by the recessing frames which encompass distinct sections of the tragedy. Framing,
as I will suggest,20 is a structuring criterion which massively influences
the audience reaction to the play. Although framing and metadrama are
likely allies, framing by itself need not be metadramatic, and both terms
should thus be retained in order to account for two different aspects of
Senecan tragedy. As a working definition, which I will refine as I proceed,
I will therefore consider to be metadramatic the elements in the play that
are explicitly concerned with the structural arrangement and the internal
16
17
18
19
Important insights on metadrama, and further bibliography, can be found in Hornby (1986). I have
benefited mostly from Schmeling (1982), as well as from Hamon (1977), Prince (1977), Hutcheon
(1984) and Stam (1992). Although his notion of metatheatre does not directly bear on my argument,
Abel (1963) who coined the term is essential reading. Calderwood ((1971) 47) has useful remarks
on the notion of metadrama that I will employ, which is not limited to forays across or . . . around
the borders between fiction and reality, but is based on the assumption that plays are also about
plays.
Abel (1963).
I adopt a distinction suggested by Schmeling (1982) 10. See Schmeling (1982) 5 for a list of various
phenomena, all to a certain extent metadramatic.
20 See below, pp. 4561.
See Slater (1985), and especially M. Barchiesi (1981) 14774.
15
organization of the drama, especially as they focus on the authorial role of

certain characters, even if those references do not trespass the boundaries
set up by the scenic space. For example, I will regard as metadramatic those
parts of Thyestes which are extensively concerned with the preparation and
the mise en scène of Atreus revenge, although neither the prologue nor act 2
breaks the barrier of scenic illusion. Atreus and other Senecan characters
who also wear the robe of inspired creator transcend their role as characters
in the play and go on to assume, implicitly but clearly, some of the functions that other forms of poetry assign to internal narrators. Endowed with
a knowledge of events that is far superior to that of their fellow-characters
(it is they, after all, who steer the plot in the desired direction), Atreus
and his metadramatic colleagues double up as authors-on-stage and constantly remind us of the non-realistic nature of the staged events. It is no
coincidence that metadrama plays such a vital role in tragedies which can
only be products of an intensely self-conscious literary project. In Senecas
Rome, tragedy is a form of expression which has by now lost the relative
ritual and political immediacy which it enjoyed in its original Greek
setting, and even, it could be argued, in the early stages of Roman literature. When Seneca writes his tragedies, writing tragedy inevitably appears
to be a problematic, regressive operation: the metadramatic layers detectable in many Senecan plays testify to the harrowing complexity of that
project.
In other ways, of course, Senecas Rome is also one big theatrical stage,
where power is constantly enacted and represented, and where, to borrow
Duponts phrase, the actor is king and the king is an actor. If power is
necessarily predicated on a careful orchestration of symbols, Imperial Rome
is in many ways the quintessential theatre of power.21
A final caveat. For twenty-first-century readers (and critics) the concept
of self-reflexive, metadramatic or narcissistic texts is reasonably familiar,
if nothing else because of the extensive metanarrative inclinations of that
most successful modern literary genre, the novel. Hardly less important
is the use of metanarrative structures in contemporary cinema.22 Whether
or not we accept the suggestion that the novel has always harboured from
the very beginning the seeds of a narcissistic reading,23 we must acknowledge that metanarrative devices abound even in novels which have
nothing in common with the most explicit products of nouveau roman.
Yet it is precisely our familiarity with these ideas that risks impairing our
21
22
See Dupont (1985), with Woodman (1992) and Bartsch (1994).

The topic is well treated by Stam (1992), with ample bibliography.
23
Hutcheon (1984) 23.
16
understanding of its ancient counterpart. In modern fiction, metanarration

is actively opposed to realism: the reader is constantly reminded of the fictional status of the representation. In the case of Senecan tragedy it would
be misguided simply to see narcissistic tendencies set against aspirations
to realism. What forms of realism, if any, we can detect in ancient literature
is, of course, a major question of literary history; among other factors, significant epistemological issues are involved here.24 Whereas Greek tragedy
flirts constantly with the temptation or illusion of portraying reality,
Senecan tragedy is seemingly oblivious to its allure. Therefore an understanding of metanarrative in the tragedies must necessarily be attuned both
to the specific issues raised by Senecas writing and to the inclinations of
Latin literature at that time.25 A distinction should, of course, be observed
between realism and reality. Senecan drama is emphatically alien to realistic forms of representation, despite the fact that some elements of reality
(itself a tricky term) can indeed find their non-realistic representation there.
However, more often than not the attempts to read some of the plays as
tragedies a` clef , starring Nero as Oedipus or Agrippina as Phaedra, look
reductive and unconvincing.26 What the tragedies tell us about Senecas
Rome is more interesting and less obvious.
iv
It is a mark of self-reflexivity in Senecas tragedies that the character who
controls the dramatic action and displays superior knowledge and power
on stage can often be seen as embodying the playwright, and can thus offer
implicit insight into the poetics of the play. There are several candidates for
this metadramatic role: Juno in Hercules furens, Medea, Atreus.27 I want to
focus again, however, on a less typical and more complicated case, that of
Oedipus.
First, a brief detour. A passage from Medea offers an interesting introduction to the way in which certain parts of Senecas tragedies can tell us
a great deal about how the author represents his own function. This does
not mean, obviously, that they should be taken as public confessions of
24
25
26
27
It would be interesting to combine Auerbachs (1959) treatment of realism with an analysis of

metanarrative structures. This is particularly true in the case of a text of enormous density such as
Petronius Satyricon, whose metafictional elements are prominently displayed.
I will return to the whole question sketched here several times in the course of this book.
See Calder (197677), an excellent discussion.
I will deal in ch. 3 with how Atreus fits into this group of female characters.
17
the historical author, or that, qua metadramatic, they should have a higher
claim to authenticity and univocity than anything else in the tragedy: they
merely represent important moments for the text to reflect on itself and its
poetics.
In the prologue Medea seeks to transform the storm of her emotions
(mens intus agitat, 47) into a revenge-plot. In doing this she is the prime
mover of the play, and thus already close to embodying a quasi-authorial
function. Medeas decision to find a way (viam, 40) for her revenge and,
later, her selection of the most appropriate means to do so, and her careful
realization of her plans all constitute the decision to create and represent
a tragedy. In this respect, Medea is similar to other characters who occupy
a central position in Senecan plays: Juno in Hercules furens or Atreus in
Thyestes. Indeed, Atreus first line on stage 176: ignave, iners, enervis . . .,
undaring, unskilled, unnerved . . . echoes Euripides, Medea 8078: let
no one consider me impotent (), weak (
) or spiritless
( ).28 All appear on stage debating their vengeful plots out loud
and giving voice to the torments of creation.29
In search of inspiration for her actions Medea invokes divine powers
with an ominous voice, voce non fausta (12). The invocation30 to her
idiosyncratic Muses follows the regular form of klesis (1317):
nunc, nunc adeste sceleris ultrices deae,
crinem solutis squalidae serpentibus,
atram cruentis manibus amplexae facem,
adeste, thalamis horridae quondam meis
quales stetistis . . .
Now, now, come to help me, goddesses who avenge wickedness, your hair defiled
with dishevelled serpents; grasping black torches in bloodstained hands, come to
help me, as grim as you were when you stood outside my wedding chamber.
Now, while she prays that the Furies approach with their dirty hair and
black torches, she echoes the poets invocations for divine inspiration and
concludes her proem, some thirty lines later, with a clear indication of the
28
29
30
The connection is particularly interesting in the light of the metaliterary resonances of line 176 itself;
see below, pp. 1312. Note also the possible connection between Thy. 257 and Eur. Med. 37680,
where Atreus and Medea discuss various options of revenge. The connection may be flagged by via
at 244 (profare, dirum qua caput mactem via), cf.
at Med. 376. A connection with
Accius may also be discernible, see p. 81, n. 24.
On the prologue to Hercules furens see below, pp. 1834.
See Petrone (1984) 1314 and Nussbaum (1994) 445.
18
forces she intends to rely on. Ira and furor, Medea claims, will drive her
actions, and the plot with them (4552):
effera ignota horrida,
tremenda caelo pariter ac terris mala
mens intus agitat: vulnera et caedem et vagum
funus per artus levia memoravi nimis:
haec virgo feci; gravior exurgat dolor:
maiora iam me scelera post partus decent.
accingere ira teque in exitium para
furore toto.
Wild deeds, unheard-of, horrible, calamities at which heaven and earth alike shall
tremble, my heart deep within is planning wounds, slaughter, death, creeping
from limb to limb. Ah, too trivial the deeds I have rehearsed; these things I did
in girlhood. Let my grief rise to more deadly strength; greater crimes become me,
now that I am a mother. Gird yourself with wrath, and prepare for deadly deeds
with the full force of madness.
Medea seems to be aware of the essentially literary nature of her pursuit.

Not only in the sense captured by Wilamowitzs dictum that she must
have read Euripides tragedy about herself,31 but also because she explicitly
hopes for literary recognition of her deeds. Directly after the invocation to
her Muses which we have just read, she goads herself by saying let your
repudiation be told as equal to your wedding (paria narrentur tua | repudia
thalamis, 523). The tragedy we are watching fulfils this wish. In a similar
fashion, Atreus vows that his revenge must not be approved by anyone in
future, but nor must it be passed over in silence: age, anime, fac quod nulla
posteritas probet, | sed nulla taceat (1923).
The mise en abyme (of sorts) from Oedipus is the one that delves deepest
into the reconstruction of creative processes, and also the one that spotlights most forcefully what begins to appear as the circular nature of these
processes. Fear will lead to poetry, and poetry will produce fear. But the
metadramatic resonance of Oedipus is also different in important respects
from that of other plays. To a limited extent, Oedipus embodies the functions of the playwright, as do Medea, Juno or Atreus. His quest for truth
is the raison detre of the tragedy and its catalyst, alongside Medeas, Junos
and Atreus thirst for revenge. But he must delegate these functions to
31
Wilamowitz (1919) iii.162. It is precisely her awareness of being a Medea (910: Medea nunc sum,
which harks back to 171: NUT. Medea ME. fiam), of being part of a literary universe, that
substantiates Medeas metadramatic character.
19
other characters, presumably because his persistent trust in reason makes

him unwilling to yield fully to the forces of inspiration and poetry. In the
end, poetry and truth will come from a real prophet, a vates.
Oedipus immediately rejects the truth offered to him, relying on the
deceptive evidence that Merope is in fact married to Polybus. It will take
a whole new act of the play, and a careful analysis of traces (vestigia), for
him to accept that he himself is the culprit. This delay between Laius
revelation and Oedipus refusal to accept it intensifies the tragic irony that
permeates the play. At this point, it is really only Oedipus, ever the cunning
investigator, who continues to believe in reason and refuses to see the truth
which the chthonic force of poetry laid out in no uncertain terms. His
reaction to Laius revelation is to suspect that the vates and Creon are
plotting to seize the throne (66970). Oedipus staunch defence of his
rational methods of pursuing truth and his consequent denial that passions
can give answers to his doubts relate directly to his continued grip on
power even in the face of overwhelming adversity. His tragedy dramatizes
the relationship between poetry, passion and truth, and identifies a clear
winner, since passion is shown to contain the seeds of truth and lead to its
full discovery.
We are now able to appreciate a fundamental difference in the way
that Oedipus fulfils his responsibilities as protagonist and prime mover
of the tragedy in comparison to Medea or Atreus. He does not enjoy the
privileged, omniscient point of view of the author as they do. Stirred by
passion, he instigates a drama, but one which he cannot control and which
will eventually turn against him. Medea and Atreus act within the plots they
have constructed, while remaining unchallenged masters of their plans.
Their authorial function is always foregrounded and never challenged.
Oedipus, on the other hand, quickly abandons his role as the omniscient
author-on-stage and reveals his nature as an impotent spectator, repeatedly
threatened by events outside his control. The enormous force of Oedipus
dramatic consistency is predicated precisely on his double status as author
and spectator. He sets in motion the search for poetry, which will turn
him into a desperate victim a guilty one, in fact. For us, he embodies
the dangers associated as much with yielding to, as with resisting passions.
Thus his plight dramatizes one of the hermeneutic possibilities offered to
spectators. This tragedy represents the dangers of Oedipus passions, and,
at the same time, the futility of denying that passions have a valid claim to
truth. Poetry is a passion, not only for the enthusiastic author who creates
it, but also for the audience which receives it.
20

v
The contrast between passion and reason, which I have chosen to foreground, is often named as the crucial tension animating these tragedies.
However, I should point out again that I will not focus primarily on
the usual issues concerning the articulation of passions in the plays. My
main topic in this preliminary chapter remains a more specific one: how
passions can be described as the driving force not just behind the actions of
several characters but also behind the very existence of the tragedies as we
read them, and especially how this genetic function is represented in the
tragedies. As we will see in chapter two, this genetic force operates at more
than one level, since passion underlies the Furys determination to put in
motion Atreus revenge, just as passion for Aerope and for power had
motivated Thyestes initial attack on Atreus.
By entertaining the hypothesis that passions might generate tragic poetry we are forced to face a set of familiar questions about the relationship
between the tragedies and the rest of Senecas corpus. Yet one could reasonably claim that those questions are, in a sense, irrelevant. For instance,
it could be argued that we should read each tragedy as a separate and selfstanding unit. Or that the attempt to relate the tragedies at any cost to
Senecan philosophy is a petitio principii: we ask how the tragedies can be
compatible with the authors philosophy because we have already decided
that they should be since they were written by the same person. But, in a
sense, to do so would take away much of the fun. After all, no reader of
Medea or Phaedra can avoid wondering how works of such extraordinary,
even tropical luxuriance could have been penned by the same author who
fiercely (if anything too fiercely), advertises elsewhere the virtues of stylistic
restraint and moderation. The fluid state of Senecan chronology in general,
and not just that of the tragedies, makes it impossible to advance a model
of diachronic evolution and compels us to read the corpus as an unnaturally
static organism, with all its lines of tension prominently and seductively
displayed.
The critical debate on the relationship between philosophy and tragedy
in Seneca revolves around a predictably limited range of options. It is fair
to say that the presumption of a connection, or even the desire to establish
a solidarity of intents between the two domains, is still widespread. The
emphasis, of course, varies widely, between those who tend to see in the
tragedies and the prose work a similar ideological bent, and those more inclined to read in the tragedies a denunciation, if not a complete subversion,
of the prose works restrained optimism. In an interesting, if somewhat
21
dogmatic, monograph Joachim Dingel has argued that the tragedies stand
as a collective rejection of the Stoic philosophical principles advocated
in Senecas prose works, and that they give voice to the authors truer
and deeper feelings.32 One of the fundamental shortcomings of Dingels
Romantic approach lies in the sweeping generalization on which it is
based. By pitting tragedies and prose works against each other he forces
very different texts into two seemingly homogeneous and compact categories. Surely no one would deny the consistency which characterizes the
tragedies as a whole, in matters both stylistic and psychological. And a
similar assessment can probably be made about the prose works, which
display a broader range of modes, styles and tendencies, in part because of
their diverse generic affiliations. But any interpretation of the relationship
between the prose works and the tragedies which downplays the specific
characteristics of each individual work is unsatisfactory, and may at best
offer a general suggestion almost a metaphor of that relationship. As
I mentioned briefly above, it would certainly be rewarding to highlight in
Senecas prose the inner tensions it is often unable to repress.33 Even there,
for instance, the well-known contradiction between the theoretical dictates
of stylistic immediacy and the actual richness and rhetorical complexity
of the style is a strong enough indication that significant conflicts may be
lurking not too deep beneath the surface.
vi
Having briefly attempted to establish the premise that Senecan poetry and
Senecan prose should be considered equally relevant in the attempt to
understand the principles of the authors poetics, I would now like to turn
to the explicit remarks regarding the nature of poetry and poetic inspiration
which Seneca offers at several points in his essays.
In the process of representing the evolution of their plots, Juno, Medea,
Atreus and, to a certain extent, Oedipus reveal the passions of furor, ira
and metus as the sources which will inspire and animate their endeavours.
These passages seem to amount to a very strong case for the genetic connection between passions and poetry, since they indirectly represent poetry
as arising from a deeply passionate realm which has no room for reason.
In his prose writings Seneca confronts this very issue and tries to resolve
in several different ways the obvious tension between an enthusiastic view
of poetry and his teaching on the dangers of passions. In the pages that
32
Dingel (1974).
33
Interesting observations in Moretti (1995) and Too (1994).
22
follow, I address especially these attempts at resolution and connect them

with the metadramatic features that I have already discussed, with the aim
of reconstructing some aspects of Senecas theory of tragedy and tragic
passions.
The connection between poetry and furor (in the sense of enthousiasmos)
dates back to Democritus and Plato.34 According to the Phaedrus (245a),
which Seneca translates at De tranquillitate animi 17.10, the sane mind
(compos sui) knocks in vain at the door of poetry.35 The enthused poet
who transgresses his human limitations to reach out to the sublime nature
of creation is mentioned several times by Seneca, and De tranquillitate
animi offers a particularly interesting set of reflections. In the first chapter
of the dialogue, Serenus voices his misgivings (1.14):
Then again, when my mind has been uplifted by the greatness of its thoughts
(cogitationum magnitudine), it becomes ambitious of words, and with higher aspirations it desires higher expression, and language issues forth to match the dignity
of the theme; forgetful then of my rule and of my more restrained judgement,
I am swept to loftier heights by an utterance that is no longer my own (oblitus tum
legis pressiorisque iudicii sublimius feror et ore iam non meo).
Seneca replies that this need not be interpreted as the sign of a continuing
sickness, but as the natural oscillation of a body not yet accustomed to its
new health. By the end of the book, it is Seneca himself who admits that
a moderate amount of relaxation (remissio, 17.5) and moderation of efforts
(temperamentum, 17.7) are a necessary counterbalance for even the most
temperate of souls. Even a certain degree of drunkenness (ebrietas) can be
welcome (17.8):
At times we ought to reach the point even of intoxication, not drowning ourselves
in drink, yet succumbing to it; for it washes away troubles, and stirs the mind
from its very depths and heals its sorrow just as it does certain ills of the body; and
the inventor of wine is not called the Releaser (Liber) on account of the licence
it gives to the tongue, but because it frees the mind from bondage to cares and
emancipates it and gives it new life and makes it bolder in all that it attempts. But,
as in freedom, so in wine there is a wholesome moderation.
This leads rapidly to the conclusion of the dialogue, the locus classicus
for the Senecan theory of the enthused poet (17.1011):
34
35
Indeed enthousiasmos appears to be the invention of philosophers; see Tigerstedt (1970), with
Finkelberg (1998) 1920.
The most explicit Latin statement for this notion of poetic enthusiasm, ultimately Democritean
and Platonic, is probably Cic. De or. 2.194; see also Tusc. 1.64, and Peases commentary ad loc. for
further indications.
23
For whether we believe with the Greek poet that sometimes it is a pleasure also to
rave, or with Plato that the sane mind knocks in vain at the door of poetry, or with
Aristotle that no great genius has ever existed without some touch of madness: be
that as it may, the lofty utterance that rises above the attempts of others (grande
aliquid et super ceteros) is impossible unless the mind is excited (mota). When it has
scorned the vulgar and the commonplace, and has soared far aloft fired by divine
inspiration, then alone it chants a strain too lofty for mortal lips (aliquid cecinit
grandius ore mortali). So long as it is left to itself, it is impossible for it to reach
any sublime (sublime) and difficult height; it must forsake the common track and
be driven to frenzy and champ the bit and run away with its rider and rush to a
height that it would have feared to climb by itself.
Although Seneca is not engaged here in an explicit declaration of poetics,

and is addressing rather the issue of philosophical reflection, the presence
of the Platonic quotation and the term cecinit (17.11) suggests that the
same state of enthusiastic lack of control lies behind artistic creation and
philosophical excitement.36
The idea that magnitudo animi, greatness of soul is
inextricably connected with magnitudo ingenii, elevation of thought37 and that the latter finds expression in the sublime
() is rooted in Cleanthes theory that poetry, thanks to metre, song
and rhythm, is the only means which can adequately express divine greatness (

).38 Seneca, too, shares the idea that the beauty of things
generates enthusiasm, as he declares in rather extreme terms in Letters to
Lucilius 108.7:
A certain number are stirred by high-sounding phrases, and adapt themselves to
the emotions of the speaker with lively change of face and mind just like the
emasculated Phrygian priests (Phrygii . . . semiviri) who are wont to be roused by
the sound of the flute and go mad (furentes) to order. But the true hearer is ravished
and stirred by the beauty of the subject matter, not by the jingle of empty words
(rapit illos instigatque rerum pulchritudo, non verborum inanium sonitus).
The explicitly irrational overtones that mark the vocabulary of inspiration in this passage and make it so similar to its poetic counterparts
are quite surprising in the light of the Stoic strictures against passions.39
36
37
39
I will return more extensively later to the importance of the sublime in the poetics of Thyestes. See
below, pp. 12732. On the issue in general see Michel (1969).
38 SVF 1.486. See Mazzoli (1970) 47.
See Ps.-Longinus, Subl. 7.3.
It is even more important that these passages confirm quite explicitly that yielding to passions
constitutes also a superior form of knowledge. The enthusiasm of the poet or the philosopher is
what enables him to apprehend

which prose, i.e. a rational and controlled form of
expression, would be unfit to represent (Cleanthes puts it precisely in these terms in the fragment I
alluded to above, n. 38).
24
While, in this particular context, the orgiastic frenzy of the converted is

justified by their sources of inspiration and their goals, more complex problems arise if we try to apply this theory to poetry in general. In his wideranging analysis of the issue, Giancarlo Mazzoli has argued that, following
Posidonius strictures against Chrysippus theory of apathy (
),
Seneca is here embracing Peripatetic elements, namely the notion that
a controlled and moderate excitement can in fact lead to cheerfulness
(!
).40 According to Mazzoli, this explanation holds true for poetry
as a whole. But the vocabulary of Letter 108.7, with its references to the
Phrygian eunuchs (Phrygii semiviri), does not suggest moderation and
control. And even if this explanation can be considered satisfactory in the
specific case of philosophical enthusiasm, it becomes more difficult to apply
it to poetry in general, since poetry is not bound to the exclusive representation of philosophical examples (paradeigmata). Cleanthes Hymn to
Zeus would plausibly fall into this category, but Senecas tragedies, with
their powerful representations of negative examples, are a different matter
altogether.
This problem is enhanced by other Senecan passages which show a considerable degree of ambivalence towards poetry. At De brevitate vitae 16.5
Seneca attacks the frenzy (furor) of the poets that nurtures the errors of
men by offering lascivious images of the behaviour of gods. Furor is recognized here explicitly as an error of the poets who abandon moral themes
and educational messages.41 As other passages make eloquently clear, poets
are not bound by the respect for truth or morality.42 Their inspiration is
potentially dangerous precisely because it transcends, like an oracle (more
oraculi),43 the limits of human rationality: this can lead to the possibility of speaking, with a voice greater than human (grandius ore mortali),44
great philosophical truths, or, on the contrary, of depicting falsehoods in
appealing terms and deceiving mankind. Like Hesiods Muses, who can say
many true things but also many false things resembling truth (Theog. 278),
the poets are ambiguous and ultimately unreliable sources who should be
constantly checked for accuracy and moral worthiness.
I emphasize, again, the chthonic aspect of poetry and the ambivalent
nature of the poet because these are the elements that resonate most dramatically in the tragedies. When he describes the soul of the irate man,
at De ira 2.35.4, Seneca compares it with the terrible fictional underworld
created by poets:
40
43
Mazzoli (1970) 556.

See De brevitate vitae 2.2.
41
See De vita beata 26.6.

44 See Tranq. 17.11.
42
See Ben. 1.3.10; 1.4.5.
25
As is the aspect of an enemy or wild beasts wet with the blood of slaughter or bent
upon slaughter; as are the hellish monsters of the poets brain, all girt about with
snakes and breathing fire; as are those most hideous shapes that issue forth from
hell to stir up wars and scatter discord among the peoples and tear peace all to
shreds; as such let us picture anger . . .
Erinys and the Furies are just fictions, Seneca says, and in fact, as he points
out in Consolatio ad Marciam 19.4, the whole apparatus of punishment in
the underworld is a product of poetic craftmanship:
Reflect that there are no ills to be suffered after death, that the reports that make the
Lower World terrible to us are mere tales, that no darkness is in store for the dead,
no prison, no blazing streams of fire, no river of Lethe, that no judgement-seats
are there, nor culprits, nor in that freedom so unfettered are there a second time
any tyrants. All these things are the fancies of the poets, who have harrowed us
with groundless terrors (luserunt ista poetae et vanis nos agitavere terroribus).
Poets create fictional representations devoid of truth and use them to stir
human souls with empty terrors: the close connection established here
between play (lusus) and terror is particularly striking.
Poetry springs from the same form of enthusiastic furor as that which
generates the inspired sublimity of the philosopher, but it is not confined
to great moral truths. Its morality and educational potential are linked to
its contents: poetry can instruct, but can also deceive and mislead, and
exploit its capacity to strike the mind for objectionable purposes. It has a
positive function when it celebrates the beauty of divine greatness (

), as, again, in Cleanthes Hymn to Zeus. But what happens when
it generates examples of vices, as do so many tragedies? How do these
examples, presented through the powerful means of poetic expression, affect
the audience?
These are the main questions that will accompany this exploration of
Thyestes, a play dominated by the ambiguously attractive figure of Atreus;
a play, to be sure, that makes it exceedingly difficult to evaluate the nature
and purpose of poetry, which is rooted, as it clearly is in Thyestes, in the
disruptive world of passions. I will leave to the very end of this book, after
engaging at length with appealing portraits of evil and less than compelling
attempts at moral rectitude, to examine the effects that the tragic text might
have on the audience, and how it could be deemed compatible with the
Stoic requirement that poetry have an educational function.
chapter 2
Staging Thyestes
the poetics of furor

libet reverti
(Seneca, Agamemnon 12)
Quis inferorum sede ab infausta extrahit

avido fugaces ore captantem cibos?
quis male deorum Tantalo visas domos
ostendit iterum? peius inventum est siti
arente in undis aliquid et peius fame
hiante semper?
Who drags me forth from the accursed abode of the dead, where I snatch at food
ever-fleeing from my hungry lips? What god shows Tantalus again the homes he
saw to his ruin? Has something worse been invented than parching thirst in the
middle of water, worse than ever-gaping hunger?
The Thyestes begins by staging the process of its own construction. Tantalus
not only wonders at the unexpected turn his punishment is taking, but also
questions the very existence the theatrical essence of the drama that
is bringing him on the scene. His questions, while ostensibly bearing on
his fate as a mythic character, also look in anguish at the unfolding of
the tragic action, as if he watches himself from the outside becoming a
character of a dramatic text. Who dragged him from the depths of the
underworld and forced him onto this stage? What is this novel situation
that is worse than hell, one where, paradoxically, he is punished by being
forced to punish others? Similarly, the subsequent fight between the Fury
and Tantalus shadow embodies a creative conflict between passive forces,
on the one hand, which try to resist the dramas violence, and active forces,
on the other, which create and further the dramatic action.
This initial self-reflexive gesture is one of the strategies that complicate
the audiences perception and suggest with increasing intensity a fractured,
conflictual understanding of the text. Violence is encoded already in the
26
Staging Thyestes
27
prologues dialogic form, which opposes two parties with different opportunities and levels of power to reinforce words with deeds. Division, fracture
and conflict impose themselves as dominating forces from the very first
lines, when Tantalus evoked shadow addresses his yet unknown counterpart. Dialogue, it should be noted, appears only in this Senecan prologue,1
and we can better evaluate the implications of this form of expression if we
compare this prologue with that of Hercules furens. The enraged Juno who
delivers the entire prologue of Hercules furens is in many respects parallel
to the Fury: both superhuman characters provide the impetus which sets
in motion the dramatic action, and both correspond in function to the
creative momentum which underlies each tragedy as a whole. Moreover,
Junos words and attitude establish a close connection with the role of Juno
in Aeneid 7, a text which, as we will see shortly, forms an essential backdrop
to the prologue of Thyestes. Yet the two scenes differ significantly: Junos
speech is not a dialogue and does not stage a conflict between a sinful
creative impulse and a moral resistance to the creation of nefas, a conflict
represented in Thyestes by the Fury and Tantalus shadow respectively. This
structural difference deprives the prologue in Hercules furens of the dialectical contrast between silence and speech one which, I will now argue, is
central to Thyestes.
Tantalus has a dramatic consistency which is not altered by the Furys
final, inevitable victory. He is a guilty man, as he readily acknowledges:
I should be the one to suffer punishment, not to inflict it (Thy. 867:
me pati poenas decet | non esse poenam). His moral opposition to the Furys
demand is heightened by this admission. The Fury wants Tantalus, who
resists in vain, to arouse new, terrible scelera on earth. Those scelera are
the tragedy itself, since Thyestes is precisely the tale of a compelling and
memorable scelus.
Textual markers are uniformly pointed: this prefatory debate might be
read as a symbolic enactment of the birth of the play and an open (although
far from neutral) window onto the forces that preside over its creation.
Indeed, if Tantalus firm appeal to moderation had succeeded, there would
be no Thyestes at all. Tantalus is appalled at the request to come back to
earth; his anguish is clear in the repeated questions in lines 15, which I
quoted above. Iterum is the keyword here.2 Tantalus questions the senseless
1
2
Hine (1981) offers a persuasive analysis of the prologue and its thematic links to the rest of the play.
Iterum is often a metaliterary mark. Haupt suggested early on that at Ov. Fast. 3.4712 the adverb
signals Ovids allusion to Catullus Ariadne (Haupt (187576) 71, with Conte (1985) 38). The verb
soleo has similar functions; see Leo (187879) 14955. See later, p. 193, n. 44, for further observations
on this topic and its thematic relevance.
28
drama of re-enactment, and the novel invention (see inventum, line 4).
Indeed, the Fury reminds him that his scelus would not be original: let
the banquet be spread you will come as a guest to a feast of crime well
known to you (623: epulae instruantur non novi sceleris tibi | conviva
venies). But what is personally and morally unacceptable is precisely what
this tragedy and its poetics are made of: re-enactment, repetition, obsessive
return of, and return to, what could (and should) best be left unsaid.3
The tragedy firmly rejects the moral option of silence. The question that
Tantalus utters here for the first time is also the key question of the play as a
whole: why again?4 The Fury inspires scelera, the very scelera that make up
the whole of Thyestes. Thus the Fury effectively inspires this poetry: from
the very beginning of the play, there is no escaping the daunting connection
between poetry and scelera.
Tantalus does try to resist. After much remonstration he finally assumes
a firm and fierce stance (905); he simply will not obey: here I will stand,
and prevent the evil deed (95: stabo et arcebo scelus). His attempts to impart
moral guidance and avoid errors would befit a sage, perhaps even a Stoic
sage. His resistance, however, does not last long, as the Fury tortures him
on stage (96100):
quid ora terres verbere et tortos ferox
minaris angues? quid famem infixam intimis
agitas medullis? flagrat incensum siti
cor et perustis flamma visceribus micat.
sequor.
Why do you terrify me with the sight of your lash, and fiercely threaten me with
your twisted snakes? Why do you rouse pains of hunger deep in my innards? My
heart burns with fiery thirst, and in my burnt-out vitals a flame is darting I follow
you.
Tantalus words attest to the impossibility of his moral stance and of his
didactic purpose. The vehement language which describes his intentions
3
The prologue of Agamemnon is centred as well on the topic of return and reiteration. While the
relative chronology of Ag. and Thy. in Senecas literary production cannot be certain, in mythical
time the actions of Ag. come after those narrated in Thy. and are in fact a direct consequence of
them (on the dating of both plays see p. 5, n. 7). The ghost of Thyestes in the prologue effectively
recalls Tantalus shadow, especially since they both claim to prefer the underworld to the devastation
awaiting them on earth. Thyestes exclamation at line 12 libet reverti means precisely that he
would rather return to the underworld than assist in the terrible revenge which is about to happen.
It should be noted, however, that according to Fitchs metrical study (Fitch (1981)), Ag. would have
been written before Thy. For a more sceptical position on the dating of Ag., prior to Fitch (1981), see
Tarrant (1976) 56. The whole issue is reassessed in Nisbet (1990).
It is tempting to charge transcribor (13) with metadramatic resonances, if for no other reason than
its etymological reference to writing. The verb is seldom used in poetry, see Virg. Aen. 5.750, 7.422;
Ov. Ibis 187; Met. 7.173. See Tarrant (1985) 89 and Jacobi (1988) 153.
Staging Thyestes
29
(moneo, stabo, arcebo)5 is suddenly and irrevocably reversed in the bitterly ironic repetition of a Stoic-sounding sententia: sequor is what the
sage should say when facing destiny, since it is better to follow willingly
than be dragged.6 Naturam sequi following Nature is the paramount
principle of a truly Stoic life: Tantalus does after all respect this intimation, his true nature being germane, rather unsurprisingly, to that of the
Fury.7 The Furys power is the power of unavoidable destiny. The Fury is
the Muse of scelus, and her victory is the victory of poetry (of this particular brand of poiein) against the repressive silence advocated in vain by
Tantalus.
Tantalus pained questions at 969 synthesize a number of associations
that Thyestes will repeatedly explore. First, his doomed resistance to the
Furys instigation recalls the similar reaction that seers display when the god
violently overpowers them and forces them to speak compare the violence
of Apollo on the Sybil at Aeneid 6.7780 and 1001.8 Secondly, the particular choice of images makes the languages of erotic desire and creative
impulse intersect, and cruelly deprives both of their comforting metaphorical value. The standard association of fire and eros connotes the Furys
order as an irresistible, sinful desire redolent of erotic passion.9 At the same
time, these images are bound up with the vocabulary of poetic enthusiasm,
the burning power which moves poets to create.10 In this respect the passage anticipates a central moment later in the play, where a merciless Atreus
5
9
10
Compare the behaviour of virtus in De vita beata 15.5 (illa fortiter stabit et quidquid evenerit feret); the
fact that the sapiens remains unperturbed in the face of natural disasters: stabit super illam voraginem
intrepidus (Q Nat. 6.32.4); Ben. 5.2.4: (vir bonus) ad ultimum usque vitae diem stabit paratus et in
hac statione morietur; and Jocastas attempt to stop the massacre at Thebes: ibo, ibo et armis obvium
opponam caput, | stabo inter arma; petere qui fratrem volet, | petat ante matrem (Phoen. 4079). The
original suggestion, however, is probably Virgilian; see Aen. 7.3735 (Allecto and Latinus): his ubi
nequiquam dictis experta Latinum | contra stare videt, penitusque in viscera lapsum | serpentis furiale
malum totamque pererrat.
Sequor was athetized by many editors, starting with Bentley, but modern texts retain it (see Hine
(1981) 2678 for a defence of Bentleys decision). Its relevance for the characterization of Tantalus
is crucial, and the parallel with Thyestes sequor at 489 seems decisive. See also Plaut. Trin. 12, a
passage with interesting points of connection with the whole scene at hand; see below, n. 18. There
are other instances of half-lines in the Senecan corpus, such as Phoen. 319, Tro.1103 and Phaed .605.
See Calder (1984); Tarrant (1985) 103; and Zwierlein (1986) 298.
Cf., for instance, Senecas use of sequor in De vita beata 15.6 (with Schiesaro (1996)), where he
claims that the principle always to be followed is deum sequere, follow god! An emphasis on sequi
occasionally lends Aeneas a Stoicizing connotation; see M. W. Edwards (1960).
At Aen. 6.7780 the language (excussisse, fatigat, domans, fingit premendo) alludes to the taming of
horses, which could also be suggested by the use of verbere at Thy. 96 (cf. the whipping motion
suggested by 101, on which below, p. 178, n. 6).
For the metaphoric associations of ardere and related images see Fantham (1972) 1011 and 878.
On the warmth of inspiration see Ov, Fast. 6.56: est deus in nobis, agitante calescimus illo: | impetus
hic sacrae semina mentis habet and especially Ov. Tr. 4.1.414, in explicit connection with Bacchic
inspiration: utque suum Bacche non sentit saucia vulnus, | dum stupet Idaeis exululata modis, | sic
ubi mota calent viridi mea pectora thyrso, | altior humano spiritus ille malo est. For the association of
30
strives to find inspiration for his new creation the plot of his revenge and
of the rest of the tragedy (2504):11
dira Furiarum cohors
discorsque Erinys veniat et geminas faces
Megaera quatiens: non satis magno meum
ardet furore pectus, impleri iuvat
maiore monstro.
Let the fearful band of Furies come, the discord-sowing Erinys, and Megaera,
shaking her twin torches; the frenzy burning in my breast is not great enough;
some greater horror must fill me.
Atreus, far from being tortured by the Furies, appeals to them for the
impulse to act creatively, to produce in effect a dramatic text which he
will perform and inflict upon Thyestes. Both passages implicitly depict the
Furys impulse as a sinful desire which finds its final realization in poetry.
This furor cannot be resisted, and the play enacts the sinister force of a
victory against morality, reason and fas.
In this same prologue Seneca offers intimations about the literary background in which we should situate these declarations of poetics. Three texts
in particular stand out, belonging to three authors whose presence looms
large in Senecas tragic world: Euripides, Ovid and Virgil.12
Friedrich Leo first pointed out a connection between the prologue of
Thyestes and the dialogue between Iris and Lyssa in Euripides Hercules
furens (82274).13 The actual verbal coincidences are faint, but the overall
structure of the dialogue is close to Senecas scene in important details. At
the beginning of the second part of the play, the two infernal characters
appear on stage to lay the curse of madness on Heracles. Like Tantalus
shade, Lyssa does try to resist Iris commands, albeit unsuccessfully (Eur.
HF 84354). No physical torture ensues, but Iris replies scornfully to Lyssas
noble attempt at changing her mind: Zeus wife did not send you here
to display wisdom (857:

).
11
12
13
calor and prophetic inspiration, see Ov. Met. 2.641 incaluitque deo, quem clausum pectore habebat
(Ocyrhoe).
On this passage see p. 46 below.
A connection with Accius Atreus, cautiously suggested by La Penna (1979) 136, n. 1, cannot be ruled
out. According to La Penna, inc.inc. lviii and lix Ribbeck2 must refer to Tantalus, and might thus
belong to the prologue.
Leo (1912) 2012. Calder ((1983) 1856) compares the scene with the dialogue between Hephaestus
and Kratos at the beginning of Aeschylus Prometheus (187), mediated through Sophocles Oedipus
at Colonus, and argues that Sophocles, not Euripides, is Senecas primary model.
Staging Thyestes
31
As she prepares to yield, Lyssa invokes the sun as witness of her resistance
and unwillingness to accomplish the monstrous deed demanded by Iris
(85873). At the end of her speech, which illustrates in great detail the events
that will soon follow (and is thus similar to the anticipatory function of the
Senecan prologue), Lyssa sends Iris back to Olympus and enters the house
where she will wreak destruction (8723).14 In Thyestes Tantalus seems to
perform both actions, since he first enters the house, and then is sent back
to the underworld by the Fury.15
Senecas Fury herself offers at least one other strong signal of intertextual
self-awareness in her first, most effective speech (Thy. 547):
ornetur altum columen et lauro fores
laetae virescant, dignus adventu tuo
splendescat ignis Thracium fiat nefas
maiore numero.
Decorate the lofty column and let the doors be green with festive laurel; a fire
worthy of your arrival must shine brightly then let the Thracian crime be done,
but multiplied.
Through allusive amplification (maiore numero)16 the Fury inaugurates here

a map of intertextual connections which will prove crucial for the whole
play: the Thracian nefas par excellence is the bloody story of Tereus, Itys and
Procne, especially as told in Ovids Metamorphoses.17 The Fury underlines
her truly metadramatic function by showing her knowledge of mythical
and literary history, and by explicitly alerting us to the allusive resonances
of the play. The Furys intertextual competence is one of the ways in which
she acquires a metadramatic status: her references to other poetic texts reflect the genesis of the play and its modes of signification, and go well
beyond the dramatic level acted on the stage. What she effectively proposes
and realizes is a self-conscious mise en abyme of Ovids and Sophocles story
which will span the remainder of the play. The agonistic stance expressed
in maiore numero acknowledges the new dimension that this dramatic repetition of nefas will assume. As we will see shortly, any repetition of nefas
14
15
16
17
The chorus notes the Bacchic connotation of her actions in the house:

(897), a
significant detail in view of the Bacchic overtones of Thyestes; see pp. 1338. A further element of
contact between this scene and Senecas play can be detected at HF 8656, where Lyssa specifically
points out that Heracles will not be aware that he is killing his own children.
The succession of these actions is somewhat problematic; see below, pp. 17880.
On the thematic importance of maius and related concepts see pp. 1301. Maius in relevant contexts
can often also encode a statement of poetics, as in Virg. Ecl. 4.1 and Aen. 7.44 (where the Iliadic
part of the poem is greater than the first part, just as the Iliad is greater than the Odyssey).
See later, pp. 17980.
32
is necessarily worse than its model more obsessive, more painful, more
guilty. At the level of poetics, the repetition will encourage the exploration
of a more intense and emotionally loaded language of recursive patterns
and of elaborate internal echoes.18
The most evident intertext for the prologue of Thyestes, however, is
the opening scene of Aeneid 7.19 The sequence of events in Virgils poem
is more intricate, but the fundamental pattern is very similar. Enraged
by the apparent triumph of the Trojans, who have finally landed in Italy
(7.286322), Juno summons Allecto and commands her to bring discordia
and destruction into the Latin field (7.33540), thus igniting the war against
Aeneas and his people:
tu potes unanimos armare in proelia fratres
atque odiis versare domos, tu verbera tectis
funereasque inferre faces, tibi nomina mille,
mille nocendi artes. fecundum concute pectus,
dissice compositam pacem, sere crimina belli;
arma velit poscatque simul rapiatque iuventus.
You can take brothers who love each other and set them at each others throats.
You can turn a house against itself in hatred and fill it with whips and funeral
torches. You have a thousand names and a thousand ways of causing hurt. Your
heart is teeming with them. Shake them all out. Shatter this peace they have agreed
between them and sow the seeds of recrimination and war. Make their young men
long for weapons, demand them, seize them!
The Furys orders to Tantalus strongly echo Junos words (Thy. 836):
ante perturba domum
inferque tecum proelia et ferri malum
regibus amorem, concute insano ferum
pectus tumultu.
First throw your house utterly into confusion, and bring in strife along with you,
bring passion for the sword, the bane for rulers, and with wild upheaval strike the
savage breast.
18
19
In spite of some notable differences (Calder (1983) 196, n. 22), it is also worth noting the possible
connection (Leo (1912) 202) with the prologue (13) of Plautus Trinummus, where Luxuria and
Inopia, mother and daughter, set in motion the action of the comedy: LU. sequere hac me, gnata, ut
munus fungaris tuom. | IN. sequor, sed finem fore quem dicam nescio. | LU. Adest. em illae sunt aedes,
i intro nunciam. There is no traumatic conflict in this dialogue, which is quickly resolved by Inopias
only slightly puzzled obedience. Alone on the stage, Luxuria embarks on an extended and explicitly
metadramatic monologue, addressing the audience directly and giving precise indications about the
play that has just begun.
Tarrant (1985) has useful notes (listed at p. 85, n. 2) on several of these passages. The model is
pointed out by Monteleone (1980) 77. See Timpanaro (1981) 1278, for a thorough analysis of the
connections between Juno in Hercules furens where the goddess acts again as the primum mobile
of furor and in the Aeneid .
Staging Thyestes
33
The vivid expression concute pectus as well as the insistence on domus as

the target (Aen. 7.336 and Thy. 83) and the use of inferre (7.337 and 84) link
the two texts. Furthermore the results of Junos and the Furys destructive
orders are alike; the Fury points out to a bewildered and reluctant Tantalus
how his presence has affected the house of the Pelopides (Thy. 1037):20
sentit introitus tuos
domus et nefando tota contactu horruit.21
actum est22 abunde. gradere ad infernos specus
amnemque notum; iam tuum maestae pedem
terrae gravantur.
Your house feels your entering and has recoiled in horror from your unutterable
contagion. Enough! More than enough! Go to the caves of the underworld and
your familiar river; already your step falls heavily on the saddened earth.
The Furys words resonate with Junos final admonition to Allecto (Aen.
7.5524):
terrorum et fraudis abunde est:
stant belli causae, pugnatur comminus armis,
quae fors prima dedit sanguis novus imbuit arma.
There is enough terror and lying. The causes of war are established. They are
fighting at close quarters and fresh blood is staining whatever weapons chance first
puts into their hands.
When Allecto, at Junos request, returns to Acheron, she relieves earth

and sky of their painful burden. Terrae gravantur at Thyestes 107 elaborates
on Virgils description of the forces of evil once again oppressing the earth
(Aen. 7.56871):
hic specus horrendum et saevi spiracula Ditis
monstrantur, ruptoque ingens Acheronte vorago
pestiferas aperit fauces, quis condita Erinys,
invisum numen, terras caelumque levabat.
Here they point to a fearful cave which is a vent for the breath of Dis, the cruel god
of the underworld. Into this cave bursts Acheron and here a vast whirlpool opens
its pestilential jaws, and here the loathsome Fury disappeared, lightening heaven
and earth by her absence.
20
21
22
Lines 1037 foreground the issue of the temporal structure set out in the prologue: see below, ch. 5,
pp. 17880.
On the possible Bacchic overtones of this image see below, p. 138, n. 60.
This use of ago (which is absent in the Virgilian model) can be considered a technical theatrical
term; see OLD s.v. 25 and 43. But the very form actum est is often used to describe a situation which
has deteriorated beyond repair, and in the Furys mouth it is appropriate to preserve especially this
sense of ultimate destruction.
34
In Senecas prologue, however, the impact of this founding scene extends

further, beyond the direct association between the Furys and Allectos acts of
pollution. The second part of the dialogue between the Fury and Tantalus,
for instance, echoes Allectos fateful visit to Turnus at Aeneid 7.40674.
The metaphoric taedae with which Allecto excites Turnus after his initial
refusal (with these words she threw a burning torch at the warrior and it
lodged deep in his heart, smoking with black light sic effata facem iuveni
coniecit et atro | lumine fumantis fixit sub pectore taedas, 7.4567) become the
all too real fires employed by the Fury to bend Tantalus well-intentioned
reluctance at Thyestes 989. Here we should contrast Turnus self-assured,
even overtly mocking reaction to Allecto, appearing to him in the shape of
old Chalybe, with Tantalus high-minded, explicitly moral concerns.
Another significant point of contact between the two texts can be established. Before approaching Turnus, Allecto had successfully stirred Amata
to action (7.38590):
quin etiam in silvas simulato numine Bacchi
maius adorta nefas maioremque orsa furorem
evolat et natam frondosis montibus abdit,
quo thalamum eripiat Teucris taedasque moretur,
euhoe Bacche fremens, solum te virgine dignum
vociferans
Not content with this, she flew into the forests, pretending that she was possessed by
Bacchus, and rose to greater impieties and greater madness by hiding her daughter
in the leafy woods, hoping to cheat the Trojans out of the marriage or delay the
lighting of the torches. Euhoe, Bacchus! she screamed. Only you are worthy of
the virgin . . .
The diffracted allusion to this passage is to be found in the second prologue

of Thyestes, where Atreus deliberates the best way to obtain his revenge. In
act 2 Atreus is under the inspiring spell of furor and ira introduced into
his house by Tantalus and the Fury: structurally, this position coincides
with that of Aeneid 7. Atreus words at Thyestes 2524 the frenzy burning
in my breast is not great enough; some greater horror must fill me (non
satis magno meum | ardet furore pectus, impleri iuvat | maiore monstro) are
redolent of Amatas maius nefas and maior furor at 7.386. Maius, at any
rate, is the hallmark of Atreus monologue throughout.23 The connection
between the two scenes, supported by structural as well as lexical arguments,
sets up an association between Atreus and the world of Bacchic frenzy: this
23
Seidensticker (1985) identifies in the maius-motif the binding element of this play, and indeed a
fundamental characteristic of Senecan drama in general.
Staging Thyestes
35
is a relationship whose importance, as we will see, goes well beyond the

scope of this specific instance.24
By pointing directly at Aeneid 7 and acknowledging Virgils archetypal
role as a poet of furor, Seneca reconstructs a meaningful sequence of literary
history and invokes a powerful model for his own nefarious endeavours.
Seneca establishes an important continuity with the second half of the
Aeneid.25 The thematic connection, as we have seen, is clear. Junos very
words to Allecto you can take brothers who love each other and set them
at each others throats (7.335) leave us in no doubt that the new battles
in store for the Trojans will be of a different kind from those told in the
first six books. The merging of Trojans and Latins at the end of the poem
retrospectively casts their conflict as a civil war. Seneca now shows that the
fratricidal origins of Roman history, symbolically enshrined in the conflict
opposing Romulus to Remus, reach even further back.26 Yet it is precisely
this tale of horrors which is Virgils greater work (maius opus, 7.45), his
higher order of things (maior rerum . . . ordo, 7.44), or, indeed, the maius
nefas of Amata (7.386) which Atreus, too, will strive to emulate. These
Virgilian echoes also show how Thyestes condenses the horrors of civil strife
in the polarized contrast between two brothers. The theme had obvious,
obsessive resonances in the culture of the first century.27 By alluding to
Virgil, Seneca implicitly reflects, too, on a crucial issue of poetics, the same
issue that Tantalus had raised in his opening speech: why again? He also
defines his own writing as repetition, as a painful, irresistible return to
horrors which have already been sung. Seneca thus situates his tragedy
in a tradition of Juno-inspired poems (and actions) whose authoritative
model he traces back to Virgil: these poems are characterized by the violent
subversion of an ordered world structure guaranteed by Jupiter, and allied
with the chthonic (and, crucially, female) forces of irrational passions and
desires.28 But Thyestes differs from the Aeneid in another relevant detail.
Whereas in Aeneid 7 Allecto is instructed by Juno, in the tragedy her
counterpart, the Fury, acts of her own accord, and the absence of a divine
figure prevents a further displacement of moral responsibility on the gods.
The Fury has learnt her lesson and now acts on her own initiative.
The second half of the Aeneid impresses on the reader a set of ethical
dilemmas which the first, for all the suffering it described, did not present
24
25
26
27
28
See ch. 3, passim.

On Seneca and Virgil see especially Putnam (1995) 24685.
Hardie (1993) 23; Quint (1993) 79; see Hor. Epod. 7.1; Luc. 1.95; Virg. Aen. 7.317; 12.313.
A useful survey can be found in Frings (1992).
On the model of epic inspired by Juno, as opposed to the epic under the sign of Jupiter, see now
Hershkowitz (1998) 95124.
36
so poignantly. (The final books of the poem are also much more extensively
engaged in the detailed description of killing one thinks particularly of
the large battles in 10 and 11 and thus confront the reader with the same
questions about the aesthetic appeal of violent representation which are
inevitable for Senecas and Lucans audience. In this respect it is important to stress the continuity of first-century literature vis-à-vis its Virgilian
model.29 ) At the beginning of book 7 a happy ending is within reach;
hence Virgils (and Junos) decision to start the poem all over again, as it
were, with a second proem, and to expand it considerably with a detailed
account of a quasi-fratricidal strife is all the more disturbing. Poetic innovation and moral responsibility run hand in hand; the latter six books are
more troubling because they represent new and unnecessary amplifications of the plot, because they reproduce the physical horrors of war which
the Trojan exiles hoped to have left behind, and above all because they give
voice to the unsurpassed evil of civil war.
All the moral implications of these poetic strategies are active in Senecas
text, which as a whole stands as a challenge to the repressive decorum of
silence. Thyestes repeatedly presents ethical instances (which are more or less
convincing, more or less hypocritical) and pits them against a subversive
passion which generally gains the upper hand. It is around ethics and its
enemies that the play enacts the struggle between repression and subversion.
From an ethical point of view, then, Seneca is as guilty as Virgil, since
he chooses to retell a story whose devastating contents he knows well: once
again, to sing of nefas is in a sense to perpetrate it.30 Seneca raises the stakes
of his moral conflict by giving voice at the beginning of the play to an
alternative which the Aeneid had only implied: Tantalus does proclaim his
intention to steer away from the Fury and her orders, but his ultimate
defeat only amplifies the horror of nefas. Yet this is precisely what Virgil
had done, and Seneca attempts in turn to displace moral responsibility
by invoking such a mighty predecessor. The spiral of violence and poetry
about violence, it seems, offers no escape.
ta n ta lu s to n g u e
libet loqui pigetque
(Seneca, Phaedra 637)
The presence of a perceptible metadramatic level in the prologue of Thyestes

implies a complex of voices, motives, contrasting forces and cross-references
29
See Narducci (1979) 809.
30
A point very effectively made by Masters (1992), esp. 10.
Staging Thyestes
37
which must undermine a moral, didactic reading. The metadramatic dimension acts as a bent mirror, which multiplies and distorts, complicates
and blurs our perception. The central action of the play Atreus revenge
will have to be perceived by the audience within the alienating frame
provided by the prologue, with its discordant attempts at establishing responsibility and causal connections. In experiencing the play as a whole,
we cannot forget what the prologue implies for the rest of the tragedy: an
intrinsic complicity between tragic nefas and its representation is imposed
on the audience in the revelation of the Furys backstage deliberations.
The audience is made to realize that the aesthetic pleasure afforded by the
play is coextensive with that nefas, since the prologue, in offering a tangible
embodiment of the power of poetry and its source of inspiration, has established that connection before their very eyes. We can walk out there and
then. But if we keep watching (or reading), we forfeit our claim to nave
innocence. The plays very existence, the prologue tells us, is guaranteed by
the unrelenting evil of the characters who perform it, such as the Fury and
Atreus.
Only if we neglect the structural importance of the prologue is it possible
to locate the essence of the play in the contrast between two ethical types,
Atreus and Thyestes, whose actions may bear comparison with, for instance,
a character described in De ira or some sort of Stoic proficiens.31 Constrained
by the supposedly clear-cut choice between a Stoic proficiens with suspect
credentials and a blood-thirsty, monstrous tyrant, the audience might have
reasons to doubt the poignancy of the play and its emotional impact. The
real dramatic and emotional crisis of the tragedy lies not so much in the
tension between Thyestes and Atreus both of whom, for different reasons
which I will consider further on,32 are unlikely ethical prototypes as in
the ethically troubling connotations of the very act of representation that
is foregrounded by the prologue. From this point onwards the audience
will be continuously forced to negotiate the conflicting aspects of that
representation pleasure and pain, moral horror and, at a certain level, an
implicit acceptance of that very horror.
The Fury and Tantalus establish in a tense dialogue the connection between
their deeds and the tragic text they will bring to life. The Fury intensifies
31
32
For information on previous treatments of the play along these lines, and a new proposal, see Lefèvre
(1985). I will consider below (pp. 16376.) to what extent the chorus is left unscathed by the cognitive
turmoil that is forced on the audience in the prologue and can thus be seen as a reliable incarnation
of a superior moral stance.
See below, pp. 13951.
38
the tragic plot by conquering Tantalus resistance and instigating him to

pollute the house of his descendants (Thy. 239):
perge, detestabilis
umbra, et penates impios furiis age.
certetur omni scelere et alterna vice
stringatur ensis; nec sit irarum modus
pudorve, mentes caecus instiget furor,
rabies parentum duret et longum nefas
eat in nepotes;
Forward, cursed shade, and drive your sinful house with fury. Make them vie in
every kind of crime and draw the sword on either side; let there be no limit to
shame in their anger; let a blind fury incite their souls; make the rage of parents
last, and make the long trail of sin reach their childrens children.
What she is plotting is indeed a nefas, as she reiterates at line 56: let the
Thracian crime be done, but multiplied (Thracium fiat nefas | maiore numero). In a series of instructions spanning lines 52 to 67, the Fury effectively
organizes the staging of the drama. Fiat nefas (56), epulae instruantur (62:
let the banquet be spread) and, finally, mixtus in Bacchum cruor | spectante
te potetur (656: let blood mixed with wine be drunk before your eyes)
articulate in a careful sequence the fundamentals of her plot, the specific
form the pollution will take, and the audience which is supposed to be
watching the performance. The Fury insistently declares her intention to
subvert the moral order of events, to give space (and voice) to what is normally repressed and silenced. The language of subversion defines her speech
at several critical junctures. She begins by voicing her desire that the two
restraining qualities of modus (26) and pudor (29) give way to furor (27),
rabies (28) and, of course, nefas (28). In a similar vein, she cautions that no
repulsion for novel crimes will be tolerated: give no one time to hate a past
crime let a new one unceasingly arise (2930: nec vacet cuiquam vetus |
odisse crimen: semper oriatur novum). Ira will know no bounds, will in fact
overturn all prohibitions: let there be nothing which wrath deems forbidden (39: nihil sit ira quod vetitum putet). In the war between morals and
immorality that she is at the same time describing and waging, there will
be one clear winner, which will rejoice over previously powerful leaders of
people: Libido victrix (Lust triumphant, 46). The hierarchy of crimes will
also be overturned: in the house of the Pelopidai, she assures, stuprum will
become levissimum (a trivial crime, 47). Her conclusion is sweeping and
unequivocal: let fraternal sanctity and faith and every right be trampled
under foot (478: et fas et fides | iusque omne pereat). Nefas will hold sway in
Staging Thyestes
39
a completely subverted, topsy-turvy world, from which fas has been utterly
banished.
It is worth lingering on the formulations that the Fury adopts in describing her project. She insists on a subversion of values which can best be
described as a denial of accepted norms and values: fas will be replaced by
nefas. Both of these terms encode an important linguistic aspect: originally,
fas and nefas referred to days in which certain kinds of utterances were
allowed or forbidden.33 By ordering fiat nefas, the Fury is precipitating the
speaking of unspeakable crimes which will exist for us precisely because
they will be written, spoken, represented. Tantalus desire to preserve fas,
and resist the attack of nefas, is equally couched in linguistic terms (8995):
ducam in horrendum nefas
avus nepotes? magne divorum parens
nosterque (quamvis pudeat), ingenti licet
taxata poena lingua crucietur loquax,
nec hoc tacebo: moneo, ne sacra manus
violate caede neve furiali malo
aspergite aras. stabo et arcebo scelus.34
Shall I, their grandfather, lead my grandsons into horrible crime? O great lord of
gods, and my father too (though this fact may cause you shame), even though my
tongue be condemned to severe punishment and tortured for speaking, I will not
withhold even this; I warn you, do not defile your hands with execrable slaughter;
do not stain your altars with a madmans crime. Here will I stand, and prevent the
evil deed.
The dialectic of free speech and repression articulated in these intricate

lines is revealing. In the subverted world foreseen by the Furys forceful
advocation of nefas over fas, in both its linguistic and moral dimensions,
Tantalus feels that his own ability to announce the moral injunctions he
wants to deliver is painfully restrained by the Furys torture. His lingua . . .
loquax (92) will be punished for trying to advocate the values of fas, which
no longer have any place in a world dominated by the Fury. Later on,
the play will pointedly pit Atreus resourceful wordiness against Thyestes
33
34
As Varro explains in his definition, according to which the dies nefasti are those in which the praetor
cannot utter (nefas fari) the official formulae do, dico, addico (De lingua Latina 6.30); cf. Ov.
Fast. 1.478.
Scelus and nefas are used in largely overlapping fashion by Seneca (in spite of, for instance, the
distinction suggested by Cic. Paradoxa 25). Nefas, however, foregrounds in its very semantic structure
a conflict between talk and silence which is central to my interpretation of the play. A significant
instance of this contrast is found in the exchange between the messenger and Theseus at Sen. Phaed .
9913: NUNT. o sors acerba et dura, famulatus gravis, | cur me ad nefandi nuntium casus vocas? | TH.
ne metue cladis fortiter fari asperas.
40
inability to recognize that the words he hears are shifty signifiers in a scheme
that is too elaborate for him to understand. Even at his most tragic Thyestes
is not nearly as articulate as his rhetorically proficient brother. Groping in
vain for an explanation of his despair as the tragedy reaches its denouement,
he privileges non-verbal forms of communication, tears, moans, gestures:
grief loves accustomed tears; miserable people have an ominous desire to
weep. I feel like uttering ill-omened laments; I feel like tearing apart my
clothes, full of Tyrian purple; I feel like shrieking (9526: maeror lacrimas
amat assuetas, | flendi miseris dira cupido est. | libet infaustos mittere questus, |
libet et Tyrio saturas ostro | rumpere vestes, ululare libet). Soon afterwards,
once Atreus crime has been unveiled in all its magnitude, words again seem
to fail him: what cries in my misery shall I utter, what complaints? What
words will suffice me? (10367: quas miser voces dabo | questusque quos? quae
verba sufficient mihi?).35
Tantalus, on the other hand, had once been a victim of his own excessive verbosity. The mention of his impertinent tongue recalls a salient
aspect of his mythological record, the fact that he had been punished by the
gods for revealing their secrets to human beings. The prologue of Euripides Orestes refers explicitly to Tantalus intemperance (), and,
largely through Euripides influence, he is later consistently identified as the
paramount example of supremely audacious verbal hybris.36 In Senecas
Thyestes, Tantalus hybris, paradoxically, consists in verbalizing a moral restraint: the cruel thwarting of his attempt in the midst of painful torments
graphically exposes how, in a Fury-dominated world, there is only room
for advocating nefas, for voicing and acting the language of pollution and
crime.37
The prologue of Thyestes provokes an uneasy reflection on the very nature
of theatrical experience,38 but its intricate and ambivalent intersection of
silence and speech has much in common with another first-century literary
work, Lucans Bellum Civile. Just before the climactic moment of the battle
at Pharsalus, Lucan, who is fraught with doubts and worries about the
35
36
37
38
The messenger confesses his difficulty in reporting Atreus crime at 684: quis queat digne eloqui?,
a topical declaration of inadequacy (see Tarrant (1985) 189). But Thyestes words are motivated
well beyond their topicality.
Willink (1983) 32; cf. Willink (1986) 7980. Ovid, for instance, calls him a taciti vulgator (Am. 3.7.51)
endowed with a garrula . . . lingua (Am. 2.2.44).
Critics have connected the contrast between voice and silence in authors such as Ovid with the
issue of free speech in the Principate; see Baldo (1989) and Feeney (1992).
It is useful to keep in mind as a background the Romans tormented attitude towards actors and
their ethical status; see C. Edwards (1993) 99, and the whole chapter for discussion of the actors
ambivalent role in society.
Staging Thyestes
41
very nature of his task, entrusts a pained reflection to the narrators voice
(7.5526):39
hanc fuge, mens, partem belli tenebrisque relinque,
nullaque tantorum discat me vate malorum,
quam multum bellis liceat civilibus, aetas.
a potius pereant lacrimae pereantque querellae:
quidquid in hac acie gessisti, Roma, tacebo.
Mind of mine, shun this part of battle and leave it to darkness, and from my words
let no age learn of horrors so immense, of how much is licensed in civil war. Better
that these tears and protests go unheard: whatever you did in this battle, Rome,
I shall not tell.
The prologue of Thyestes explodes the contradiction which Lucan so acutely

faces, and which pervades the very structure of his poem. Having chosen to
sing of Pharsalia and the evils of civil war, Lucan atones for his guilty projects
by repeatedly delaying the narrative process, postponing (or even silencing)
the revelations of truths that are as historically inevitable as they are morally
shocking.40 But in Thyestes there is no narrator who can negotiate the
conflicting demands of his project. The stage for the play is bare, and good
and evil must confront each other in isolation. The outcome of the match
is predictable, yet it is still important to notice how little opposition nefas
receives, here or elsewhere, in the play. We will see a number of instances in
which delay or resistance are at work, but they will be silenced even more
swiftly than Tantalus compunctions are laid to rest in the prologue. Unlike
an epic poet, the dramatist Seneca is not statutorily supposed to sing the
great (positive) deeds of the past,41 and the proem to Thyestes carries Lucans
logic one step further as it radically denies the act of resisting the poisonous
advance of nefas.
Tantalus, as we have seen, is punished a second time, but for reasons
antithetical to those which warranted his penalty in the well-ordered divine
cosmos of Greek myth: there he had been punished for revealing divine
secrets, for voicing a nefas; here he is punished for trying to voice a fas,
a lesson of moral restraint. The comparison between his current and past
predicaments highlights very clearly the violent subversion of rules which
39
40
41
Narducci (1979) 33; OHiggins (1988) 21516; Feeney (1991) 277.

A point which is brilliantly stressed by Henderson (1987) and Masters (1992).
Feeney ((1991) 277, n. 119) rightly refers to Virg. Aen. 9.4469 and especially 10.793 as explicit
declarations of this purpose. It might be added that a direct, phonic echo of 10.7913 (tuaque optima
facta, | . . . | non equidem nec te, iuvenis memorande, silebo) could be traced in Luc. 7.556 (quidquid
in hac acie gessisti, Roma, tacebo); for similar examples of phonic allusion see Conte and Barchiesi
(1989).
42
follows the gods fall from a position of control in the universe and their
substitution by the Furys infernal power. Tantalus has not been able to
understand that the world has changed, that, in the world of nefas (of Libido
victrix), fas is inevitably suffocated: his restraint would effectively amount
to the suppression of the nefas embodied by the Furys poetic performance,
which cannot, will not, be tolerated. Tantalus, himself a victim of repetition,
is once again forced to heed his mythical propensity for verbal hybris; he
tries to give his hybris a moral purpose by putting it to the service of morals,
but he is quickly and violently disappointed.
Tantalus own past error and present behaviour underscore the essential
nature of the confrontation between himself and the Fury as a contrast between a repressive silence, which would deny nefas any means of expression,
and an avoidance of scelus, which is also realized in linguistic terms. At the
same time, this reference buttresses the equation of silence and inaction on
the one hand, and word and action on the other.
The prologues articulation of a dialectic between repression and its removal scaffolds the creative struggle represented in the play. Indeed it invites consideration of the basic structure of the play in the light of the
assumption that literature might indeed be a return of the repressed made
available to a community of men but rendered harmless by sublimation and
fiction.42 In a series of powerful monographs now grouped under the general title Literature, reason, and the repressed. Three Freudian Studies,43
Francesco Orlando has over the past thirty years developed a coherent
theory of literature that is rooted in Freuds reflections on the linguistic
characteristics of Witz. By exploring extensively the linguistic analogies of
Witz and poetry, both of which are indebted to the peculiar logical forms of
the unconscious,44 Orlando has argued that literature exploits the formal
return of the repressed which creates pleasure by a number of expressive
devices precisely as Witz does and at the same time conveys, in an
institutionally acceptable form, contents which would be subject to partial
or total social censorship. Orlandos own masterful readings of Racines
Phèdre45 and Molières Misanthrope,46 as well as of a number of philosophical texts from the Enlightenment, testify to the theorys wide applicability
42
43
44
45
Orlando (1978) 19 and 1378.

Orlando (1971), (1973), (1979) and (1982): the first two volumes are available in English as Orlando
(1978). See also, more recently, Orlando (1993).
In his later works Orlando draws extensively on the ground-breaking work of Ignacio Matte Blanco,
who provides an exhaustive formulation of the logic of the unconscious; see Orlando (1993) passim,
and especially Matte Blanco (1975).
46 Orlando (1979).
Orlando (1971).
Staging Thyestes
43
and impressive heuristic potential. Yet it is clear that certain literary texts
seem to embody more strongly than others the idea that they represent
a return of the repressed; furthermore, the rhetorical complexity of these
texts can be readily considered to be one of the ways in which repressed
contents are camouflaged. In the case of Senecan tragedy both assumptions would prima facie withstand examination: in the case of Thyestes,
however, the analysis of the prologue that I have proposed highlights a persuasive thematization of the return of the repressed which warrants further
investigation.
In a very important sense Thyestes can be read as an experiment in the
nature and limits of tragic (poetic) language and an answer to the problem
of the relationship between poetry and reality. The antagonisms raging in
the play invest a number of different spheres of human nature and activity.
Language is one of these spheres, and the conflict between Tantalus and
the Fury is also a friction between the words of tragedy and the silence
preserved by their avoidance. The prologue thus represents poetry as the
medium through which scelera and nefas can be expressed, and against
which the moral restraint personifed by Tantalus shadow remains fatally
impotent. Not only are words actions, as the double aspect of scelus and
nefas itself powerfully suggests, but the words of poetry represent a decisive
victory against the repressive morality of silence. If we extrapolate the conflict between (repressive) silence and words already encoded in the word
ne-fas, we can better visualize the reversal brought about by voicing deeds
and words (fas) which had been deemed worthy of censure and perhaps
oblivion. It is not by chance that evil Senecan characters such as Atreus
(age, anime, fac quod nulla posteritas probet, | sed nulla taceat: up, my soul,
do what no coming age shall approve, but none forget!, 1923) and Medea
(paria narrentur tua | repudia thalamis: let the story they tell of your divorce
be like the one they tell of your marriage, 523) are obsessed with the hope
that their actions will not be passed over in silence. Their rebellion, which
subverts normative codes of conduct, demands a similar overturning of the
decorum of silence and is coextensive with it. Atreus will do unspeakable
things, and that is precisely why he wants to ensure that they will be spoken
and talked about for ever.
In giving voice to nefas, poetry reverses the repressive instance that ne-fas
would encode. Silence was, after all, the standard, expected reaction to nefas.
At the political level, for instance, damnatio memoriae awaited the enemies
of the state whose very names, let alone actions, were consigned to eternal
oblivion by a stroke of the pen. For a long time, literature seemed bent
on extolling virtues rather than on giving any space to offensive conduct.
44
Seneca himself refers explicitly to the difference in treatment (Consolatio

ad Marciam 1.4):
legitur [sc. Cremutius Cordus], floret, in manus hominum, in pectora receptus
vetustatem nullam timet; at illorum carnificum cito scelera quoque, quibus solis
memoriam meruerunt, tacebuntur.
But he is now read, he lives, and ensconced in the hands and hearts of men he
fears no passing of the years, but these cut throats even their crimes, by which
alone they deserved to be remembered will soon be heard of no more.
It will fall to Tacitus to subvert this basic rule and to claim for his
historical prose, which does not shrink from mentioning negative behaviour
if necessary, the hitherto little-explored function of a deterrent (Ann. 6.7.5):
neque sum ignarus a plerisque scriptoribus omissa multorum pericula et poenas,
dum copia fatiscunt aut, quae ipsis nimia et maesta fuerant, ne pari taedio lecturos
adficerent verentur: nobis pleraque digna cognitu obvenere, quamquam ab aliis
incelebrata.47
And I am not unaware that many writers omit to discuss the dangers and punishments of many men, either because they flag at the quantity, or because they are
afraid to afflict their readers with experiences which they have themselves found
excessive and sad: as far as I am concerned, many things came to my notice which
I consider worthy of record, even if they have been left unrecorded by others.
At a more radical level, the presentation of Atreus actions as an intrinsic

victory over repression, together with the collusion established between
poetic word and scelus, gives scelus an emotional appeal which defies the
feeble attempts at moral correctness that are ostensibly advocated in other
parts of the play. Also, as we have seen, any attempt made by the audience
to identify their emotions with any given character is distorted by the
plays intricate metadramatic structure; for any action represented in the
play must not only be interpreted and judged per se: the very form of its
representation also carries upsetting ethical connotations.
The most important case in point is Atreus himself. The text constantly
challenges its audience to assess the ethical status of Atreus and his actions,
but also questions the ethics of the authors choice to represent them. Thus
the possibilities of emotional identification offered to the audience are multiplied and result in a potentially endless set of conflicts. The audience must
constantly assess the plausibility of Atreus complaints and the explanation
47
The issue is discussed by Luce (1991) 291213, to which I am indebted. He identifies in Diodorus
1.1.5 and 15.1.1 the only other instance of history acting as a deterrent in ancient historiography.
I will explore below (pp. 228ff.) whether Senecas tragedies, too, might be considered a deterrent
against morally repulsive behaviour.
Staging Thyestes
45
he offers for his actions, and at the same time deliberate whether they can
enjoy the aesthetic emotions offered by Senecas poetry without colluding
ipso facto in its powerful violation of fas.48 Conversely, the audience might
sometimes sympathize with Tantalus advocacy of silence and simply wish
that the tragedy did not exist at all.
The function of this polymorphic prologue, as I have tried to show, is
radically to complicate a straightforward opposition between the oppressing
force of Atreus violence and the moral values defended, however faintly, by
Thyestes, the satelles and the chorus. The prologue introduces a dynamic
which subjects otherwise clear-cut values to multiple transformations and
interactions. The prologue functions as a vital metadramatic frame for the
drama as a whole, ensuring that dramatic and metadramatic dimensions
are always co-evident. It also casts in a different light a dilemma which I
have already touched on,49 namely whether we should read the tragedies as
negative illustrations of values and ideas advocated in Senecas prose works.
The very architecture of the play promotes a multiplication and diffraction
of meanings that makes summary comparison with other texts necessarily
absurd. What the prologue impresses upon us is first and foremost the
lacerating power of the poetic word, which imposes on its creator and its
public a set of moral implications that cannot necessarily be bound into a
reassuring unity.
framing thyestes
libet videre
(Seneca, Thyestes 903)
i
The prologue sets up a pattern of representation which is essential to the
structural organization of the tragedy as a whole. The Fury and Tantalus
cease to act, but they are not meant to disappear. The tragic action will
unfold before their eyes, and Tantalus, too, will be forced to watch the
monstrous banquet he has unavoidably, if unwillingly, precipitated: let
blood mixed with wine be drunk before your eyes (656: mixtus in Bacchum
cruor | spectante te potetur). The conflict between the Fury and Tantalus
draws a line between active and passive forces, performers and spectators,
48
49
This is the most explosive aspect of the return of the formal repression which Orlando posits as
one of the forms in which the repressed returns: certain densely figurative parts of the work offer a
direct aesthetic pleasure.
See above, pp. 2ff.
46
power and powerlessness; it is rewarding to read it also as a meditation

on the implications of poetry and its effects on both author and audience.
Tantalus, once he has performed his primary task, becomes a spectator
himself, an impotent, horrified prisoner. As he lurks in the background
throughout, his perspective remains an imposing, if latent, counterpart to
our own troubled experience as spectators of Thyestes.
The conflicting drives personified by the Fury and Tantalus in the prologue infect the play as a whole. This contrast generates in the text a series
of oppositions that are subordinate to and dependent on the basic polarization of silence and tragedy instigated by the Fury and equally liable to
be construed as a conflict between repression and its removal.
Just as the contrast between Tantalus and the Fury pits one of the principal moving forces of the tragedy against the potentially most effective
obstacle to it, further dramatic confrontations in the play mirror the same
antithesis. The satelles initial reaction to Atreus plans in the second act is
precisely one of resistance. Functionally, then, the satelles and Tantalus
are paired together in their vain attempt to stop Atreus and the Fury
respectively. Atreus complains about his inactivity (1768: ignave, iners,
enervis . . . | . . . | inulte, undaring, indolent, nerveless, unavenged), and
braces himself for new, spectacular actions, indeed for a new scelus (203). The
satelles, on the contrary, advocates the restrained morality which Tantalus
had unsuccessfully embraced, and tries to counter Atreus machinations
with an invitation to desist from the proposed scelus. However, as we will
see shortly, he ultimately changes his initial attitude, accepts Atreus point
of view and tries to expose the weak points of his plot rather than insist
on the need for restraint. As in the prologue, if Atreus yielded to the
satelles invitation, there simply would be no tragedy at all. Once again,
the intimate connection between the unfolding of the plot and Atreus
responsibility is foregrounded. The equivalence between nefas as an action
and a poetic representation of that action is made emphatic in Atreus
self-presentation at 176204. Atreus plans to perpetrate a nefas (193) of
unrivalled atrocity, one which can therefore aspire to immortal fame, to
never being silenced (1923).50 The remarks that follow, at 2504, after a
brief altercation with the satelles show that Atreus is again a victim of the
same furor of poetic creation as is symbolically represented by the prologue, when Tantalus is forced to submit to the Furys irresistible force of
inspiration.
50
See Aaron in Titus Andronicus 3.1.1335: What shall we do? Let us that have our tongues | Plot some
device of further misery | To make us wondered at in time to come.
Staging Thyestes
47
The same pattern is repeated in the dialogue between Tantalus and his
father, Thyestes, at 40490. Here, too, the contrast between father and son
focuses on Thyestes reluctance to further the dramatic plot (436: placet ire,
pigris membra sed genibus labant, I would like to go, but my limbs waver
on my shaky knees) and Tantalus insistence that he follow the prescribed
series of events (440: evince quidquid obstat et mentem impedit, overcome
whatever opposes and thwarts your will). Thyestes final words (4889:
eatur. unum genitor hoc testor tamen: | ego vos sequor, non duco, let us on.
Yet this one thing your father does declare: I follow you, not lead), in fact,
pointedly echo Tantalus shadows ultimate confession of defeat: sequor,
I follow (100).51 Finally, in the first meeting between the two brothers,
Thyestes incarnates a role which by now we have learnt to recognize as
ineffective and morally dubious. Thyestes does indeed try to resist Atreus
enticements, much as Tantalus had done in the prologue, and as the satelles
had perfunctorily attempted to do in the first act, but his sudden and
rhetorically startling surrender to Atreus argument is not comparable either
to Tantalus suffering while he is tortured or to the undeniably weak position
of the satelles: Thyestes, whose superficial determination had begun to
vacillate while he was talking to his son, acquires in this central scene of
the play the new role of a moving force. In this way he signals his final
transformation from victim to accomplice.
The whole tragedy thus hinges on the antithesis of two sets of functionally similar characters: on the one hand, the Fury, Atreus and Tantalus, on
the other, Tantalus shadow, the satelles and Thyestes. The two groups possess different degrees of textual knowledge and stand in different positions
vis-à-vis the metadramatic aspect of the play. The first group includes all the
forces that work towards the tragedys resolution, those whose furor in carrying out the proposed nefas is coextensive with the removal of the repression
that lifts the play from silence into existence. The second group is made up
of characters who in one way or another want to uphold that repression,
those who try, with different degrees of determination and credibility, to
stop the ruinous pattern of events instigated by their antagonists. This contrast is reinforced, as we will see, by the fact that the two groups represent
authors and victims of deception.52 The moving forces of the tragedy, furor,
nefas and furor-inspired poetry, are embodied by consummate deceivers,
against whom Tantalus, Thyestes and the satelles moralizing attempts are
51
52
Note also that Tantalus exhortation in the two following lines (48990: respiciet deus | bene cogitata.
perge non dubio gradu) echoes a similar command on the Furys part: perge, detestabilis | umbra
(245).
On Tantalus see below, pp. 489.
48
completely ineffectual, partly because they are predicated on an incomplete,

flawed knowledge and assessment of the events.
The systematic correspondences among the characters of the play map
the opposition between repression and its removal that was emphasized in
the prologue and reiterate and expand the basic conflict between poetic
expression and morally justified silence. A further set of oppositions is also
implied: between nefas and fas, ratio and furor, honesty and deception. The
conflict between Atreus and Tantalus on the one hand, and the satelles and
Thyestes on the other, reflects the contrast between the Fury and Tantalus
shadow in the prologue, since it dramatizes the fundamental dichotomy
between furthering and suppressing the dramatic plot. The prologue sets up
a conflict which pervades the whole play, leading the audience to suspend
judgement on the utterances and actions of the characters, not only in
the prologue, but throughout the play. At the same time, we must also
be prepared to recognize that Atreus, the playwright with bloody hands,
stands out in the play as the incarnation of a victory against the constraints
of moral repression, a triumph which is inevitably connected with the force
and pleasure of poetry.
Before turning in the following chapters to analyse the main characters
of the play and their epistemological horizons, I should make a closer examination of the structural relationship between the various scenes and
the different levels of the tragedy that I have outlined. The play unfolds at
different levels. The first is represented by the prologue and the indication
of the apparently limitless agency of the Fury, whose presence will have to
be felt, though not seen, in the rest of the play. Indeed, if spectante te at
line 66 is to be taken literally, Tantalus is throughout the play an unseen,
incapacitated spectator of the events that he has been forced to precipitate.
As a hidden spectator of the events that will unfold, Tantalus would parallel
Hermes in Euripides Ion, a prologue character (! ) who hides
away (probably behind a panel shaped as a bush) as soon as he notices
Ion arriving on the scene at the end of the prologue (7880), and never
reappears.53 Although hiding-scenes at the end of a prologue are to be found
in several authors,54 Senecas te spectante retains the defining features of Ion
53
54
See Halleran (1985) 102. In Senecas probably Euripidean model Tantalus remained to see the banquet
(Lesky (192223) 533); according to Steidle (194344) and Hine (1981), however, Senecas Tantalus
does leave once and for all when he has polluted the house. This issue is closely connected with
the interpretation of the Furys seemingly contrasting orders at 66 and 1056 (gradere ad infernos
specus | amnemque notum): does the Fury change her mind in the course of the prologue and relent
on her initial order that Tantalus watch the ensuing actions (so Tarrant (1985) 98), or do the two
contrasting orders follow each other in a compressed temporal sequence? I discuss the issue in
ch. 5, pp. 17880.
Aeschylus (Taplin (1977) 3345), other tragedians (Soph. El. 7785; Eur. Hipp. 513; Hec. 524), Old
and New Comedy (after Leo (1908) 68, see, e.g., Fraenkel (1962) 226 and Handley (1965) 1712).
Staging Thyestes
49
(a play, incidentally, deeply concerned with the issues of viewing, spectatorship and representation), where the ! is a divine character
who remains on the scene to watch the events but does not reappear later
in the play.55
At a further, included level, we can place Atreus deliberations on how
to fulfil his wish for revenge, his own author-like plotting. At the third
level, finally, we watch Atreus turn into an actor of the play he has devised,
and Thyestes being taken in by the elaborate performance of his brother.
It might be useful to sum up these distinctions as follows:
First level
Second level
Third level
(a) Prologue
(b) Play
(c) Atreus play
1 Fury, Tantalus (1121)

[T
2 A
I chorus (12275)
Atreus, satelles (176335)
II chorus (336403)
N
3
Thyestes, Tantalus (40490)

T
A
4 L
5 U
Atreus alone (491507)

Thyestes, Atreus (50845)
III chorus (546622)
chorus, mess. (623788)
IV chorus (789884)
Atreus alone (885919)
Atreus, Thyestes
(9201004, agnoscis)
S]
Atreus, Thyestes (10051112)
ii
An analysis of the second level of dramatic action in Thyestes reveals a close
structural similarity to the prologue, since the dialogue between Atreus
and his satelles parallels that between the Fury and Tantalus shadow.56
The specific metadramatic implications of this part of the play are particularly evident in the characterization of Atreus, whose plotting is insistently
associated, as we have seen, with the creative activity of a poet.
This second part of the play opens with a monologue by Atreus, which
parallels Tantalus opening speech. Seneca favours the use of an expository monologue at the beginning of several plays (Hercules furens, Troades,
Medea), or, as in this case and in Oedipus, an opening monologue which
55
56
Eur. Hec. 524 offers another interesting point of comparison: the ghost of Polydorus disappears as
his mother, Hecuba, arrives on the scene, having dreamt, he claims, of his own presence.
I will discuss later (ch. 4) the possible political implications of the dialogue between Atreus and the
satelles.
50
then turns into a dialogue for the rest of the first act:57 Atreus first appearance is thus overtly characterized as a programmatic gesture. There is
no need to suppose that the Fury and the satelles should arrive on stage
with their respective counterparts, although their initial words pick up at
half-line:58 in both cases it is a solitary speech, heavily punctuated by questions, which has pride of place. Just as the Fury did, Atreus vows to take
his revenge on Thyestes by producing a nefas of great novelty quid novi
rabidus struis?, asks the worried satelles (254: what strange design are you
plotting in your rage?) and unsurpassed audacity: fiat hoc, fiat nefas |
quod, di, timetis (2656: let it be done, let a nefas be done at which, o gods,
you take fright). His nefas, as we will see, is an eloquent poetic artefact, a
deceitful ploy whose author (quid sit quod horres ede et auctorem indica, says
the chorus at line 639: tell what it is that makes you shudder, and point
out its author) fully exploits the potential of words to ensnare and betray.
Atreus nefas is the core action of the whole tragedy, the well-devised and
well-acted scheme to which Thyestes is doomed to succumb.
While the pairing of nefas and poetry was already clearly established in
the prologue, Atreus declaration of poetics adds significant details to our
understanding of exactly what kind of poetic activity the text is referring to.
The scene opens, as the prologue did, by pitting Atreus against the satelles
in a debate on what should be the presiding force of creation (24854):59
sa.
nulla te pietas movet?
at. excede, Pietas, si modo in nostra domo
umquam fuisti. dira Furiarum cohors
discorsque Erinys veniat et geminas faces
Megaera quatiens: non satis magno meum
ardet furore pectus, impleri iuvat
maiore monstro.
s a. Does no sense of Piety move you?
at. Away, Piety, if ever you have had a place in our house. Let the fearful band
of Furies come, the discord-sowing Erinys, and Megaera, shaking her twin
torches; the frenzy burning in my breast is not great enough; some greater
horror must fill me.
57
58
59
The beginnings of Phoenissae (a dialogue throughout) and of Phaedra (a monody followed by

two scenes) privilege other options. As Tarrant ((1976) 157) notes, Senecas predilection for the
introductory monologue is Euripidean . In addition to Tarrant see Anliker (1960) 239.
At lines 23 and 203 respectively.
As Tarrant ((1985) 1289) rightly points out, many features of Atreus behaviour closely resemble
Senecas description of an iratus (cf. especially the angry mans inclination towards the sublime, Thy.
268 with De ira 1.10.2). On Atreus as an iratus see the important article by Staley (1981). While my
emphasis here is different, his perspective should certainly be kept in mind.
Staging Thyestes
51
As the Fury invoked the precedent of Procne and Philomela, here Atreus,
echoing in maiore monstro (254) the Furys maiore numero (57), names her as
his Muse. The tortures inflicted by the Fury on Tantalus shadow become
the self-conscious ardour of the new poet, the metaphoric fire of inspiration
or the unsettling force of poetic enthusiasm (969):
quid ora terres verbere et tortos ferox
minaris angues? quid famem infixam intimis
agitas medullis? flagrat incensum siti
cor et perustis flamma visceribus micat.
Why do you terrify me with the sight of your lash, and fiercely threaten me with
your twisted snakes? Why do you rouse pains of hunger deep in my innards? My
heart burns with fiery thirst, and in my burnt-out vitals a flame is darting.
In two later, closely related passages Atreus insists he is being dominated

by the sweeping force of inspiration. First he confesses his emotional distress
(2602):
at. fateor. tumultus pectora attonitus60 quatit
penitusque volvit; rapior et quo nescio,
sed rapior61
at. I do confess it. A mindless tumult shakes and churns my breast deep inside.
I am dragged away, I do not know where to, but I am.
His conclusion clearly recalls the Furys intimation at line 56 let the Thracian crime be done, but multiplied (Thracium fiat nefas | maiore numero):
fiat hoc, fiat nefas | quod, di, timetis (2656: let a nefas be done at which,
o gods, you take fright). Rapior provides a description of Atreus state of
mind that is paralleled in a densely programmatic Horatian ode, 3.25, where
the poet explicitly connects the force of Bacchus inspiration with his transportation into uncharted, perilous territories.62 As we will see shortly, it is
precisely this type of Bacchic, enthusiastic poetics that Atreus embodies on
60
61
62
Its Greek equivalent "

" #
also has distinct Dionysiac overtones; see Mendelsohn (1992)
111. Attonitus is first found in poetry in Virgil, but an Ennian model is possible. For Bacchic
overtones see Aen. 7.580 attonitae Baccho . . . matres and Hor. Carm. 3.19.14 attonitus . . . vates (with
Livys description of revellers at Bacchanalia at 39.15.9 vino . . . attoniti). On Senecas abundant and
nuanced use of the adjective in the tragedies see Pasiani (1967).
The structure of lines 2612 recalls the famous Catullan distich (85) about the epistemological
quandary caused by erotic passion: odi et amo. quare id faciam, fortasse requiris. | nescio, sed fieri sentio
et excrucior.
Quo me, Bacche, rapis tui | plenum? quae nemora aut quos agor in specus | velox mente nova? (lines
13). Cf. lines 1718: nil parvum aut humili modo, | nil mortale loquar. On this passage see Pasquali
(1920) 14 and 54950 (with bibl. 81314); and La Penna (1971). See also Sat. 2.1.10: si tantus amor
scribendi te rapit, and Ovids Medea (fr. 2 Ribbeck2 ): feror huc illuc, ut plena deo. I will return later
to the Dionysiac aspects of Atreus character, pp. 1338.
52
the stage of Thyestes, and that we, the spectators, should consider to be the
driving force of the play.
Subsequently, as the satelles is reduced to a completely impotent sparring
partner, who does little more than feed his master the next line, Atreus
caps his own poetic prologue by describing in further detail the grandiose
nefas he is plotting,63 and by invoking his own special Muses, Procne and
Philomela (26777):
at. nescioquid animus maius et solito amplius
supraque fines moris humani tumet
instatque pigris manibus haud quid sit scio,
sed grande quiddam est. ita sit. hoc, anime, occupa
(dignum est Thyeste facinus et dignum Atreo,
quod uterque faciat): vidit infandas domus
Odrysia mensas fateor, immane est scelus,
sed occupatum: maius hoc aliquid dolor
inveniat. animum Daulis inspira parens
sororque; causa est similis: assiste et manum
impelle nostram.
at. My mind swells with something greater, larger than normal, and beyond the
bounds of human custom, and it urges on my sluggish hands I do not know
what it is, but it is some great thing. So let it be. Haste, my soul, take hold
of it. (It is a crime worthy of Thyestes, and worthy of Atreus, such that each
of them could perform it.) The Odrysian house once saw an unspeakable
banquet this is a monstrous crime, I grant, but it has been done before;
let my grievance find something worse than this. Inspire my soul, o Daulian
mother and sister; my case is like yours; help and urge on my hand.
The nature of the poetic creation which Atreus envisages is evidently

very close to the one imposed by the Fury on Tantalus. His deeds will have
to be sublime, literally exceeding human boundaries supraque fines moris
humani (268). They will be original, as the satelles unwillingly points out by
asking quid novi rabidus struis? (254).64 And, above all, they will have to outdare all precedents, just as the Fury had demanded: Thracium fiat nefas |
maiore numero (567). These features in particular the programmatic
combination of novelty and awareness of the tradition65 would be enough
to lend Atreus declaration of intents a distinct literary colour, even if he had
not used a number of key terms which Seneca elsewhere applies explicitly
63
64
65
I will analyse the character of the satelles below, pp. 15464.

On Atreus sublimity see the next chapter, p. 127.
Compare especially Letters to Lucilius 79.6, in which Seneca discusses the relationship between
invention and tradition with regard to a poem on Aetna which Lucilius might be tempted to
compose.
Staging Thyestes
53
to poetic creation.66 The most important of such parallels is undoubtedly

with De tranquillitate animi 17, where, as we have seen in chapter one,67
Seneca offers the most overt endorsement of the Democritean and Platonic
theory of poetic enthusiasm, and praises the emotional excitement that
leads to sublime poetry.68 It is worth quoting again the last few lines of the
dialogue (17.1011), important elements of which are closely paralleled by
Atreus:
For whether we believe with the Greek poet that sometimes it is a pleasure also to
rave, or with Plato that the sane mind knocks in vain at the door of poetry, or with
Aristotle that no great genius has ever existed without some touch of madness
be that as it may, the lofty utterance that rises above the attempts of others is
impossible unless the mind is excited (mota mens). When it has scorned the vulgar
and the commonplace, and has soared far aloft fired by divine inspiration, then
alone it chants a strain too lofty (grandius) for mortal lips. So long as it is left to
itself, it is impossible for it to reach any sublime (sublime) and difficult height; it
must forsake the common track and be driven to frenzy (efferatur) and champ the
bit and run away (rapiat) with its rider and rush to a height that it would have
feared to climb by itself.
Atreus emphasizes in a similar fashion the connection between the irrational force of poetic inspiration one that literally snatches away
and a grandiose, sublime form of poetry which deserves the highest praise.
Indeed, poetic enthousiasmos can transform man into a semi-divine entity, a
vates whose utterances have a higher claim to truth.69 It is especially worth
noting that the last sentence of the passage appears radically to subvert the
hierarchical principle on which Stoic ethics is predicated, by relegating the
rector to a subordinate position. Poetry, as an irrational force, smashes
the barriers of both rationality and decorum, and overrides any hesitation enforced by timor (Tranq. 1.14): then again, when my mind has been
uplifted by the greatness of its thoughts, it becomes ambitious of words,
and with higher aspirations it desires higher expression, and language issues forth to match the dignity of the theme; forgetful then of my rule
and of my more restrained judgement, I am swept to loftier heights by an
utterance that is no longer my own (rursus ubi se animum cogitationum
magnitudine levavit, ambitiosus in verba est altiusque ut spirare ita eloqui
66
67
69
This comparison is developed in full by Picone (1984) 5661. Picones book offers a fascinating
reading of Thyestes, though he places a different emphasis on Atreus as doppio negativo del vates
(59).
68 Mazzoli (1970) 509.
Above, pp. 223.
I agree with Mazzoli ((1970) 52, n. 94) that this passage is connected with the description of
the Sybil at Virg. Aen. 6.4851. See Picone (1984) 59, n. 39, and further references to De ira 2.35.6;
De vita beata 2.2; 9.2; Ep. 108.26.
54
gestit et ad dignitatem rerum exit oratio; oblitus tum legis pressiorisque iudicii
sublimius feror et ore iam non meo).
On the basis of these connections, it is now clear that the metadramatic
implications of the Furys actions, perceptibly if indirectly suggested in the
prologue, constitute an essential dynamic of Thyestes as a whole: act 2 confirms more explicitly what the audience already had cause to suspect at the
very beginning of the play. As for Atreus, we are now effectively invited to
consider him a magnum ingenium as well as a sublime poet, one who attains greatness by overstepping the boundaries of timor and mores through
a superior form of enthousiasmos. It would still be possible to suppose that
Atreus poetic activity is represented here as an anti-model of the true
poet, a negative mirror for the moral conscience of the Stoic author.70 Such
an interpretation, however, is rooted in the assumption that Senecas prose
works try to justify the un-Stoic aspects of the Democritean-Platonic theory of poetic inspiration by subordinating them to a morally praiseworthy
goal, and by trying to restrain the potentially disruptive effects of the sublime. I have already tried to show71 that even if we confine analysis to the
theoretical prose works, it is very difficult to arrive at such an unambiguous assessment of the moral implications of poetry. I would suggest that
metadramatic passages such as Oedipus 50965872 emphasize the feeling
of ambivalence and awe surrounding the process of poetic creation, and
indeed set inspiration and creation in a twilight zone of horrific, haunted
groves. In the case of Atreus, however, any comforting solution seems to be
denied by the context itself. Nothing in Thyestes suggests the notion that
Atreus should be imagined as a negative model of the poet. The principles
he advocates originality, knowledge of the tradition, desire to reach the
highest peaks of creation are nowhere accompanied by a critique of the
notion of good or successful poetry. Perhaps we might argue that the text
frames Atreus poetry as bad by exposing its association with unethical
deeds, and that we should therefore be able to distinguish between his
abstract principles of poetics and his wholly unprincipled conduct. But,
even if we set aside for the moment a more nuanced evaluation of Atreus
motives and behaviour,73 we are still bound to recognize that his poetic
success is intrinsically twinned with his thirst for revenge: he devises his
seductive and successful plot precisely as a means to exact retribution from
his brother.
70
72
71 See ch. 1, passim.

Picones thesis (1984) 59.
73 Ch. 3, passim.
Ch. 1, pp. 810.
Staging Thyestes
55
However, if, as I have suggested, act 2 replicates in a more obvious way a

set of ideas already clearly established in the prologue, we should feel fully
justified in reading the connection between Atreus and the sublime poet
described in De tranquillitate animi as thoroughly consistent, in so far as it
upholds the equation between poetry and the lifting of repression, and is
predicated on the inherently ambiguous notion of poetry as a transgression
of psychological, moral and expressive limits. All sublime, grandiose poetry
is nefas, is inevitably implicated in transgressive actions, since it abandons
self-composure in a heady atmosphere of semi-prophetic creation. Atreus
is but one example of sublime poetry, one which concretizes the emotive
alliance between poetry and nefas. As it repeats itself for the second time,
the contrast between poetry and repression so forcefully voiced by Tantalus
and the Fury seems even more skewed in favour of the latters violent,
trangressive force. In the dialogue between Atreus and the satelles, even
Tantalus credible, if short-lived, attempt at resistance has disappeared. So
has the sense of impotent despair with which he surrenders himself to the
overwhelming force of the Fury. The satelles is an ineffective dialectical
adversary, whose ethical considerations will be summarily dismissed by
Atreus swift, pragmatic attitude.74 The victory achieved by the Fury in the
prologue has been decisive, and its effects are felt throughout the drama
over which she grimly presides.
iii
Atreus reappears on stage as the crafty author and director of his own tragic
play in a brief aside in act 3: he is in front of the royal palace as Thyestes
and his children approach, and comments smugly on the favourable turn
that events have taken. The beast (his brother) has been captured. All
that is required now is to keep up the fiction, to start acting as planned.
The emphasis on deception is explicit: when rage scents blood, it cannot
be concealed; yet it must (5045: cum sperat ira sanguinem, nescit tegi |
tamen tegatur).75 The ensuing performance will be based, unsurprisingly, on
deception: Atreus last words before addressing his brother, praestetur fides
(507), are revealing, since they mean both let me be true to our promise
(ironically), and let me put on a display of fides. This scene marks the
transition between the second and third levels of the plot: it opens with
Tantalus and Thyestes approaching the royal palace unaware of what has
74
See below, pp. 15461.
75
On lines 497505 see below, pp. 99100.
56
been plotted within (40490), it gives space to Atreus aside (491507), and
it ends on the two brothers meeting in what is for Atreus a masterful display
of intrigue (50845). It is in this third and innermost level of the tragedy,
where the revenge finally takes place, that Atreus doubles up, not unlike
a cunning Plautine slave, as an actor in the play he has himself plotted.
But the performance has begun even before he appears on the stage: it falls
to Thyestes and his son ignorant actors to begin the dramatization of
the plot that Atreus has carefully orchestrated; they have been duped into
coming here by Atreus, who, as we soon realize, is watching the proceedings
very closely.
Thyestes arrival in the vicinity of Argos clearly represents a fresh beginning. We might compare it, for instance, to Orestes arrival at Argos at the beginning of Aeschylus Choephoroe, or Odysseus landing on
Lemnos in Sophocles Philoctetes. In the first few lines of Sophocles Oedipus
at Colonus the blind Oedipus wonders where he has finally got to, and
Dionysus arrives in Thebes as Euripides Bacchae begins.76 The most interesting parallel, however, can be found in the prologue of Sophocles
Electra. There Orestes reaches Mycenae accompanied by the old pedagogue, who, like Thyestes, first talks about the satisfaction of his old desire
to return (810): and look! before us, at our very feet you see Mycenae of the
golden hoard, and there the grim palace of Pelops line, deep stained with
murder.
The dialogue between Thyestes and Tantalus is structured along the
lines of the two dialogues that precede it, namely that between the Fury
and Tantalus in the prologue and the debate between Atreus and the satelles
in act 2, both directly or indirectly charged with inceptive functions. The
closing lines of this scene Thyestes tormented sequor (489) again mark
with explicit verbal echoes the connection with the prologue. As the vigilant Fury presides over the actions she has initiated, we must imagine that
Atreus is watching the first act of his new play. Frames and levels of spectatorship multiply, and, as we will see, inevitably complicate our own act of
viewing.77
Indeed, a great deal of prominence is assigned in this scene to the issue
of viewing and understanding. Thyestes approaches his city and the royal
palace as a confused spectator of the performance laid on for him by his
brother (407: cerno). Although he cannot fully understand the dangers
awaiting him, he nonetheless hesitates to be drawn onto the stage that
76
77
The verb $ is often used to mark the arrival of supernatural entities on the stage at the very
beginning of a play; see, for example, Eur. Bacch. 1, Hec. 1, Tro. 1, Ion 5.
Ch. 6, passim.
Staging Thyestes
57
Atreus has prepared. He is aware of the distance between appearance and

reality, of the seductive illusion of spectating, though ultimately he is unable
to draw the right conclusions from these premonitions (41416):
clarus hic regni nitor
fulgore non est quod oculos falso auferat:
cum quod datur spectabis, et dantem aspice.
there is no reason why the shining splendour of power should mislead my eyes
with its false radiance; when you look at a gift, check who is giving it, too.
The spectators point of view is aligned with the privileged viewing position of Atreus (and also of the Fury and Tantalus ghost). We know full
well that Thyestes intimations of impending doom are justified, and his
words, like much in the ensuing dialogue with his son, are pregnant with
tragic irony. As he first sees Thyestes at line 491, Atreus is certain that he
has succeeded in his ploy: his viewing is endowed with full awareness and
understanding. The repetition of cerno in the third line of his speech (493)
echoes the presence of the same verbal form in the fourth line of Thyestes
own opening remark (407), and in so doing calls attention to the very different levels of awareness and understanding that accompany Thyestes and
Atreus viewing. This time, however, there is irony in the tyrants words,
and the exclamation with which he greets his brother at line 508 fratrem
iuvat videre (it is sweet to see my brother) underlines the fact that Atreus
viewing is geared towards a full emotional satisfaction, rooted in the complete control of the situation. It is the pleasure of poets, but is also the
pleasure of spectators who have been privileged with an inside knowledge
of the plays creative mechanisms. Atreus manipulation of the emotional
connotations of viewing is clear at line 525, when his exhortation
oculis . . . nostris parce (spare my eyes) craftily exploits the duplicity of
his perspective. Far from being disturbed at the sight of Thyestes, he enjoys both his current pitiable state, and the anticipation of the fact that if
Thyestes renounces his present misery and accepts to share the throne, he
will fall prey to an even worse fate.
Several other details in the scene emphasize how different the characters
level of cognition can be, and therefore call our attention to the different
fictional levels operating in the scene. For instance, Thyestes praise of quies
at line 469, which he himself is ultimately unwilling to transform into actions, is expressed in words that shortly afterwards will be echoed by Atreus
with completely different implications: as praestetur fides (507) echoes
praestatur fides (469) in the same metrical position, we are almost invited
to suspect that Atreus has listened to his brother and is now exploiting his
58
own words to express very different contents. A similar ironic twist can be
detected in Tantalus final words at 48990: a god will regard with favour
what has been well devised. Haste on with assured step (respiciet deus |
bene cogitata. perge non dubio gradu). Not only is there no intimation in the
play of the gods moral agency,78 but in fact we are led to assume shortly
afterwards that Atreus who would certainly not object to being called a
god has been watching this scene all along.
iv
In the fourth act of the tragedy a messenger relates to the chorus and the
audience a series of actions which they cannot witness directly and, by
sharing with them events which Thyestes will ignore until a later stage
(105268), aligns them with Atreus true intentions and behaviour.79 This
double narration of the pivotal scene of the tragedy effectively thematizes
the reiteration of nefas. Each different level of narration no longer encases
a subsequent one, but instead the same action can be watched more than
once from different perspectives corresponding to different audiences. This
new element reinforces the density of the plays structure, a topic on which I
will shortly focus: the key action of the plot is never performed on the stage
(as tragic conventions prescribe), but its absence is reflected in multiple
mirrorings, in an attempt to fill the void with competing narratives, each
bearing different emotional connotations, and each involving a different
audience and hence different reactions.80 But here, too, it is tempting to
invoke that particular concept in order to account for the multiplication
of an object which still denies us direct access and can only be perceived
at a certain remove. It has often been said that Senecan tragedy suffers
from an excess of parataxis, an inability to privilege an organic plot vis-à-vis
the centripetal tendency to align somewhat detached scenes. Thyestes, with
its largely organic plot, would be the exception that proves the rule. Yet
Thyestes precisely suggests that this paratactical arrangement might be more
the product of a specific epistemic and aesthetic Weltanschauung than an
index of artistic weakness. The structure of the final act of Thyestes is
symmetrical to that of act 3, which interposes Atreus monologue between two scenes en abyme. In act 5, however, it is the staged dialogue
between the two brothers which occupies the intermediate position: the
78
80
79 I will discuss other aspects of this speech later, pp. 87ff.

Ch. 4, pp. 1524.
I will discuss Senecas fractured treatment of time in ch. 5.
Staging Thyestes
59
beginning and end of the act are reserved instead for what I termed the
second level.81
At the beginning of act 5 (885919) Atreus celebrates his triumph shortly
before revealing to his brother the gory details of the massacre. At line 889
a verbatim repetition from his prologic speech (279: bene est, abunde est,
it is well, more than enough)82 marks the close connection between the
two sections, both of which emphasize the exceptional grandiosity of the
nefas he has perpetrated (8858):
aequalis astris gradior et cunctos super
altum superbo vertice attingens polum.
nunc decora regni teneo, nunc solium patris.83
I move on a level with the stars, and, above all others, touch with proud head the
lofty heavens. Now I hold the glory of the realm, now my fathers throne.
These opening lines clearly recall the end of Horaces first ode: quodsi
me lyricis vatibus inseres, | sublimi feriam sidera vertice (Carm. 1.1.356: but
if you include me among lyric bards, I will hit the stars with my exalted
head). Later, as Atreus congratulates himself at the sight of his brothers
desperation, a more tenuous reference to the same poem surfaces again,
as nunc meas laudo manus, | nunc parta vera est palma. perdideram scelus, |
nisi sic doleres (10968: now I praise my handiwork; now is the true palm
won. I would have wasted my crime, if you werent suffering this much)
echoes the contrast between true and Olympic celebration which Horaces
ode develops at lines 36 after Virgil, Georgics 3.1020 (palma figures in
both passages).84 The allusion to such well-known programmatic passages
again highlights the fact that Atreus sees his actions as artistic achievements
comparable to those of famous poets even as he plans the final step of his
revenge revealing to Thyestes the full import of the nefas.
In this last section of the tragedy Atreus insists with renewed emphasis
on the visual dimension of his actions, and confirms his awareness of the
81
82
83
84
The second and third scenes are separated in the tragedy by two choruses (546622; 789884), a
long speech by the messenger (623788) and Atreus monologue at 885919, which all contribute to
alter the spectators point of view.
I discuss these important internal links again in ch. 5, p. 179.
These lines are a plausible model for Tamoras cries of joy at Titus Andronicus 2.1.14: Now climbeth
Tamora Olympus top, | Safe out of fortunes shot, and sits aloft, | Secure of thunders crack or
lightning flash, | Advanced above pale envys threatening reach, where they are mischievously shot
through mock-Stoic images of the serene security of the sage such as, for example, Thy. 3658.
Atreus Priamel should be read in conjunction with Thyestes own reference to the same Horatian
ode in his proemial speech at 40420: the palma (410) he enjoyed most was that of racing
paterno . . . curru (410), while Atreus, following Horaces own predilection, has chosen the rewards
of poetry. See below, p. 116.
60
centrality of spectatorship. The first indication of this awareness surfaces

when, having just embarked upon his plan, Atreus is able to visualize the
scene that will later unfold on stage: tota iam ante oculos meos | imago caedis
errat (2812: already before my eyes flits the whole picture of the slaughter). Now Atreus wants to make sure that all concerned have a clear picture
of the nefas: the gods should be kept back so that they can see what has
happened (8945: ut ultricem dapem | omnes viderent, so that they all may
see the avenging banquet), or at least Thyestes should: quod sat est, videat
pater (895: but it is enough if the father alone sees it). The next scene, in
which Atreus is still acting the self-assigned role of the loving brother until
the anagnorisis shatters the fiction of this second level of the representation, will afford him the pleasures of spectatorship, as Thyestes slowly and
painfully discovers the truth. Atreus will watch Thyestes watching, will be
the spectator of a pained spectator: it is a pleasure to note, when he sees
his childrens heads, how his complexion changes, what words his first grief
pours forth (9035: libet videre, capita natorum intuens, | quos det colores,
verba quae primus dolor | effundat).85 The pleasure will reside primarily
in the slow unfolding of Thyestes reactions in front of his brother: I do
not want to see him miserable, but his becoming so (907: miserum videre
nolo, sed dum fit miser). The daedalean structure of the play once again
multiplies the levels of spectatorship and thematizes it.86 As we have seen
in chapter one, Stoic theory devotes a considerable amount of attention
to the plight of spectators, torn between a critical analysis of the actions
they are watching, and the ever-present risk of being too deeply affected
by them.87 As Atreus steps back and watches the performance that he has
staged, he impersonates just such an affected, undetached spectator, who
derives direct satisfaction from witnessing the spectacle: libet videre, capita
natorum intuens . . . (903). In a further indication of structural affinity,
Atreus prescribes for Thyestes what the Fury had initially intimated to
Tantalus ghost: mixtum suorum sanguinem genitor bibat (917: his sons
mingled blood let the father drink) echoes mixtus in Bacchum cruor | spectante te potetur (656: let blood mixed with wine be drunk before your
eyes).
Before the anagnorisis marks the tragedys sudden return to its second
level of action, Atreus can enjoy a spectacle similar to, but hugely more
satisfactory than the one he had been treated to at Thyestes arrival. We
observe Thyestes emotional monologue with him and, to a certain extent,
85
86
A particularly perverse desire. See Virg. Aen. 10.443 with Harrison (1991) 186.
87 Ch. 6, pp. 23543.
Ch. 1, pp. 1415.
Staging Thyestes
61
through him. Thyestes is relapsing into the doubtful, sceptical mood that
had coloured his dialogue with Tantalus, but once again he seems unable
(or unwilling) to excavate his doubts: for Atreus, and for us, his words are
hopelessly entrapped in a tragic irony which is only intensified by Atreus
double entendres in the ensuing dialogue.
The interplay of different levels of tragic action foregrounds the power of
Atreus deceit. We see Atreus plotting his revenge and carrying it out while
completely fooling his brother with a charade of friendliness. I will discuss
at the end of this book whether we are entitled to read in this multiplication
of levels a cautionary tale that by watching the poets cunning behaviour
we can learn to mistrust his ploys or whether his deceitfulness inevitably
forestalls any such resistance on our part. Anagnorisis plays a pivotal role in
the shift between these different levels of the tragedy.88 Anagnorisis reveals
again89 a state of things which should have never been forgotten: Thyestes is
tragically forced to recognize Atreus true nature only because he has failed
to heed his own confused but true premonitions, hesitantly voiced when he
arrived on stage. The deceiving power of poetry, it appears, is such that it can
make people forget even what they already sense and ought to understand.
But in the present context anagnorisis also points up the plays dramatic
structure and exposes to Thyestes his irredeemable epistemic inferiority.
While the audience sides with Atreus superior form of understanding, it
is also aware of a further vantage point which is denied even to Atreus,
since they have been watching the events of the play alongside two closer
spectators the Fury and Tantalus ghost.
t r ag e d y, t e r m i n a b l e a n d i n t e r m i n a b l e
humanae vitae mimus . . . (Seneca, Letters to Lucilius 80)
i
The tortuous relationships between different levels of dramatic action in
Thyestes heighten the audiences perception of the play as an artificial literary product, and its metadramatic structure prominently foregrounds the
process of artistic creation. The audience, rather than being confronted
with a finished, self-contained entity, faces a work in progress, as the plays
ingredients are slowly and meticulously cooked up, live. By shattering the
illusion of fictionality, metadrama carves up a schism between the readers
88
Ch. 6, p. 248.
89
See a more extended discussion of this topic in ch. 5.
62
involvement in the artistic creation and their awareness of its fictional status.
This paradox accents the readers active rather than passive involvement
in the process of decoding and understanding.90 By turning the fictional,
theatrical nature of the play into such an upfront thematic concern, Thyestes
occupies a distinctive place in Senecas tragic corpus and bears comparison with the other mastertext which develops similar reflections, Euripides
Bacchae.91
I have argued that metadrama inevitably introduces a critical dimension,
a distance between the events represented on the stage and the audiences
perception of them. By complicating the structure of the play and directing
the audiences attention towards its inner mechanisms, metadrama lures
the audience to reflect on what exactly they are watching and how it is
constructed. They are watching a tragedy, a performance provoked as in
this particular instance by revenge, by the regressive desire for retribution
which spans generations and involves human as well as superhuman agents.
It is a revenge plot based on deception and lies, and the audience is granted
a privileged epistemic viewpoint by being shown, from the very beginning,
how that dissemblance operates.
It would be too optimistic, however, to conclude that this critical gap is
what ensures the philosophical and moral viability of plays which revolve
to a disturbing extent around anger, violence, nefas and horror. On the
contrary, the frames encompassing different levels of action in a play such
as Thyestes may well be seen to complicate further the effects of the performance precisely because they raise the audiences threshold of awareness.
As I have already pointed out, framing ensures the multiplication of points
of view and the production of competing accounts of and reactions to the
events, thus reducing the possibility of a unified response on the audiences
part.
But framing, and especially repetitive, multiple framing, inevitably produces a more general disturbing effect. The compact boundaries of the
dramatic action are highlighted, and yet eo ipso effectively annihilated in
the apparently unstoppable fugue of frames.92 The violent actions encased
in Atreus inner play and, at a remove, in the prologue are portrayed as the
direct result of a violent sequence which, evidently, the frames are hardly
able to control. As the next frame portrays Atreus struggling to devise an
original plan, apparently propelled only by his own thirst for revenge, the
audiences awareness of his motivations shatters his image of uncaused
90
92
91 See pp. 1338.

Hutcheon (1984) 7, referring to the modern novel.
Or, with Derrida (quoted by Goldhill (1991) 259), frames are always framed.
Staging Thyestes
63
cause. But this epistemic superiority is also bound to provoke further worries: as frames appear inevitably enframed, so the starting point of violence
and nefas, ever more elusive, is constantly pushed back without any guarantee that it might find an ultimate resting point. How can we safely exclude
the possibility that there has been a prologue behind the prologue? Who
sent the Fury? Does someone know more than we do, in spite of all the
indications attesting to our privileged point of view? It is precisely through
this mechanism of regress ad infinitum thematized in the play by three
inset levels of dramatic action that frames lose their potential as controlling, ordering devices and rather turn out to be no more than flimsy fences
easily trespassed by uncontrollable forces. Frames, multiplied in a sequence
of horrors, become the signposts and the harbingers of the trouncing
of order.93
In yet another sense, frames have a crucial role in the plays signifying
strategies. Framing introduces into the tragedy a structuring device which
appears to ensure the orderly succession of embedded levels of action. Yet
we are soon led to realize that the markers provided by frames guarantee
that each successive level of action is perceived as dialectically juxtaposed to
the previous one, and that critical ingenuity is required in order to extract
further meaning from this interplay of difference.94 Framing imposes
comparison between the various dramatic techniques at work in different
levels of the play, as we will see, and also between the characters who share
analogous functions in those different levels: the Fury, Atreus and Tantalus
junior, we noted, all press ahead with the development of the plot which
Tantalus ghost, the satelles and Thyestes himself resist in vain. Finally, framing introduces a comparison between the various aspects of each character
as he enters a new frame: which Thyestes is more credible the worried
exile who hesitates to follow Tantalus enthusiastic recommendations or the
gullible, power-greedy man who blithely accepts his brothers pleas? Here,
too, frames separate, and at the same time ensure that we juxtapose and
compare the actions they encase.
By privileging what they encase95 and almost literally setting each
scene against a well-defined backdrop, frames give those comparisons a
sharper profile. Seneca privileges in his plays detached scenes which do
not necessarily follow each other with compelling coherence and urgency.
93
94
95
The interplay of frames recalls in some ways Euripides most Euripidean play, Orestes, which also
shows the traditional tragic forms, saturated by too many options crowding in from the past, as it
reaches a breaking-point. See Zeitlins (1980) brilliant analysis (quotations from pp. 51 and 57).
Goldhill (1991) 236, a very perceptive discussion of the role of framing in Theocritus.
A role of frames aptly stressed by Caws (1985) 21, 262 and passim.
64
Yet, far from laying bare the rhetorical nature of Senecan tragedy, frames
underline the most profoundly dramatic aspects of the play.96 Frames define
boundaries and thus mark separation and even detachment, but precipitate
comparison and contrast. They dissect different parts of the play, but they
hardly arrange an orderly, inert sequence of tableaux; on the contrary,
they highlight the inevitable collision of dramatic levels and the relentless
conflicts that plague successive generations.
ii
As I turn now to a specific aspect of the critical detachment inherent in
Thyestes, my argument becomes more speculative. I propose to analyse how
certain features of the play invite a critical reflection on the particular form
of tragedy that Seneca is writing, and on its position in the history of tragic
forms. My starting point is the distinction between different levels of action
which I summarized earlier in the chapter. Through a succession of frames,
Thyestes isolates an inner core, a deepest level, where Atreus acts out a plot
of his own devising and successfully punishes his brother. As I observed
earlier, this third level of the tragedy is composed of a compact sequence of
scenes. It begins in the second act with the exiles arrival to his land, and the
dialogue between himself and Tantalus (40490). After Atreus monologue
at 491507, which reverts to a higher level of the tragic structure, the two
brothers meet (50845), thus sealing the fate of Thyestes children. Before
reaching the third and final scene of this third level of action, the climactic
confrontation in which the horrific reprisal is declared (9201004), the
tragedy sets aside the orderly succession of different levels observed so far
and turns to a lengthy and elaborate messengers speech.97
The three scenes which occupy the innermost frame of Thyestes constitute
a miniature tragedy. This is not, strictly speaking, an instance of playwithin-the-play; and yet there are strong enough signs that this part of the
play has a specific, well-outlined structure. (In fact, this could be seen as a
very bold and experimental example, ante litteram, of the play-within-theplay form, one in which the boundaries of fiction and reality are more fluid.)
It is certainly a performance through which Atreus achieves, all too directly,
his ultimate goal of revenge. The first scene devised by Atreus, Thyestes
arrival, is a clear marker of dramatic beginning. The emotional centre of the
play is devoted to the studied deception: the blindfolded Thyestes offers his
children as pledge of his faith (520: obsides fidei), or, in Atreus words, as
96
On this topic see ch. 6.
97
On which, see pp. 16970.
Staging Thyestes
65
destined victims (545: destinatas victimas). In the final encounter between

the two brothers Atreus revels in its image of oblivion, and even more in
Thyestes reaction to the final revelation. This structure is not very different,
for instance, from that of Sophocles Philoctetes, which hinges on Odysseus
deception. In Thyestes, this beginning, middle and end structure is encased
within a prologue and an epilogue, which problematizes the premises and
consequences of the included play. The five acts of Thyestes operate as a
framing device which embraces the three-act inner tragedy, organized by
Atreus and to a noticeable extent modelled on Greek standards. It is as
if Senecas posteriority is thematized here by an encompassing structure
which becomes a showcase for, and a commentary on, a type of tragedy
that no longer exists and is no longer viable.98
iii
Of the three levels of tragic action that I have described, only the third and
innermost level of the tragedy finds a coherent closure on stage: Atreus
revenge is plotted, carried out and revealed in the three acts that I have
analysed. But the two other levels of the tragedy the first one, in which
the preliminary decisions about the nefas were taken, and the second, in
which Atreus mulls over his plan and later reveals it to Thyestes remain
undoubtedly open and portend no shortage of future evils. In this case the
interplay of differences marks a contrast between Atreus teleological plot
and the unending potential for tragedy embedded in his familys curse. In
this way Thyestes affords its audience both the reassuring satisfaction of
closure and the tormenting promise of renewed terrors. Nothing could be
more final than the cannibalistic banquet, a funeral of sorts,99 and nothing
more open-ended than the final words exchanged by the brothers, a curse
fulfilled in subsequent stages of the mythical narrative, which the audience
can already begin to fathom (and, indeed, remember): TH. vindices aderunt
dei; | his puniendum vota te tradunt mea. | AT. te puniendum liberis trado tuis
(TH. The gods will be my avengers; my prayers deliver you to them for
punishment. AT. For punishment, I deliver you to your children, 111012).
Ultimately, even at the apex of his success, Atreus, too, cannot achieve
98
99
The five-act structure differentiates Senecas plays from their Greek models, though not the comic
ones; see Tarrant (1978) 21819, esp. 219 and n. 40 and Hunter (1985) 3542. On act division in
Senecan drama see Anliker (1960) 4993. Antecedents can indeed be found in Greek comedy,
certainly in Menanders Dyscolos and possibly in four more of his plays (Aspis, Epitrepontes, Samia
and Sicyonios). See Horaces Ars P. 18990, with Brink (1971) 24850.
See Fowler (1989) 85 = (2000) 249.
66
complete satisfaction, because his desire is so boundless that it could never,

by definition, be completely fulfilled. The micro-dramas which make up
the play try in vain to order and set boundaries onto a flow of emotional
cravings which ultimately reject all constraints.100
This contrast between closure and lack of closure, entrenched in the very
core of the plot, might be read as a reflection on the nature of tragic action
and the forms that it can take on stage. The inner plot, a self-contained
unity, does achieve a clear-cut conclusion in the horrendous anagnorisis
at line 1004. By dropping his mask as an actor and returning to the role
of author, Atreus denudes the foundations of his closed tragedy. At the
same time, he indicates his dissatisfaction with that form: there is, in fact,
no real possibility of tragic closure. Closure, as achieved on stage by the
banquet, can only be a momentary illusion, predicated on deceiving an
unguarded spectator like Thyestes. We will observe later that this is one of
the many ways in which Thyestes is portrayed as a tragic hero of old, deaf
to the cunning devices and linguistic creativity of his brother, and therefore
inevitably bound to be defeated.101 For the audience, which has been alerted
to the metadramatic organization of the play, as for Atreus, closure remains
an elusive option which the open-ended finale shatters without recourse.
There is no guarantee that the evils visited upon the House of Argos by
Tantalus reappearance have really ended. True, the Fury did say that she had
seen enough (105), but her intimation that Tantalus watch Thyestes while
he drinks his childrens blood (102), coupled with the absence of closure at
the end of the play, mean that Tantalus and his curse ominously overshadow
the plays final moments. Here, too, we can observe the discomforting effect
of open frames, of frames which fail to fulfil their task and thus highlight
the absence of a reassuring resting point. Similarly, Atreus cannot claim
that he has achieved a conclusion. He has succeeded in taking revenge, and
he has without question won this round of confrontation with his brother.
But his very last words, which in a way do signal closure by capping the
second level of tragic action he had set in motion at 176204, are in fact
totally open-ended. The frame is now complete, but its closure is more
formal than substantial.
Greek tragedy, too, is deeply sensitive to the signifying potential of closure and openness, and often hints within the play, or even towards its
conclusion, at future developments of the family plots, thus announcing
new stages of literary invention. It is only in the Oresteia, however, that
we can find something approaching the open-endedness of Thyestes. At
100
See later, p. 94.
101
I develop this line of analysis in ch. 3.
Staging Thyestes
67
the end of Agamemnon (166273), for instance, the repartee between the
coriphaeus and Aegisthus, centred on predictions of violent retribution,
offers a relevant parallel to the last few lines of Senecas play.102 The conclusion of Choephoroe, though projecting an open window onto Orestes
future, is less explicit in its heralding of specific events, and the curtain falls
on Eumenides with a convincing finality. In Sophocles it is only Oedipus
at Colonus that closes with the announcement of a sequel, as Antigone
pleads with Theseus to be sent back to Thebes, where the fight between
her brothers is imminent.103
Against the background of Greek tragedy, we can better appreciate the
novelty of Thyestes reflection on closure, yet it would be misguided to posit a
radical opposition between a tightly closed Greek model and Senecas openended finale. While it is important to recognize that the ending of Thyestes
is indeed unusual in its degree of openness, it is preferable to couch the
discussion in terms of different balances between elements of closure and
openness which characterize different tragic experiences. Sophocles intersperses his main narrative with inset stories which lack a definite ending and
repeatedly announce future developments, but the main plot does normally
achieve its telos, which thus appears to be challenged by the divergences of
the inset narratives.104 Instead, Seneca opposes the openness of his main
plot to the included closure105 of Atreus mise en abyme. The Sophoclean
dialectical model is also present, of course, in Virgil. The unending debate
between openness and closure in the Aeneid, the constant struggle between
Jupiters teleology and Junos passion for delays, digressions and alternative
plots is extended within the text, while the poems end offers (ostensibly) a
Jovian, closural ending.106 Seneca turns this model inside-out, upturning
102
103
104
105
106
Aeg. But to think that these men should let their wanton tongues thus blossom into speech against
me and cast about such gibes, putting their fortune to the test! To reject wise counsel and insult
their master! Chorus It would not be like men of Argos to cringe before a knave. Aeg. Ha! Ill visit
you with vengeance yet in days to come. Chorus Not if fate shall guide Orestes to return home.
Aeg. Of myself I know that exiles feed on hope. Chorus Keep on, grow fat, polluting justice, since
you can. Aeg. Know that you will make atonement for your insolent folly. Chorus Brag in your
bravery like a cock beside a hen. Clyt. Care not for their idle yelpings. I and you will be masters of
this house and order it aright (trans. Smyth).
Ant. . . . Then send us back to Thebes, if yet we may heal this mortal feud and stay the self-wrought
doom that drives our brothers to their tomb. Theseus I will do that . . . (176976). Note, however,
that the chorus has the last word: and now cease your laments. Everything has been ordered
appropriately (1779).
On Sophocles: Knox (1964); Roberts (1989); Kraus (1991). On Aristophanes: Bowie (1993). See also
Foley (1985).
A term I borrow from Fowler (1997a) 18 = (2000) 302.
See Hershkowitz (1998) 68124; Quint (1993) 5096. On epic, not only Virgilian, closure: Hardie
(1997).
68
the balance of the two principles. He reflects on epics ambivalent desire

for closure as much as epic reflects upon tragedys, and calls into question
both genres hope (or illusion) of actually achieving a definite closure.
The lack of closure at the end of Thyestes is compatible with the fact that
the play carefully eschews a conclusive moral statement. Thyestes appeal
to an absolute moral order is immediately quashed by Atreus mocking
disillusionment, and even the chorus, as we will see, is denied (literally) the
last word.107 Openness is the vehicle which finally expresses the absence
of absolute points of reference, of divine guarantees of order and morality
to which characters and audience alike can turn in search of an authoritative take on events. Here, again, Seneca seems to have gone one step
further in the deconstructive approach displayed by the Aeneid; it is no
longer a question of opposing Jupiters teleology to Junos openness, since
we are not even sure that a divine figure can effectively embody superior
principles.108 Significantly, as Atreus questions Thyestes gesturing towards
divine sanctions (both moral and narrative), he is ultimately interrogating
the very nature of his victory. He succeeds in undermining Thyestes, but he
is nonetheless confessing that closure must be temporary and illusory for
himself as well: he might not care in this moment of triumph, but we know
that he will indeed care later. His mythical saga is emphatically not over.
We understand now how important it is that the tragedys conclusion
forcefully recalls and returns to its beginning. The very structure of the final
exchange between Atreus and Thyestes mirrors the one between Tantalus
and the Fury. The parallelism again reinforces the sense of openness repeatedly suggested in the play: indeed, it is almost as if the tragedy closed
(momentarily) with a new prologue, a preview of coming attractions. We
will see that this deconstruction of beginnings and endings carries with
it a sense, among other things, that finality is elusive and that repetition
inevitably prevails. At the beginning of Agamemnon it will be Thyestes who
returns to earth as a ghost and sets in motion a new plot of revenge.109
Atreus ability to bring his own masterplot to a definite close is a function
of his power in the tragedy. His intellectual and material resources are
definitely superior to those of Thyestes and allow him to execute his plot to
the letter. He is impotent, however, to control the tragedys final openness,
the promise of future evils which will tip the scale against him and his
descendants. As a playwright, and as a tyrant,110 Atreus power is shown to
107
108
110
Ch. 4, pp. 1756.

109 See p. 203.
On this final reference to the gods see p. 152.
Note that the language of poetic achievement evoked at 8858 (above, p. 59) neatly doubles up as
a reference to the rulers katasterismos.
Staging Thyestes
69
be almost absolute, but temporally limited. His victory, complete as it might

seem, is hardly an instance of imperium sine fine: the end of Thyestes proves
the inevitability of a next instalment. Thyestes, by triggering the illusion of
teleology in its inset play, is the only one of Senecas tragedies to develop a
significant metadramatic reflection on this issue, and in a sense it provides
a theoretical explanation for the uncertain closures which we find in other
plays. If the ability to close off narrative is indeed a sign of power, then
the lack of closure which characterizes Thyestes and other Senecan tragedies
must be seen as an indicator of the fact that power structures and roles are
inherently fluid, and that the stable certainties of Stoic fate are ultimately
predicated on the unstable and unforeseable workings of Fortuna.
The constrast between the internal closure and the openness staged by
frames can also offer some insight into the relative artistic appeal of these two
methods of structuring a text. Atreus plot converges on the intense pleasure
of anagnorisis, which implies for him the ultimate success of his plot, and
for the audience is the most intense point of the whole play, the moment
of revelation. The plays final postponement of closure, however, defers
fulfilment and pleasure, pointing towards an as yet unseen tragedy. These
are two different principles of aesthetic pleasure, and Atreus ultimately
proves his mastery in both; in this, too, he shows a multiple personality
consistent, as we will see, with his Dionysiac nature.111
Although Atreus fashions himself (and, to a great extent, actually is) controlling, he is in turn controlled by the Fury and the overarching authority
of the author. It is Senecas decision, ultimately, to deny his play an organic
and convincing closure, and to let it end with the ominous announcement
of future nefas. In writing the tragedy Seneca has already defied the intimation of closure that came to him in the form of literary tradition, has
reopened in a self-conscious, painful fashion a tragic discourse whose vitality had by his time already been questioned. At the end of the play it falls
once again to the authorial voice, masked by the silent Fury, to declare the
issue forever open.
111
Ch. 3, passim. Atreus also appears to be flouting gender boundaries, as discussed in ch. 3. Narrative
closure and openness (or, better, different degrees of closural definition) are susceptible to a genderoriented interpretation. See Fowler (1997b) 10 = (2000) 293, who relates closure and openness to
logo- and phallocentricity, building on Cixous (1986) 88, (1991) 4950 and Gallop (1982).
chapter 3
A craftier Tereus
t h r a c i u m n e fa s
t i t. How now, Lavinia? Marcus, what means this?
Some book there is that she desires to see.
Which is it, girl, of these? . . .
b oy. Grandsire, tis Ovids Metamorphoses . . .
(Shakespeare, Titus Andronicus 4.1)
i
Explicitly invoked by the Fury in the prologue and by Atreus in his inspired
monologue, the allusive pattern which links significant moments of Thyestes
to the Ovidian tale of Tereus and Procne (Met. 6.412674) is crucially important. Both stories culminate in the revengeful slaughter of children who
are then cooked and served to their ignorant fathers in perverse banquets;
neither narrative spares its readers the goriest details.
Senecas recognition of Ovids Tereus as the foremost archetype of narrative violence will be heeded centuries later by Shakespeare in Titus
Andronicus, a play steeped in classical sources (Seneca and Ovid), and
routinely criticized for its grotesque excesses of violence and goriness.1 As
Marcus first catches sight of Lavinias violated body, he not only evokes the
Ovidian model, but reiterates the agonistic comparison with Ovid inaugurated by Senecas maiore numero (2.4.3843):
Fair Philomel, why she but lost her tongue,
And in a tedious sampler sewd her mind;
But, lovely niece, that mean is cut from thee.
1
On Titus classical background see Waith (1957); Tricomi (1974); Miola (1983) 4275; Bate (1995)
902; James (1997) 4284. The emphasis on the Ovidian model is justified, but should not detract
from the importance of Thyestes, much favoured by an earlier generation of critics, as appears now
to be the case (Baker (1939) 11939, endorsed by Bate (1995) 29, n. 2). The strong metatheatrical
component of Titus is arguably inspired by Seneca rather than Ovid.
70
A craftier Tereus
71
A craftier Tereus, cousin, hast thou met,

And he hath cut those pretty fingers off,
That could have better sewd than Philomel.
Lavinia overcomes the silence to which mutilation has doomed her by

pointing her family to the relevant section of the Metamorphoses, just as
the horrors of Thyestes emerge from the silence that Tantalus ghost had
advocated in vain, thanks to the powerful inspiration of Ovids Thracium
nefas.
Thematic parallelism offers an ostensibly adequate explanation for the
intertextual relationship, a relationship which is further encouraged by
the connection that the Metamorphoses establish, in turn, not only with
Sophocles prototypical (and lost) Tereus, but also with Accius Atreus, one
of Senecas most influential models.2 On a more general level, it is hardly
surprising to find the Metamorphoses registering in the intertextual background to Senecan tragedy, whether one might want to explain this pervasive phenomenon as a generic (Silver?) stylistic affinity or a more pointed
signifying strategy.3 Plotting, however, is not the only aspect of these texts
that bears comparison, and I will in fact postpone this issue for the time
being.
As we have seen, the Fury acknowledges the hellish atmosphere of the
Tereus story4 and places her own endeavours under the aegis of Tereus
nefas, advocating a new Thracian nefas5 with even more victims.6 This
strategy of excess announced at the outset is clearly a very important aspect
of the way in which Thyestes will negotiate its relationship with the Ovidian
model. But even more significant is the presence of such a clear programmatic intention at the core of a section of the play with strong metadramatic
resonances. Thracium fiat nefas is the founding gesture of a tragedy which
will come to light under the ominous auspices of its astonishingly violent
precedent. Indeed this opening announcement instructs any comparison
of the two texts to take into account the whole span of Ovids episode,
and not just its climactic resolution, even if the Fury stresses (if the text is
2
3
4
5
6
The most significant points of contact between Accius Atreus and Ovid in this episode are registered
by Bomer (196982) ii. 117. On the possible connections with Accius Tereus, and in general, see
Frankel (1945) 37781. Jacobi ((1988) 153) argues against a direct connection between Seneca and
Accius.
The issue of the relationship between Seneca and Ovid has received considerable attention (recently
from Jacobi (1988)), but a satisfactory critical analysis is still missing.
This atmosphere and its implications will be discussed by Ingo Gildenhard and Andrew Zissos in
their forthcoming monograph on Ovids Metamorphoses.
The whole episode of Tereus, Procne and Philomela begins in Ovid with the words Threicius Tereus
(6.424), almost an internal heading.
Maior can also carry metaliterary overtones, as, famously, in Virg. Aen. 7.445 (see above, p. 31).
72
sound) that the finale will indeed form a privileged point of contact: not
yet does Thyestes cry for his sons and when will he [sc. Atreus] lift his
hand? (589: nondum Thyestes liberos deflet suos | ecquando tollet?).
The story of Tereus and Procne, to be sure, is not just a celebrated tale of
violence, revenge and moral ambiguity which displays significant analogies
with the plot of Thyestes. More importantly, the myth of Philomela can
be considered the ur-myth about the origin of a certain type of poetry,
one which is produced by the disorder of relations and the confusion of
identity represented as incest, cannibalism or civil war,7 and is fuelled by
the dark forces of violence and vengefulness. In the background of Thyestes
stands the incestuous relationship between Thyestes and his sister-in-law
Aerope. This concern for perturbed family connections (dubius sanguis,
240)8 will lead Atreus to pour his nephews blood into his brothers cup,
to find a cannibalistic resolution to the fratricidal strife that for a Roman
audience inevitably recalls the horrors of civil war. By positing at the outset
such a strong correlation with the Procne episode, the plot of Thyestes alerts
us to its powerful implications for a reflection on poetry, its power and its
dangers.
This set of associations is recognized in Shakespeares Titus Andronicus,
where it is actually amplified and extended, as several characters deploy
their passions lust, revenge, ambition in self-consciously metatheatrical
fashion. The arch-villain Aaron is a master of words as his name suggests9
who engineers the larger part of the plot and is fully conscious of his
metatheatrical role (5.1.636):
For I must talk of murders, rapes and massacres,
Acts of black night, abominable deeds,
Complots of mischief, treason, villainies,
Ruthful to hear yet piteously performed.10
His lover and accomplice Tamora is equally aware that she is setting in
motion the complot of this timeless tragedy (2.3.265) as she writes and
delivers a fake letter which will precipitate the death of Titus sons. Later
in the play she tries to deceive Titus by staging a nightmarish pantomime
(I must ply my theme, 5.2.80) in which she acts as Revenge (I am not
Tamora: | She is thy enemy and I thy friend. | I am Revenge, sent from
thinfernal kingdom . . ., 5.2.2830) and her own children are cast in the
7
9
10
8 See below, p. 102.

Kilgour (1990) 33.
An Elizabethan audience would have known that the biblical Aaron had an eloquent, persuasive
tongue (Exodus, 4.1016) (Bate (1995) 125).
Lamentable to hear about, yet done in order to excite pity (Bate (1995) 247).
A craftier Tereus
73
roles of Rape and Murder. Titus himself, as he prepares to take his own
revenge, invokes the Ovidian mastertext, For worse than Philomel you
used my daughter, | And worse than Procne I will be revenged (5.2.1945),
and proceeds to stage his own version of the Thyestean banquet: Ill play
the cook (5.2.204).
Most strikingly, Shakespeares Lavinia collapses the two Ovidian
archetypes of female ability to overcome censorship and repression through
a mediated form of quasi-artistic expression. She is a novel Philomela, as she
makes explicit by pointing out the relevant sections of the Metamorphoses,
but she then proceeds to write down the names of the villains with a staff
on a sandy plot (4.1.69), much like Io had done, once metamorphosed
into a cow (Met. 1.64950). Both writing and poetry are born under the
same sign that of a violated woman who cunningly devises alternative
means to tell her story.11
Words play an important role already in the first part of the story of
Tereus. Overcome by his desire for Philomela, Tereus showers her with
emotional language which is supposedly conveying Procnes desire for her
sister to come to visit her, but in truth merely encodes his lust into acceptable
(speakable) forms: facundum faciebat amor (love made him eloquent, Met.
6.469). It is thanks to this elaborate and deceitful speech that Tereus is
able to overcome his father-in-laws disapproval and depart with Philomela
(6.4734). Beguiling words become once again Tereus preferred weapon,
when he falsely reports to his wife that Philomela is dead and misleads her
with a moving description of commenta funera (6.565). Only once in this
first part of the tragedy does language function transparently. Significantly,
this happens when Tereus, alone with his prey, unveils to her the true nature
of his unlawful and unholy desire (6.51926):
iamque iter effectum iamque in sua litora fessis
puppibus exierant, cum rex Pandione natam
in stabula alta trahit silvis obscura vetustis
atque ibi pallentem trepidamque et cuncta timentem
et iam cum lacrimis, ubi sit germana, rogantem
includit fassusque nefas et virginem et unam,
vi superat frustra clamato saepe parente,
saepe sorore sua, magnis super omnia divis.
The voyage now is done, and now they leave the weary ship and land on their own
shore; and then the king drags off Pandions daughter up to a cabin in the woods,
remote and hidden away among dark ancient trees, and there pale, trembling,
11
Cunning being a trait most associated with the feminine domain: Zeitlin (1996) 349, 358.
74
fearing everything, weeping and asking where her sister was, he locked her, and
revealed his own black heart and ravished her, a virgin, all alone, calling and calling
to her father, calling to her sister, calling, even more, to heaven above.
This time words do tell the truth, and this truth is the central crime of the
scene. Indeed, Tereus consistently embodies the connection between nefas
and words, both when he exploits them in order to conceal his desires,
and when, once the restraining factors represented mainly by the presence
of Pandion have been overcome, he is finally free to remove the mask of
repression and voice his desire.
The privileged relationship between words and nefas is also apparent
ex negativo in the brief section where Philomela vows to report the whole
truth about Tereus crimes. She will set aside her modesty and speak up
(6.5448):
ipsa pudore
proiecto tua facta loquar; si copia detur,
in populos veniam; si silvis clausa tenebor,
inplebo silvas et conscia saxa movebo.
audiet haec aether, et si deus ullus in illo est.
Ill shed my shame and shout what you have done. If Ive the chance, Ill walk among
the crowds: or, if Im held locked in the woods, my voice shall fill the woods and
move the rocks to pity. This bright sky shall hear, and any god that dwells on
high!
Her wish is brutally crushed by the mutilation that Tereus inflicts on

her as she tries to speak, luctantem loqui (6.556). The violent removal of
Philomelas tongue shatters her illusion that words, if clear and explicit, can
still have a role in the world of nefas. It is interesting to contrast the dark
setting in which Tereus reveals his passion to her (6.524: fassus . . . nefas),
redolent of the obscure, menacing shadows that haunt infernal or semiinfernal landscapes one is reminded of the locus horridus in Oedipus and
Thyestes with Philomelas vain invocation of a divine order represented
by the aethers listening to her (6.548).
The second act of Ovids narrative is taken up with Philomelas cunning
attempt to reveal the hidden truth, and Procnes plotting of a terrible revenge
against her husband. It is in this context that words spoken or otherwise
conveyed again play a crucial role. Philomela has no way to express her
feelings verbally. Her mouth lacks a means of expression (6.574: os mutum
facti caret indice), but she can resort as an alternative to ingenium and
sollertia (6.575).12 The novel indicium sceleris will thus be a craftily textured
cloth (6.5769):
12
Ingenium being, of course, a poets virtue; see OLD s.v. 5.
A craftier Tereus
75
stamina barbarica suspendit callida tela

purpureasque notas filis intexuit albis,
indicium sceleris, perfectaque tradidit uni,
utque ferat dominae, gestu rogat;
on a clumsy native loom she wove a clever fabric, working words in red on a white
ground to tell the tale of wickedness, and, when it was complete, entrusted it to a
woman and by signs asked her to take it to the queen.
The outcome is indeed a carmen13 (6.582), which her sister can read in
the fabric in spite of the constraints and the repressive violence to which
its author has been subjected: Philomela, though horribly mutilated, has
found what T. S. Eliot will call her inviolable voice (The Waste Land
101). Philomelas muted words actualize the double meaning of textus as
both cloth and text and evoke the metaphorical association between
weaving and plotting:14 her message thus overcomes the repressive force
of violence described in the earlier part of the story. To the strength of
Tereus actions Philomela opposes the silent reproach of her embroidered
messages, powerful enough to unleash Procnes avenging furor (6.5816).
I touch here upon an issue of crucial importance in the ideological texture
of Metamorphoses. There, too, Philomelas web strengthens the equation of
poetic word and return of the repressed which I have highlighted in Thyestes.
The structural function of Philomelas words is not dissimilar to that of
Tereus normal verbal utterances. His dissemblance breaches the decorum
of silence that he should be respecting and displaces his feelings. The limitations imposed upon Philomela are those of violence and confinement, yet
words function in the same way, by slyly overcoming the barrier of silence
and inaction. It is precisely this consistency in the way words operate that
makes them ambivalent and double-edged. Words are inherently disruptive
because they can subvert moral principles just as easily as nefarious ones.15
Even in the latter case, however, the ultimate balance of good and evil
is difficult to ascertain. Certainly Philomelas encrypted words overcome
Tereus immoral orders, and manage to reveal the fate she has suffered.
The impressive might of her words is emphasized by their positioning in a
13
14
15
Ovid does not reveal explicitly whether the carmen is composed of images or letters (see Bomer
(196986) i i.158). The connotations of carmen, however, are unmistakable.
Bergren (1983) 715; Scheid and Svenbro (1996).
Other emphases, of course, can be preferred. In a perceptive article (Segal (1994)), for instance,
Charles Segal has chosen to insist rather on Procnes ability to keep her reactions at bay when
she deciphers her sisters messages. This is certainly an important aspect of the narrative, which
touches more on the issue of the readers response to the (poetic) message. As far as emotional and
cognitive dimensions are concerned, however, this episode underlines the liberating potential of the
poetic word. I would also be inclined to argue that what Segal considers to be the unsatisfactory
pseudosolution of the final metamorphosis is in fact a compromise between an attempt to somehow
condemn the cruelty of the tale and the need to preserve the storys confused moral balance.
76
linear sequence which connects crime to punishment: Philomelas writing

becomes an indispensable instrument of revenge. Yet this revenge is highly
problematic, because Procnes actions will reduce the moral chasm between
herself and Tereus to dangerously narrow proportions. Procnes perverse
revenge problematizes the readers ability to side emotionally with either
the victims or the villains.16 The words used to describe Procnes reaction
on reading the woven carmen fasque nefasque | confusura ruit poenaeque in
imagine tota est (she stormed ahead, confusing right and wrong, her whole
soul filled with visions of revenge, 6.5856) apply equally well to the
readers confusion, as they suspect in Procnes muted rage the intimation
of further unspeakable violence to come. Struck by the unusual message that
she has received, Procne reacts with apparently restrained emotion. Once
again, the words and cries that her dolor would have normally elicited must
be stifled in the presence of a violent, repressive force. In this case, however,
Procnes conscious and voluntary gagging of her emotions is geared towards
a delayed yet fuller satisfaction, one that will come not from words but from
deeds. For us, reading this scene of reading, this is yet another confirmation
of the psychological impact of words, which rouse emotions and stir up
violence.
Since they are the cause of Procnes avenging murder, Philomelas words
testify once again to the close connection between words, which work
against repression, and nefas.17 By revealing to her sister the nefas she has
endured, and stirring her Bacchic furor,18 Philomela is functioning here as
a quasi-divine mover of events. If Procnes reaction is reminiscent of Virgils
Amata the Bacchic woman who sets in motion, at the human level, the
violent narrative of the second half of the Aeneid Philomela is structurally
analogous to Juno and the Fury, whose decision and responsibility it is to
stir Amata to action. Thus, by speaking unspeakable words, Philomela is
endowed with the same inceptive function assumed by the Fury in the
prologue to Thyestes.
16
17
18
A point well stressed by Segal (1994).

I am thus inclined to disagree with Segals rather optimistic conclusion that as the web of words that
calls attention to its textual origins, it [the weaving] objectifies the crime and in that way enables
the reader to take the full measure of its horror (Segal (1994) 266).
The Dionysiac connotation is reinforced by the Horatian model for fasque nefasque at Met. 6.585,
that is, Carm. 1.18.711 (at 10): ac ne quis modici transiliat munera Liberi, | . . . | . . . monet Sithoniis
non levis Euhius, | cum fas atque nefas exiguo fine libidinum | discernunt avidi. Note the reference to
Sithoniae . . . nurus immediately following at Met. 6.588. See Titus Andronicus 2.1.1335 (Demetrius,
Tamoras son, as he accepts Aarons invitation to revel in Lavinias treasury (131)): Sit fas aut nefas,
till I find the stream | To cool this heat, a charm to calm these fits, | Per Stygia, per manes vehor (the
last line is based on Sen. Phaed . 1180: per Styga, per amnes igneos amens sequar (Phaedra vowing to
pursue Hippolytus)).
A craftier Tereus
77
It is important to stress that female characters embody this force of

(written) poetic creation associated with a removal of repression, as explicitly recommended, indeed, by Ovid himself at Ars amatoria 3.61132. Even
under strict control, women will always be able to deceive dare verba
husbands and custodians: tot licet observent, adsit modo certa voluntas, | quot
fuerant Argo lumina, verba dabis (even if as many guardians watch you as
Argos had eyes, youll manage to deceive if you really want to, 3.61718).
We face here an instance of the Bacchic paradigm that endows women
and goddesses with a subversive creative power such as that of Juno in the
Aeneid,19 who challenges and sabotages at every turn Jupiters fixed, teleological prescriptions. This disruptive power is one that Atreus a Bacchic
character whose self-identification with a female character, Procne, highlights his complex gender connotations claims for himself. But we also
face a subversive transformation of the traditional prescription of female
silence and tameness into an exuberant, active and pernicious loquacity.
Readings of the Procne story that are informed by a feminist perspective
have focused especially on the voice of the shuttle, and its potential to
grant voice and power to the silenced weaving of women.20 It is a voice
whose profoundly disturbing energy is never lost sight of: it can denounce
crimes, and call for revenge, just as easily as it can unleash the powers of
hell.
Thyestes lays great emphasis on the force of poetic language. The prologue, as we have seen, represents both the bond between words and nefas,
and the violent dialectic between repression and its removal which words
precipitate. Other parts of the play, too, insist on this connection. Atreus
cunning use of words appealing, mendacious and ultimately victorious is consistently matched against Thyestes inability to look beyond
their literal surface, hence his final demise. Atreus has read his Ovid, and
displays through a number of revealing allusions a detailed knowledge of
Tereus story, in particular of Procnes avenging plans. Thyestes, on the other
hand, fails to remember Tereus plight and is thus unprepared to counter
his brothers plan. He betrays his lack of awareness when he proclaims
that lacrimis agendum est (it is time for tears to push forward my case,
Thy. 517) a doomed proposition which ironically ignores Procnes much
more effective injunction to the contrary: non est lacrimis hoc inquit agendum (Ov. Met. 6.611) and Junos authoritative precedent non lacrimis hoc
19
20
On women as catalysts, agents, instruments, blockers, spoilers, destroyers see Zeitlin (1996) 347.
See Joplin (1984), an insightful reading of the Tereus and Procne episode from a feminist standpoint.
See also the recent essay by Richlin (1992). On the theme of violence in Ovid see Galinsky (1975)
13240.
78
tempus ait Saturnia Iuno (Virg. Aen. 12.156: no time for tears, now said
Iuno daughter of Saturn). Atreus highly figurative and rhetorically powerful form of expression embodies the poets own craftmanship. Like the
Muses, who traditionally can sing the truth, but can also sing convincing lies,21 poets can bend words to express any feeling and any emotion,
whether true or false, deeply moral or astonishingly cruel. New prominence, too, is granted to the notion that the play presents itself as a way
to overcome ideological, literary and ethical limitations. The strength of
Philomelas web is indeed the strength of Thyestes as a whole.
Both texts show that the words of poetry can reveal unexpected extremes
of violence, and that there is no limit to the creativity of human wickedness.
Seneca competes with his model at a metanarrative level as well, further
blurring the distinction, already problematic in Ovid, between good and
evil. Atreus desire to surpass all previous horrors powerfully reflects the
plays agonistic relationship with its literary ancestor.22 By remembering and
repeating well-known criminal deeds, those of Tereus and Procne, Seneca
is already raising the moral stakes of his own writing, since his rewriting
will necessarily exemplify a new, bloodier advance in the literary depiction
of horrors, and will necessarily result from yet another brutal breach of the
decorum of silence. If Tereus and Procnes final metamorphosis guarantees
that their violence will forever be encoded in the bloody stains of their
feathers, then Senecas play testifies that the message has not been lost.
ii
After highlighting, in Ovids narrative, concerns and reflections which carry
perceptible metaliterary overtones, I would like to focus on a comparative
analysis of the two plots, which, by introducing for the first time a set of
concepts central to the rest of this book, will greatly assist an understanding of some key features of Senecas play. It is useful to look at analogies
and important differences between the two episodes, especially outside the
comparison between the two banquet scenes: if we focus a comparison exclusively, or even predominantly, on these sections, we may end up playing
down the extent to which the two texts clash in their articulation of the
plot.
On the one hand, Ovid plots his story on a large narrative stretch which
encompasses a series of episodes that are all closely linked to each other
21
22
Bergren (1983) is excellent on the specific connection between Muses, women, truth and language.
See Tarrant (1985) 130.
A craftier Tereus
79
and are all equally indispensable for a coherent understanding of the finale.
His story is neatly divided into two main sections, the first one leading
to Philomelas rape and mutilation, the second, chronologically distinct
(Met. 6.571: signa deus bis sex acto lustraverat anno, through all the twelve
bright signs of heaven the sun had journeyed), centred upon Procnes revenge and leading up to the banquet and the final metamorphosis. Thyestes,
on the other hand, chooses to stage only the last part of the confrontation
between the two brothers, and voices the causes of Atreus furor in a limited
number of carefully worded, almost coded references. Yet the programmatic
reference to the Thracian nefas which defines the Furys creative intention
invites further comparisons between the two plots.
To begin with, the first part of the Tereus episode foregrounds the same
themes of deceit and betrayed trust that play an equally important role
in Thyestes. In Ovids poem, the narrators voice explicitly insists on the
contrast between reality and appearance which only Philomelas cunning
stratagem will reveal to Procne. Tereus, madly in love with his wifes sister,
is able to conceal his lust under the veil of soothing words (6.46974):
facundum faciebat amor, quotiensque rogabat
ulterius iusto, Procnen ita velle ferebat;
addidit et lacrimas, tamquam mandasset et illas.
pro superi, quantum mortalia pectora caecae
noctis habent! ipso sceleris molimine Tereus
creditur esse pius laudemque a crimine sumit.
Love made him eloquent; and, if at times he pressed his pleas too far, why, Procne
wished it so; he even wept, as if shed ordered tears. Ye Gods above, how black the
night that blinds our human hearts! The pains he took for sin appeared to prove
his loyalty; his villainy won praise.
In Pandions trusting of his daughter to Tereus, we recognize what Thyestes

himself will do: both men maintain that fides will guarantee the safety of
their offspring. Pandion reminds his son-in-law of his obligations (498), and
Thyestes, in an astounding feat of self-deception, goes as far as claiming
that those children will in fact ensure his own loyalty: as pledge of my
faith, brother, take these innocent boys (Thy. 5201: obsides fidei accipe |
hos innocentes, frater).
The second segment of the story told by Ovid Philomelas rape
does not find a direct counterpart in Senecas play. It is very much present,
however, in the background to Thyestes, that part of the mythical plot
which is not directly staged but is alluded to and offers a very interesting, if
incomplete, explanation of the events Seneca portrayed. Philomela attacks
80
Tereus, blaming him for the violence she has endured, for ripping apart the
family (Met. 6.5378):
omnia turbasti: paelex ego facta sororis,
tu geminus coniunx, hostis mihi debita poena.
You have confused everything. Im made a concubine, my sisters rival; youre a
husband twice, and Procne ought to be my enemy!
The charge is not dissimilar to Atreus invective against his brother as he is

planning his revenge (Thy. 2224):
coniugem stupro abstulit
regnumque furto: specimen antiquum imperi
fraude est adeptus, fraude turbavit domum
My wife he took away with his debauchery; he stole my kingdom; the ancient
token of our dynasty he gained by fraud, by fraud unsettled our house.
This background is necessary in order to understand fully the rigorous

selection of relevant aspects of the plot which Seneca operates vis-à-vis his
model. All the intertextual pointers concur in establishing a connection
between Atreus and Procne.23 As he sets out to repeat the horrors witnessed
by Tereus family in the even more audacious form willed by the Fury,
Atreus follows in Procnes footsteps while planning the nefas and carrying
it out.
The allusion to Ovid is clarified in the crucial scene of Thyestes where
Atreus reveals his plans to the counsellor. Even in this deceptively clearcut case it is interesting to look for specific insights on how Seneca has
systematically reworked his model.
The beginning of Atreus monologue condenses in a question the experiences and emotions which Ovid had divided between Philomela and
Procne. In Atreus words to himself (17880)
post tot scelera, post fratris dolos
fasque omne ruptum questibus vanis agis
iratus Atreus?
after so many crimes, after a brothers treacheries, and breaking every law, you are
busy with idle complaints is this Atreus in a rage?
we recognize Procnes impatient exhortation to action after Tereus crime

has been revealed (Met. 6.61113):
23
Both are compared to a tigress, at Ov. Met. 6.6367 and Thy. 70714, on which see below, p. 123
and nn. 11112.
A craftier Tereus
81
non est lacrimis hoc inquit agendum,

sed ferro, sed siquid habes, quod vincere ferrum
possit.24
This is not time for tears, but for the sword, she cried, or what may be mightier
than the sword.
This intertextual strategy sheds some light on the multifaceted character

of Atreus, who, in the portion of the plot that is elaborated by Seneca, is
a new avenging Procne, but also represents himself as a female victim a
battered Philomela. It is through allusion that Atreus protestations about
his own rights acquire the special emotional value warranted by Philomelas
innocence. And a similar overtone could readily be detected in line 220,
where Atreus replies to the shocked counsellor that, when it comes to
Thyestes, the very notion of fas becomes blurred beyond recognition. Fas
est in illo quidquid in fratre est nefas (whatever is wrong to do to a brother
is right to do to him, Thy. 220) recalls, in its apparent oxymoron, the
moral justification that Procne uses to absolve herself as she contemplates
the punishment she has in mind for her husband: scelus est pietas in coniuge
Tereo (Met. 6.635: loyalty for a husband like Tereus is a crime).
It is Atreus, again, who recalls Procnes words magnum quodcumque
paravi: | quid sit adhuc dubito (Met. 6.61819: some mighty deed Ill dare,
Ill do, though what that deed shall be, is still unsure) as he announces
his plan: haud quid sit scio, | sed grande quiddam est (I do not know what
it is, but it is some great thing, Thy. 26970).25 And both Atreus and
Procne are able to imagine in detail the final outcome of their revenge:
poenaeque in imagine tota est (Met. 6.586: her whole soul is filled with
visions of revenge) prefigures Senecas tota iam ante oculos meos | imago
caedis errat (already before my eyes flits the whole picture of his slaughter,
Thy. 2812). In order to pursue his revenge, Atreus displays qualities traditionally associated with women in Greek and Roman culture: secrecy,
guile, entrapment.26 Thyestes trust in the traditionally male qualities of
steadfastness and earnestness which he advertises especially in act 3 will
prove to be no match.
24
25
26
See also the exchange between the satelles and Atreus at Thy. 257 (on which see above, p. 17,
n. 28): SAT. ferrum? AT. parum est. SAT. quid ignis? AT. etiamnunc parum est, which may also recall
Accius Procne planning her revenge atque id ego semper sic mecum agito et comparo | quo pacto
magnam molem minuam (6345Ribbeck2 = 4467 Dangel). On the possible connection between
the Dionysiac atmosphere of Accius Tereus and Thyestes see p. 133, n. 137 below (on 4467 Dangel).
See Ovids Medea (Her. 12.212): nescio quid certe mens mea maius agit, which echoes in a metanarrative
vein Propertius 2.34.66 (nescio quid maius nascitur Iliade): Bessone (1997) 3241, 2824.
Zeitlin (1996) 360.
82
In the final revelation of their plot, Atreus and Procne show their affinities once again: quidquid e natis tuis | superest habes, quodcumque non
superest habes (whatever is left of your sons, you have; whatever is not
left, you have, 10301) recalls intus habes, quem poscis (Met. 6.655: its
inside you, the son youre looking for). Also the more shaded connection between Atreus and Philomela resurfaces in the cruel joy with which
Atreus appears to accomplish what for the girl had remained an unfulfilled
desire. To Philomelas muted satisfaction nec tempore maluit ullo | posse
loqui et meritis testari gaudia dictis (6.65960: she never wanted more her
tongue to express her joy in words that matched her happiness) he can
oppose eloquent cries of joy: nunc meas laudo manus, | nunc parta vera est
palma. | perdideram scelus, | nisi sic doleres (Now I praise my handiwork;
now is the true palm won. I would have wasted my crime, if you werent
suffering this much, Thy. 10968).
The analogy established in the text between Atreus and his Ovidian
models lends him a psychological chiaroscuro, and further discourages the
temptation to oppose bluntly his supposedly all-negative ethos to the supposedly positive ethos of his brother Thyestes. From the Metamorphoses,
in fact, Seneca inherits the key issue of the ethical responsibility of the
main characters (Tereus and Procne), an issue which the final metamorphosis pointedly refuses to resolve by sealing the fate of both spouses in
a new, but eternal, condition. To Tereus responsibility Ovid opposes a
monstrous revenge with intractable moral implications. If Tereus tyrannical cruelty is neither lessened nor justified by the terrible punishment
his wife prepares for him, neither does Procne personify a fully endorsable
moral option. The pointed and systematic connection between the Metamorphoses and Thyestes reinforces precisely this precarious and destabilizing
morality.
The impact of the final banquet on the overall ethical connotation of
the Ovidian characters applies, in reverse, to the tragedy as well. For one
thing, the link between Atreus and Procne invites the reader to credit Atreus
with the same objective, if partial, justification which monstrous suffering
guarantees to both Procne and Philomela. Secondly, the association between
Thyestes and Tereus reflects upon this apparently blameless victim the
inhuman traits which make Tereus redemption impossible even in the
context of his extreme punishment.
A synopsis of the two plots offers one final insight. The Tereus episode
foregrounds the notion that victims will turn into executioners, and, of
course, vice versa. Philomela and Procnes revenge against Tereus is increasingly horrific. At the end of the tale Tereus, the violent and cunning villain,
A craftier Tereus
83
is thoroughly defeated. This intertextual model introduces in Thyestes, too,

a lesson of reciprocity and continuity that is pointedly reinforced by other
features of the play. The Fury had already made clear from the very beginning, in a series of generic statements, that the vicissitudes of revenge
and counter-revenge would continue: certetur omni scelere et alterna vice |
stringatur ensis (Make them vie in every kind of crime and draw the sword
on either side, 256). The story of Tereus, Procne, Philomela and their
successors unequivocally confirms this sinister intimation.
iii
The two lines along which I have chosen to carry out a comparison between
the Procne episode and Thyestes neglect a number of basic characteristics
of the Ovidian story, since they only concentrate on analogies and differences of plot and on thematic implications. Yet at this point it is fruitful
to take into account a more basic and specific implication suggested by
this intertextual connection. As I have already mentioned in passing, the
Fury and Atreus do not simply refer to a generic plot for the Procne story,
but explicitly invoke the specific instantiation of that mythical story-matter
accomplished by Ovid in his Metamorphoses. The story of Tereus was certainly a productive tragic theme at least from Sophocles onwards,27 yet it is
to his epic predecessor, rather than, for instance, to the tragedy of Accius,
that Seneca insistently refers. The importance of this choice is heightened,
of course, by the prominent position of the Furys initial arousal of the
Thracium nefas in a prologue fraught with programmatic, metaliterary implications. Thus the recognition of the fact that Tereus and Procne are tragic
characters only throws into sharper relief the fact that Thyestes invokes an
epic text as its authorizing Muse, and as a fundamental model that must
not only be equalled, but surpassed.
Generic affiliations become all the more pertinent when Ovid himself steps into the picture, so that simple labels such as epic (or indeed
tragic) cease to be encompassing or definitive: the Metamorphoses provide
numerous and dazzlingly complex instances of generic cross-fertilization
and manipulation. Ovid himself, to begin with, looks to Sophocles and
Accius as his models and is engaged in the same exercise of transgeneric
appropriation that we witness in Thyestes only the direction is different.
It is reasonable to assume that Seneca capitalizes on that complexity, as he
alludes in his play both to an epic poem, which in turn alludes to tragic
27
An extensive analysis of the tradition is offered by Cazzaniga (1950); see now also Ciappi (1998).
84
models, and to those tragedies directly, thus creating an intricate web of

allusive relationships which resembles the stemma of a heavily contaminated textual tradition. Seneca imitates Ovid not only in the subject matter and expressive options of his work, but also in his intricate intertextual
protocols, which thrive on multiple references. Yet again, to extrapolate the
methodological and structural aspects of the phenomenon jeopardizes a
full appreciation of its core element: why does the tragedy of Thyestes begin
with an explicit and programmatic evocation of epic?
The Fury herself provides an interesting point of comparison. The Fury
who dominates the prologue is a direct descendant of the Virgilian Fury
responsible for bringing the second half of the Aeneid into existence. An
epic Fury, she coherently invokes epic models, although she is herself closely
connected in turn with a tragic precedent, Euripides Lyssa. It is almost as
if tragedy could not refer back directly to tragedy, but should necessarily
rely on an epic filter and thus testify to the impossibility of an immediate
connection, to a hiatus in the continuous tradition of tragic writing. Here
again Senecan tragedy highlights its posteriority, its position outside the
mainstream of tragic writing. But, again, it is not just any filter that is
interposed between Seneca and, say, Sophocles and Accius. It is specifically
the filter of Ovids Metamorphoses and peculiar aspects of Virgils Aeneid
the epic of violence and horrida bella rather more than the celebration of
heroic virtues and beliefs. This might very well be the most far-reaching
implication of Senecas choice. By giving pride of place to the Fury of
Aeneid 7 and to Ovids Procne he has not only demonstrated the shifting
boundaries of generic affiliations, thus reclaiming from the start his freedom
to experiment; he has also presented epic as the expression of nefas, as the
corrupted, doomed voice of history gone sour. It is not easy for History
directly to enter the hallowed halls of mythical tragedy, especially of a
tragedy written in a land, a language and a time other than that of Classical
Athens. But the shadow of history, with its pains and burdens, can reveal
itself, indirectly, in the peculiar selection of epic themes to which we are
treated in the prologue. Epic, we are warned, is not going to provide a
sound moral counterpoint to the towering horrors of tragedy. Epic is in
fact nothing less than the explicit justification invoked by the Fury: after
Virgils civil war and stories such as Ovids Procne, only further violence
and horror are conceivable, if anything maiore numero.
At the end of this book, once we can rely on a more substantial dossier, I
will deal more again with the issue of epic elements in Senecan tragedy, and
in particular with the hypothesis that the presence of epic might encourage
in the audience the form of critical spectatorship which Bertolt Brecht
A craftier Tereus
85
considers the defining characteristic of epic theatre a detached reflection

on the actions performed on stage which would certainly befit a tragic form
that is rich with philosophical intimations.28 For the moment I emphasize
the suggestion that epic acquires early on in Thyestes a function which does
not seem to encourage such a reflection on drama. Rather, epic appears from
the very beginning of Thyestes as the voice of destruction and violence, of
endless horrors and cruel, often excessive revenges. Other instances, from
other plays, will do little to soothe our sense of surprise and anxiety as we
contemplate the polluting force of infernal epic.
c r i m e , r i t ua l a n d p o e t ry
Ill play the cook
(Shakespeare, Titus Andronicus 5.2)
A playwright, a spectator and an actor, Atreus is also high priest of his own
rites, even a god himself. In act 4, where he appears in the messengers
detailed narrative, all these aspects are revealed together:29 in the climactic
moment of his nefas, standing alone in the recesses of his palace, Atreus
shines through in all his idiosyncratic depravity, as he undertakes what
looks like a Dionysiac sparagmos30 with due respect for all the procedures
of a proper Roman sacrifice.31 It is in this murder-as-sacrifice,32 already
present to a certain extent in Aeschylus Agamemnon33 and Accius Atreus,34
that Atreus achieves the paradoxical combination of holy and horrible,
fulfilment and uncleanness, sacrament and pollution which is at the heart
of the Dionysiac experience.35
The location for Atreus sacrifice is out of this world. Behind and below
the public quarters, the royal palace splits up (649: discedit) into many
rooms, until it reaches an arcana regio (a secret spot) located in its innermost recess (650: in imo . . . secessu). Only barren trees survive in this
28
29
30
31
32
33
34
35
Ch. 6, pp. 246ff.

On this scene see esp.: Burkert (1983) 1046; Petrone (1984); Picone (1984) 947; Tarrant (1985) 180;
Dupont (1995) 1936.
For child-murder in Dionysiac rituals see Dodds (1960) xix and n. 3; Burkert (1983) 105.
Human sacrifice is considered un-Roman (Livy 22.57.6), though occasionally attested (Suet. Aug.
15). On sacrifice in Rome see Scheid (1988); on its imagery, Huet (1994).
See Euripides Cyclops for the association of cannibalism and sacrifice: there the horror of cannibalism
is intensified by the careful, civilized sacrificial practice (Seaford (1984) 152).
See Ag. 10967 (where the children of Thyestes are called sphagas), with Zeitlin (1965).
Accius 2202 Ribbeck2 = 513 Dangel: concoquit | partem vapore flammae, veribus in foco | lacerta
tribuit. The sacrificial overtones of the description are revealed by the use of focus; see Dangel (1995)
2812, with further references.
Dodds (1960) xvii.
86
locus horridus,36 a natural enclosure in the bowels of a man-made building. This space is filled with the gifts (659: dona) which played crucial
roles in the familys history the trumpets, Oenomaus chariot, the broken
wheels a pictorial documentation of the ruling houses tormented past.
Spoils of war regularly celebrate the rulers achievements, and Seneca is
here alluding pointedly to Latinus regia in Aeneid 7.37 Differences, however, abound, because Seneca takes pains to emphasize how far we find
ourselves from the public rooms of the house, those where celebratory displays are to be expected. In this deep, dark, private domain, relics of the
past line up like memories in the recesses of the mind. Everything in there
is hidden, and frightening even to mention: quidquid audire est metus | illic
videtur (whatever is dreadful even to hear of, there is seen, 6701). An old
crowd freed from ancient graves (6712: errat antiquis vetus | emissa bustis
turba, the crowd of the long-since dead come out of their ancient tombs
and walk around), and creatures more monstrous than men have known
(673: maiora notis monstra) dwell in the grove, and they make the upper
chambers of the palace freeze with terror (677: attonita)38 as they wander at
night amidst the cries of the gods of death (668: feralis deos). Even the light
of day cannot restrain the horrors of the grove: terror is not yet allayed
by day; the grove is a night unto itself, and the horror of the underworld
reigns even at midday (6778: nec dies sedat metum: | nox propria luco est, et
superstitio inferum | in luce media regnat). It would be difficult to conceive
of a locus more evocative of the fundamental characteristics of the unconscious, indeed a place where nature, in all its dark, hostile power, survives
in spite of the elaborate superstructures that encircle and delimit its sway,
and where memories of the past roam unchecked as a constant source of
fear.39 Remarkably, it is in this place of passion, violence and memory that
knowledge elects to hide: from here the sons of Tantalus are used to enter
on their reign, here to seek help when their affairs are in distress or doubt
(6578: hinc auspicari regna Tantalidae solent, | hinc petere lassis rebus ac
dubiis opem).
Several analogies connect this grove to the one where Laius is evoked
from the dead in Oedipus, and the metapoetic dimension of that scene
36
37
38
39
On locus horridus and Senecas role in the development of the motive see Schiesaro (1985), with
further bibliography. A reference to this type of description is to be found in Letters to Lucilius 41.3,
a letter which will be discussed below, p. 127.
Smolenaars (1998).
Attonitus can be used in connection with poetic inspiration, and has a distinct Bacchic connotation:
above, p. 51, n. 60.
In Freuds celebrated simile, the mental realm of phantasy is described as a nature reserve where
useless and even harmful entities are allowed to grow unchecked (Introductory Lectures on PsychoAnalysis, Freud (191517) = SE xvi .372, with Orlandos seminal treatment (1993) 17).
A craftier Tereus
87
is aptly matched in Thyestes by the ceremonial aspect of the sacrifice, a

likely catalyst for metadramatic engagement.40 Atreus, sacerdos of his rites
(Thy. 691), is here also a vates, a magician-poet like Tiresias, who performs
arcane rituals and utters a sinister carmen: ipse funesta prece | letale carmen ore
violento canit (6912: he himself with a sinister prayer chants the death-song
with a violent voice) recalls carmenque magicum volvit et rabido minax |
decantat ore (Oed. 5612: over and over he unfolds a magic song, and,
with frenzied lips, he chants a charm).41 Significantly, the trembling of
the grove at Atreus magic intonations (Thy. 6969: lucus tremescit, tota
succusso solo | nutavit aula, dubia quo pondus daret | ac fluctuanti similis,
the grove begins to tremble; the whole palace sways as the earth quakes,
uncertain in which direction to fling its weight, and seems to waver) can
be compared both with the onset of horror at Tiresias words (Oed. 576),
and with the effects of Tantalus pollution in the prologue of Thyestes, yet
another passage with strong metapoetic implications: sentit introitus tuos |
domus et nefando tota contactu horruit (your house feels your entering and
has recoiled in horror from your unutterable contagion, Thy. 1034).42
Subversion of nature, too, occurs on both occasions. The Fury observes
that waters start to flow backwards (1078), and the messenger remarks
on the startling metamorphosis of wine into blood as Atreus performs his
ritual libation (7001).
Just as Tiresias evokes Laius truth from the underworld, and Lucans
Erictho turns to a corpse in her search for knowledge, it is in these hellish,
deadly abodes that Atreus conducts his painful negotiations between passion and knowledge, past and present, prediction and memory, poetry and
death. The texts insistence on the hostile, dark nature of the place is not
a symptom of rhetorical excess, for the sacrifice must be performed in an
uncanny underworldly location if it is to display fully its connection with
the forces of nefas. But several details in the description point to a specific
significance of the locus horridus. The adytum where the sacrifice takes place
is located deep inside the house, in a deep, secluded place (650: in imo . . .
secessu), a cavern (681: specus) covered by an ancient grove (651: vetustum . . .
nemus), overshadowed by dark vegetation. The Freudian symbolic geography of sex is transparent.43 More importantly, it is very significant in
40
41
43
See Hornby (1986) 4966. A further metadramatic aspect of Atreus behaviour can be gleaned in
the details of how he cooks the boys entrails (7657). Cooking is often characterized, in comedy,
as an activity with metadramatic connotations. On cooks in comedy see Dohm (1964); on the
metadramatic implications they hold in Plautus see Gowers (1993) 50108.
42 See ch. 1, p. 38.
On this passage, see above, pp. 11ff.
S. Freud, Fragments of an Analysis of a Case of Hysteria (Dora) (Freud (1905) = SE v i i.94, 99).
See The Interpretation of Dreams (Freud (1900) = SE v .348, 355); On Dreams (Freud (1901) = SE
v .684); Introductory Lectures on Psycho-Analysis (Freud (191617) = SE xv .15860).
88
the context of the tragedys plot. Atreus crucial concern regards the true
paternity of his children, which, he would like to believe, can be ascertained
by the observation of Thyestes reaction to the death of his own offspring,
and an inspection of their entrails. Thus Atreus descent to the womb-like
arcana . . . regio (secret spot, 650) beneath the royal palace becomes a fitting symbolic exploration of Aeropes entrails, where the truth about his
dubius . . . sanguis conceivably resides. The careful investigation of the boys
entrails (7558) is a mise en abyme of the only (impossible) inspection
which could actually assuage Atreus doubts, that of his adulterous, even
incestuous, wife. It is in her womb that the original nefas has taken place,
the confusion of generations feared by Phaedras nurse: are you preparing
to mix the fathers wedding-bed with the sons, and to welcome in your
impious womb a mixed-up progeny? (Phaed. 1712: miscere thalamos patris
et gnati apparas | uteroque prolem capere confusam impio?).44 The whole
structure of the play, moving from one level of the action to a deeper, inner
one, and culminating in Atreus extispicium, deep in the womb-like recesses
of the palace, dramatizes this descent into the secrets of conception.45
This symbolism is much developed by Shakespeare, who stages the worst
horrors of Titus Andronicus in a dark forest, A barren detested vale . . .
forlorn and lean (2.3.923).46 Its central feature is an abhorred pit (98)
with a strong Senecan colouring (2.3.98104):47
And when they showd me this abhorred pit,
They told me, here, at dead time of the night,
A thousand fiends, a thousand hissing snakes,
Ten thousand swelling toads, as many urchins,
Would make such fearful and confused cries,
As any mortal body hearing it
Should straight fall mad, or else die suddenly.
The pit evokes King Lears obsessed description of female genitals (King
Lear 4.6.1238):48
Down from the waist they are Centaurs,
Though women all above:
But to the girdle do the Gods inherit,
44
45
46
47
48
The womb of Thyestes incestuous daughter is also a receptacle of nefas (Ag. 31): coacta fatis gnata
fert utero gravi | me patre dignum (334).
See Irigarays analysis ((1985) 243364) of the cave in Platos Republic as a womb-like source of all
representations (Robin (1993) 111), with Leonard (1999).
A locus of instinctual, evil and fatal force (Marienstras (1985) 45).
By a sort of metonymy, the pit in Shakespeare also comes to symbolize (metadramatically) the
classical underworld and its hellish sources of inspiration; see Tricomi (1974) 18.
Tricomi (1974) 18 n. 3; Willbern (1978) offers a psychoanalytic interpretation of the pit in Titus.
A craftier Tereus
89
Beneath is all the fiends: theres hell,

Theres darkness,
There is the sulphurous pit burning, scalding,
Stench, consumption; fie, fie, fie! Pah, pah.
But Senecas gendered landscape of the unconscious implies ramifications

which go well beyond Shakespeares important, if somewhat transparent,
imagery. Atreus arcana . . . regio, as we have seen, finds a close parallel in
the lucus ilicibus niger (530) at the very heart of Oedipus. There the imagery
is even more heavily loaded with sexual connotations. Not only do we
find a forest which is permanently kept in the dark, but the emergence of
the hellish creatures from the depth of the earth is described as a painful
birth, and a monstrous one since dead creatures are brought to the light
(57281):49
rata verba fudi: rumpitur caecum chaos
iterque populis Ditis ad superos datur.
subsedit omnis silva et erexit comas,
duxere rimas robora et totum nemus
concussit horror, terra se retro dedit
gemuitque penitus: sive temptari abditum
Acheron profundum mente non aequa tulit,
sive ipsa tellus, ut daret functis viam,
compage rupta sonuit, aut ira furens
triceps catenas Cerberus movit graves.
I have uttered prevailing words; blind Chaos is burst open, and for the people of
Dis a way is given to those living on earth. The whole forest shrank down, then
raised its foliage, the oaks were split and the whole grove shook with horror; earth
withdrew and groaned deep inside: whether Acheron did not tolerate an assault
against its hidden depths, or the earth itself broke down its barriers in a thunder to
give way to the dead; or three-headed Cerberus furious with rage shook his heavy
chains.
Poetry comes to light through a painful birthing process which gives

shape to the passions residing in the underworld, and is thus associated
with the fear-inspiring secrets of the female body. The story of Procne and
Philomela also indicated a strong connection between poetic inspiration
and womanhood. A raped and silenced Philomela had found in her thirst
for revenge the strength and ingenuity for writing up Tereus crimes.
Philomela turns her fury into the cunning plot which takes her husband
in. As she kills Itys, she overcomes her maternal function and perversely
forces on Tereus an impossible birth (Met. 6.6635):
49
Note that the sacerdos begins his rites by excavating the ground: tum effossa tellus (550).
90

et modo, si posset, reserato pectore diras
egerere inde dapes inmersaque viscera gestit,
flet modo seque vocat bustum miserabile nati.
Gladly, if he could, he would want to open his breast and eject that terrible feast,
the entrails immersed into his own, and now he cries and calls himself his sons
miserable tomb.50
Thyestes, too, will be forced to a perverse delivery as he vomits his

own children.51 But Atreus, acting as the main purveyor and creator of
poetic plots and explicitly acknowledging his identification with Procne
and Philomela, also inevitably erodes the boundaries of his masculinity.
Nefas and its poetry are described as they emerge from the feminized entrails
of the earth. Medea another avenger acting as a playwright will put it
with epigrammatic clarity: parta ultio est: | peperi (Med . 256: vengeance
is born: I have given birth).
In the merging of frenzy and control, of ira and ordo, which characterize
the sacrifice, Atreus reveals once more his deeply metadramatic role. As a
playwright, he carefully devised and executed a complex plot; as a sacerdos,
he once again plays both instigator and executor. In both cases, he fuses
inspiration and techne in the heady cocktail which provokes awed pleasure, the sacred rituality of the priest in communication with the divine,
the frenzied poet able to express divine enthousiasmos in refined, regulated
language.
The messenger himself seems to marvel at Atreus deliberate observance
of ritual, which is reported after the physical setting of the scene has been
engagingly described (Thy. 68290):
quo postquam furens
intravit Atreus liberos fratris trahens,
ornantur arae quis queat digne eloqui?
post terga iuvenum nobiles revocat manus
et maesta vitta capita purpurea ligat;
non tura desunt, non sacer Bacchi liquor
tangensque salsa victimam culter mola.
servatur omnis ordo, ne tantum nefas
non rite fiat.
After Atreus entered there in a frenzy, dragging his brothers children, the altars
are decorated who has adequate words for this? Behind their backs he forces
the noble hands of the youths, and their unhappy heads he secures with a purple
band. Nor is incense missing, nor the holy liquor of Bacchus, and the knife, which
50
Authors translation.
51
Littlewood (1997) 77.
A craftier Tereus
91
touches the victim with salted meal. Every detail is preserved, lest such a crime
take place in breach of ritual.
The scene we are invited to imagine is eerily compelling: here is the wild
tyrant dominated by furor, finally alone with his designated victims, with
no one in sight, who chooses to perform his vengeance with carefully chosen
sacrificial gestures, fussing about minutiae.
Atreus himself seems to be aware of the fact that he has been performing a real sacrifice when he later describes his actions to Thyestes: with
deep-driven sword I wounded them; I slaughtered them in front of the
altar; I appeased the sacred fires, offering their death as a vow (10579:
ferro vulnera impresso dedi, | cecidi ad aras, caede votiva focos | placavi).
These words resonate alongside Atreus question to the satelles in act 2:
tell by what means I may bring ruin on his wicked head (244: profare,
dirum qua caput mactem via).52 Although the meaning of macto in this
particular context must be closer to afflict than to sacrifice, the technical
use of the verb cannot surely be too far away, and a paraphrase such as
how I might offer sacrifice in such a way as to torment Thyestes most
aptly conveys the implications of the line.53 The messenger, too, resorts
to specific sacrificial language: he wonders whom he should first sacrifice
to himself, whom he should slaughter second (71314: quem prius mactet
sibi | dubitat, secunda deinde quem caede immolet).54 The nefas must be
performed, of all things, rite, comme il faut. The sacrifice is divided into its
customary phases, praefatio, immolatio and litatio, to be followed later by
the epulum. All the most important aspects of the ritual are mentioned in
the narrative: the altar is decorated (684), the victims head is bound with a
vitta (686), wine and incense are used (687), as is salsa mola (688). No part
of the procedure must be skipped (695: nulla pars sacri perit); order must
triumph (689: servatur omnis ordo); ordinare is indeed a source of delight
(71516: nec interest sed dubitat et saevum scelus | iuvat ordinare, it does
not matter, but still he hesitates, and has pleasure in ordering his savage
crime).
Although by far the most explicit, Thyestes is not the only Senecan tragedy
to represent murder in the guise of sacrifice.55 Hercules furens provides an
eloquent example, by way of a noticeable departure from its Euripidean
52
53
54
55
Dirum casts Thyestes in the role of a cursed victim; the expression mactare (caput) is indeed used
in sacrificial contexts see, for example, Livy 21.45.8, Sil. 5.653.
See Putnam (1995) 275.
Traina (1981) neatly sums up the case for understanding mactet sibi rather than sibi | dubitat.
On this and other perverted sacrifices in Seneca see especially Petrone (1984) 403 and Dupont
(1995) 189204.
92
model. Hercules frenzied slaughter takes place in the context of a sacrifice

he is offering to the gods, which cruelly degenerates into the killing of
Hercules own wife and children (Her. F. 8989: nunc sacra patri victor et superis feram | caesisque meritas victimis aras colam, now I shall make offerings
for my victory to my father and the gods, and honour their altars as they deserve with sacrificed victims). Amphitryon connects sacrifice and murder
by explicitly addressing his son with these charged words: nondum litasti,
nate: consumma sacrum (you have not yet made full offering, son; complete
the sacrifice, 1039). In Troades the Greeks present the deaths of Astyanax and
Polyxena as a required sacrificial offering to the dead Achilles: the youths
blood is needed to placate his rage and allow safe sailing from Troy. The
conflicting points of view in this tragedy, with the Trojans actively questioning, emotionally if not ideologically, the actions of the Greeks, problematize the equivalence between sacrifice and murder. Hippolytus death in
Phaedra, too, has distinctly sacrificial overtones. Theseus slowly and
painfully reconstructs his sons corpse in a fashion reminiscent of similar rituals after sacrificial slaughters, when the body of the slain animal is
rearranged in its proper order. Such belated pietas, incidentally, is conspicuously absent from Thyestes.56
The pervasiveness of sacrificial motives in Senecan drama invites the
audience to reflect on a religious problematic which might well have been
thought of as anachronistic and misplaced, once tragedy had severed its
connection with its traditional Greek roots. Sacrifice occupies a central
role in Greek tragedy. In Euripides especially, perverted human sacrifice,
such as the slaying of Pentheus or of Heracles children, marks a larger
social and religious crisis,57 as, in different ways, does Senecas exploitation of this particular motif. The analogy, however, stops here. Euripidean
sacrifices ostensibly attempt to heal the wound they inflict: the poet ultimately reconstructs and reaffirms tradition through the cathartic power
of sacrifice.58 Thyestes makes no overt attempt at reparation, symbolic or
otherwise, for the extraordinary disruption signalled by Atreus perverted
sacrifice. The prominence of sacrifice in Thyestes thus seems to correspond
to yet another aspect of Senecas intertextual and metaliterary strategy. I remarked in chapter two that framing the potentially self-enclosed structure
of a traditional play that is redolent of Greek forms affords a reflection
on the viability of that particular type of tragedy. Similarly, giving sacrifice
the same structural importance it enjoyed in Euripides, but depriving it
56
57
Valuable observations in Petrone (1984) 314. Oedipus displays sacrifice in two central scenes, but
they are not directly connected with murder (291402 and 530658).
58 Foley (1985).
Zeitlin (1965).
A craftier Tereus
93
of any constructive, forward-looking value, betrays the awareness that this

particular escape from nefas, too, is gone for ever.59
The Roman model for the association between sacrifice and murder
is the final scene of Virgils Aeneid. On the verge of accepting Turnus
supplication, Aeneas is struck by the sight of Pallas baldric, and buries his
sword in the neck of his enemy (12.9459):
ille, oculis postquam saevi monimenta doloris
exuviasque hausit, furiis accensus et ira
terribilis: tune hinc spoliis indute meorum
eripiare mihi? Pallas te hoc vulnere, Pallas
immolat et poenam scelerato ex sanguine sumit.
[Aeneas] feasted his eyes on the sight of this spoil, this reminder of his own wild
grief, then, burning with mad passion and terrible in his wrath, he cried: Are you
to escape me now, wearing the spoils stripped from the body of those I loved? By
this wound which I now give, it is Pallas who makes sacrifice of you. It is Pallas
who exacts the penalty in your guilty blood.
Seneca amplifies the sacrificial protocol latent in this scene (though

Aeneas immolat is telling enough) and makes it register at the forefront of
his account. Just as Aeneid 7 had been prominent in the inaugural movements of the tragedy, the end of the poem is powerfully evoked in the
climactic scene of Thyestes, in the action that effectively brings the plot
to a close, if not to a closure. The end of Aeneid 12 seems to have found
in Seneca a reader devoted to the point of obsession, who confronts that
scene and its disturbing implications time and again in his tragedies.60 The
comparison with the Aeneid yields important insights. At the conclusion
of the poem we face an almost too neatly schematic dramatization of Rene
Girards theory of the sacrificial crisis ,61 the breakdown of the foedus
which Aeneas and Latinus had finally reached (Aen. 12.161215). Aeneas
sacrificial murder of Turnus restores the violated order, but only by means
of equally transgressive violence: the reparation is far from satisfactory, and
far from final.
There is (fortunately) no need to rehearse here the vast body of criticism
on the final scene of the Aeneid; suffice it to say that the comparison is instructive. In its basic outline, Thyestes offers a similar scenario. Thyestes has
caused the violent disruption of order which Atreus concisely portrays in
the statement fas . . . omne ruptum (breaking every law, 179). His sacrificial
59
60
A similarly deflating attitude can be discerned in Euripides Cyclops, where Ulysses companions are
slaughtered with ritual accuracy; Seaford (1984) 1513 and 1801.
61 Hardie (1993) 21.
As Putnam (1995) 246 rightly remarks.
94
killing of Thyestes children fulfils the need for reparation and restoration,
so much so that, in Atreus words, it actually restores a status quo ante which
might have been thought of as irrevocably lost. Whatever our assessment of
Aeneas behaviour, his explanation for the killing of Turnus is unequivocal:
Pallas must be avenged.62 This intertext thus emphasizes once again Atreus
deep conviction that he has been wronged and is seeking a justifiable retribution. His sacrificial killing is a direct response to Thyestes violation of
fas in the seduction of his brothers wife. The sacrifice would thus seem to
heal the wound that Thyestes inflicted and restore the order that he upset.
Atreus retribution is especially apt in the light of the firmly held belief that
incest and cannibalism are homologous acts.63 Thus Thyestes intercourse
with his sister-in-law Aerope must be expiated with a similarly perverse and
unnatural action: he will be forced to eat his own children.
As he implicitly identifies cannibalism and incest, Atreus displays in all
its upsetting force the working of his peculiar form of logic. Incest pollutes
the body with the seed of a close relation; eating ones own children is a
similar form of unacceptable ingestion. Atreus identifies behind these two
very different gestures a common element which becomes central to his
thinking and on which he bases his course of action. Like Platos tyrant,
he overruns the boundaries which keep distinct facts and actions separate
and follows a form of logic which is akin to the logic of the unconscious:
analogies overcome differences and precipitate the identification of disparate actions. This form of generalizing thought was originally considered
typical of schizophrenia, but it is one of the greatest achievements of postFreudian thought to have realized that this strange logic, where symmetry
replaces the rigid conventions of Aristotelian thought, is actually an ineliminable component of the mind, given free rein in the workings of the
unconscious but normally kept at bay during conscious activity. It is interesting in this connection to look at an observation that Freud makes in
Psychopathic Characters on the Stage:64
In general, it may perhaps be said that the neurotic instability of the public and
the dramatists skill in avoiding resistances and offering fore-pleasures can alone
determine the limits set upon the employment of abnormal characters on the
stage.
62
63
64
Note that Aeneas had already ordered a human sacrifice immediately after Pallas death: Virg. Aen.
10.51720 and 11.812.
Pl. Resp. 571cd with Parker (1983) 98 and 326 and Burkert (1983) 104. On the connection between
sex and eating see Kilgour (1990).
Psychopathic Characters on the Stage (Freud (1942, but written 1905 or 1906) = SE v i i.30510
at 310).
A craftier Tereus
95
and rephrase it in the light of the observations above. Atreus abnormality

does appeal to the public precisely because it displays a form of logic and
behaviour which does away with the restrictions of adult Aristotelian
logic. He does not appeal so much to our neurotic instability as to the
ineliminable part of our thinking which chafes at the unnatural imposition
of criteria such as non-contradiction, asymmetry, hierarchy. We do not have
to be closet cannibals to be taken in by Atreus extraordinary flights of logic,
which are part and parcel, of course, of his being a poet.
In Thyestes the disruption of bonds is rooted in the alternae vices of reality.
These events, left out of the plays direct dramatic focus, and only alluded
to in more or less detail, are surely real in that they are subtracted from the
stage manipulation we witness. While we are asked to focus only on the
perversion of Atreus deeds, we might well wonder whether the reality
he is trying to control and alter is any more acceptable or normal than
his striking revenge against it. Interestingly enough, for all the emphasis
that the reversal of the sun attracts in Thyestes, it lacks the prominence
it had received in earlier texts, where it was credited with a fundamental
cosmogonic function.65 According to part of the earlier tradition, for instance, Atreus murder caused the sun to change its path once and for all.
In this play, characterized by the general absence of ethical certainties, the
sun does indeed show its disgust at the murder and abruptly disappears in
the middle of the day, only to resume its regular route the following day.
Atreus deeds, extraordinary as they may be, can only upset the order of
nature for so long, because after all they, too, are part of nature. What we
call normal and abnormal are generalizations that might well stand in
relative contrast to each other, but offer very little in the way of absolute
certainty.
Just as there is no absolute limit to ethical disruption, there can hardly
be a well-defined sense of closure and ending to human revenge. By
killing Thyestes sons instead of their father, Atreus follows a sacrificial
protocol, but he also introduces an element which is consistent with his
characterization in this play. Atreus choice obeys a homeopathic principle that is perfectly understandable within the norms of sacrifice.66 If
the personal wound that most directly aggrieves him is the doubt cast
by Thyestes relationship with Aerope over the paternity of Agamemnon
and Menelaus, then killing Thyestes sons repeats and returns the same
wound, and fits in with the plays insistence on the notion that horror is
65
66
Burkert (1983) 105 and n. 13.

Medea, too, chooses to kill her own offspring rather than Jason himself. See Girard (1972) 24 on
the issue in general.
96
self-perpetuating.67 Epicurus famously maintained that pain is either very

intense, but brief, or long, but then bearable. Thyestes seems to show that
evil can be both extraordinarily intense and potentially endless. Sparing
Thyestes not only spares Atreus double, but ensures the continuation of
their duel in the family saga, ensures that revenge can be exacted not just
once, but many times over. Thyestes invocation to the gods, to a principle
of absolute justice that would also entail a final moment of judgement, is
rejected in favour of a new phase of human action. Leaving the conflict
open, of course, exposes Atreus himself to possible future retribution, and
his lot will not be spared. Closure cannot possibly appeal to the tyrant who
had chided his counsellor for the simplistic suggestion that Thyestes be
quickly dispatched (2468). Nothing is in fact more alien to Atreus than
his feigned willingness to forgive and forget: let all our anger pass away.
From this day, let ties of blood and love be honoured, and accursed hatred
disappear from our hearts.68 An unequivocal rejection of finality resurfaces
towards the end of the play, when Atreus contrasts process and result: I do
not want to see him miserable, but his becoming so (907: miserum videre
nolo, sed dum fit miser). Killing Thyestes children will also guarantee the
additional pleasure of watching him watch their death, or at least their
corpses: it is a pleasure to note, when he sees his childrens heads, how his
complexion changes, what words his first grief pours forth (9035: libet
videre, capita natorum intuens, | quos det colores, verba quae primus dolor |
effundat . . . ). Atreus has already built repetition into the structure of his
revenge, killing the children himself a first time, and then forcing Thyestes
to kill them, as it were, all over again; and the text reinforces this repetition
by allowing the audience to hear twice, at least in part, the narrative of the
murder, first from the messenger in act 4, and then, more succinctly, from
Atreus himself in act 5 (105765).
However, Atreus obsession with the repetition of revenge, his refusal of
closure, will also prove to be his undoing. In the final line of the tragedy he
gloats that Thyestes punishment is not a hope for the future but a fact already accomplished (1112: te puniendum liberis trado tuis for punishment,
67
68
In this respect Atreus is the victim of his own logic; see above, pp. 117ff.
50911: quidquid irarum fuit | transierit; ex hoc sanguis ac pietas die | colantur, animis odia damnata
excidant. This statement is yet another one of Atreus double entendres: he surely means it when
he says sanguis colatur, but not in the way the sentence seems to convey. His own respect for (his
own) blood, together with the worry about the dubius sanguis that Thyestes has caused, is precisely
the source of his ira. Another possible ironic connotation is detected by Tarrant ((1985) 164): [t]he
sacral overtones of colatur (be worshiped) may also carry ironic force, since Atreus will in fact
turn his bloodshed into a ritual act. Tarrant also notes that sanguis recalls sperat ira sanguinem at
line 504.
A craftier Tereus
97
I deliver you to your children picks up premor . . . natis I am weighed

upon by my sons of 1051), but he also foreshadows the reversal of fortunes
that his offspring will endure. Indeed te puniendum liberis trado tuis could
be applied to Atreus himself, since Thyestes revenge will be accomplished
with the killing of Agamemnon. This following phase of the family history shows the force of repetition: once again incest (between Thyestes and
Pelopia) will lead to murder.69
The sacrificial proceedings of act 4 encapsulate the core motives of the
play and its main character, as well as, arguably, the reason for its power and
appeal. Atreus had already displayed in act 2 the strength of his Dionysiac
inspiration and had shown in act 3 how cunningly and masterfully he could
perform in order to achieve his goals. Here we finally realize that his project
goes beyond the specific objective of revenge. His ambition is effectively to
create a new world order (hence the ritual importance of ordo) in which the
traditional gods lose their power, accepted political philosophy is shown to
be useless and void, and even the traditional categories of order and frenzy
can be deconstructed and redefined. Atreus sacrifice is the most Dionysiac
of rituals: the slaughter and cooking of victims. It is a ritual which uncannily
represents both the establishment of civilization and a throwback to barbarity. Atreus does not necessarily portray Nero on stage, nor indeed should
his behaviour inevitably be collated with the anedoctal evidence of extravagant cruelty which peppers Suetonius Lives. The ritualization of violence
encoded in the murder-as-sacrifice shows that Atreus is the incarnation of
imperial power at a much more radical and discomforting level.70 Almost
from its inception that power had played an elaborate and risky game by
suggesting, increasingly, the religious dimension of the emperor. First as a
sacerdos, then as a divus, the emperor of Rome had (even in the West) relied
more on the accretion of power and mystique than a careful exploitation
of religious symbols would allow. Atreus shows the game for what it is
he is god to himself, and god to his subjects. His power makes him so.
Dionysus, too, had become under Augustus an attractive symbol of power
and regeneration, not to mention a useful figure for summoning the awed
memory of Alexander. This, too, is a symbol which Atreus transforms into
reality. In Bacchae the cunning god had shown the inevitable limitations
69
70
On this connection see, in general, Irwin (1975). For the mythical plot see Hyg. Fab. 87 and 88;
the latter offers a complicated and largely unparalleled version of the plot which, uniquely, offers a
complete closure: Pelopia commits suicide; Aegisthus kills Atreus; Aegisthus in regnum avitum redit
with his father Thyestes.
The ritualization of violence in Titus Andronicus has more specific political ramifications; Bate (1995)
234.
98
of an earthly power based on the limited intellectual and imaginative resources of a Pentheus. In Thyestes Atreus shows that a ruler can appropriate
the animal, wild strength of Dionysiac inspiration and use it for his own
purposes in a seductively creative form. We are reminded of the revolution
which Lucan had encapsulated in unsurpassable, if wholly unappreciative,
terms at the very outset of his Bellum Civile (1.2): ius . . . datum sceleri
legality conferred on crime. Thyestes makes us wonder whether ius and
scelus can be so neatly distinguished and set against each other.
t h e lo g i c o f c r i m e
Videturne summa inprobitate usus non sine summa esse ratione? (Cicero, De
natura deorum 3.69)
i
Atreus extraordinary power explicates itself on several levels. His dramatic,
larger-than-life personality has many different sides, from wild aggression
to comic penchant for punning; throughout, he is obsessed with ever bigger
pursuits, transcending, by his own admission, the boundaries of mankind
(2678) and aspiring to reach or even surpass the power of the gods. It is
the gods, indeed, who constitute Atreus ultimate point of comparison
his power over men is not open to discussion, and his doubts concern only
how, not whether, he will defeat his brother. His nefas, he believes, will be
such that even the gods will have to take notice and flee in horror (2656;
888). His nefas, he finally gloats, has lifted him to the stars (8856: aequalis
astris gradior). In this exhilarating declaration of success Atreus combines
the nefas of gigantomachy71 and the proud claims of a cultural hero such
as Lucretius Epicurus, whose intellectual victory managed to expand the
boundaries of human knowledge and exalt us mortals as high as heaven
(De rerum natura 1.79: nos exaequat victoria caelo).
As we have already seen repeatedly, it is unhelpful to import into the
complex texture of the tragedy a system of moral categories that has been
developed out of context, as the specifics of Atreus case are bound to be
bulldozed in the discussion of general principles.72 It is far more important
71
72
Interestingly enough, the motif is explicitly mentioned by Thyestes at 1084 among the guilty excesses
that Jupiter has quashed in the past. Any such divine retribution of Atreus nefas is conspicuously
absent from Senecas play. A cursory anticipation is also in the choruss words at 806.
An important analysis of Atreus is offered by Knoche (1941), who stresses his irrational and violent
features, his spirit of anti-nature, the irredeemable madness rooted in his evilness, and connects
them to Senecas own experience under Caligulas reign of terror. The date of publication of the
article, of course, is not irrelevant. See also Lefèvre (1985).
A craftier Tereus
99
to understand the means by which Atreus overpowers Thyestes and in the

process becomes the emotional fulcrum of the play. Atreus engrossing energy derives from his superior intellectual ability to manipulate the vigour
of his passions. His most powerful weapons are, firstly, the method he
brings to his madness: the epigraph of this section quotes Cottas use of
Atreus (and Medea), in De natura deorum 3.689, as evidence that ratio is
not a generous gift of the gods, because it can be turned to negative uses:
videturne summa inprobitate usus non sine summa esse ratione? (does he not
appear to have acted with the highest degree of criminality and at the same
time the highest degree of rationality?). Secondly, Atreus is able to use
language creatively (and passionately) as a weapon to overcome Thyestes
fatally narrow literalness. Thirdly, he displays an instinctive comprehension
of human nature, and an ability to foresee and manipulate his opponents
reaction. Atreus is not a madman, of course. But he shows that there is
much beyond Thyestes unbending logic and referential use of language
that the passions associated with primal instincts and desires open up different forms of logic and expression. These may abandon the reassuring
certainties of non-contradiction, but prove invaluable in the execution of
Atreus plot.
In the chthonic bowels of the palace Atreus chooses to enquire the fates
(Thy. 757: fata inspicit) by looking at the entrails of his victims. The result
pleases him (759: hostiae placuere). We have already been offered an image
of Atreus as a hunter of traces. In act 3, as he is finally ready to meet his
brother, Atreus is certain that his plot is close to completion. Thyestes, in
accepting to come back to Argos, has fallen into the trap: the prey, Atreus
gloats, is firmly bound in the nets he has prepared (491: plagis tenetur clausa
dispositis fera). The hunting imagery is extended in the image that Atreus
offers of himself immediately thereafter (496505):
vix tempero animo, vix dolor frenos capit.
sic, cum feras vestigat et longo sagax
loro tenetur Umber ac presso vias
scrutatur ore, dum procul lento suem
odore sentit, paret et tacito locum
rostro pererrat; praeda cum propior fuit,
cervice tota pugnat et gemitu vocat
dominum morantem seque retinenti eripit:
cum sperat ira sanguinem, nescit tegi
tamen tegatur.
I can scarcely contain my heart; hardly can my grief tolerate restraint. Thus a keen
Umbrian dog, when he is kept on a long leash in pursuit of wild animals, and with
lowered muzzle sniffs the traces, while through its lasting scent he perceives the
100
boar afar, he obeys and with silent tongue explores the place; but when the prey is
closer, he fights with all his head, and moans and begs the master holding him and
breaks away from his restraint: when his rage scents blood it cannot be concealed;
yet it must.
This extended simile has often prompted reservations in critics who

either fault its epic tone, dissonant in a dramatic context, or criticize its descriptive excesses.73 It would be rash, however, to underestimate the importance of this detailed passage only because similar descriptions are offered
by Ennius,74 Virgil and Ovid. Indeed, a comparison with those influential models highlights once again the specific function of these lines in the
context of the play, and offers a vivid and explicit representation of a crucial aspect of Atreus character: his passion for, and success in, attaining
knowledge and using it for his purposes.
The Umbrian-dog simile effectively depicts Atreus intents and his
heuristic methods. The dog possesses an instinctual drive which can be
compared to Atreus own furor and ira, but this is displayed only after a
diligent enquiry has enabled it to discover the prey, and should remain
subordinated to a strategy of dissimulation which can guarantee the successful outcome of the hunt (5045: . . . nescit tegi; | tamen tegatur). In this
respect Senecas accurate choice of words to describe the search (vestigat,
sagax, scrutatur, sentit) begs comparison not with generic hunting scenes,
but, more specifically, with Lucretius simile in book 1 of De rerum natura
(4048):
namque canes ut montivagae persaepe ferarum
naribus inveniunt intectas fronde quietes,
cum semel institerunt vestigia certa viai,
sic alid ex alio per te tute ipse videre
talibus in rebus poteris caecasque latebras
insinuare omnis et verum protrahere inde.
for as dogs, thanks to their nose, often find the resting place of a mountain prey,
covered with leaves, once they have trodden on certain traces, thus in such matters
you will be able to see by yourself one thing after another, and to penetrate all the
secret recesses and extract from them the truth.
73
74
For a reasoned defence and an analysis of possible models see Tarrant (1985) 162.
The use of sagax is a direct if limited point of contact with Ennius, 3324 Skutsch (3402
Vahlen): veluti, [si] quando vinclis venatica velox | apta dolet si forte <feras> ex nare sagaci | sensit,
voce sua nictit ululatque ibi acute. Cf. Hom. Il. 22.18893: but swift Achilles, relentlessly pressing
on, kept on after Hektor. And as when a dog startles a fawn in the mountains and chases it out of
its lair, through hollows and glades, and even if the fawn takes to cover and crouches in a thicket,
the dog tracks it () and runs it down even so Hektor could not get away from the
swift-footed Peleion. See also Varius, De morte, fr. 4 Courtney, though the context may have been
more ominous (Courtney (1993) 274).
A craftier Tereus
101
Verbal correspondences are significant even if it is not necessary to postulate

a direct correlation: Seneca speaks appropriately of vestigia, the fundamental
object of venatic enquiry75 and expands the description of the dogs careful
exploration (pererrat, which stresses the accuracy and scope of the search,
conveys some of the force of montivagae). Comparison with the models
strengthens the point, since neither Virgil nor Ovid devotes comparable
attention to this aspect of the search; they focus more on the final outcome
of the hunt.
This simile sets the stage for the more intriguing notion that the sacrifice that Atreus performs is also an extispicium, a procedure meant to yield
important information. The two details together open an interesting vista
on a very important aspect of the plot which only occasionally surfaces in
the text, but at all times stays firmly at the back of Atreus mind. Atreus
is uniquely able to combine the forceful determination of his willpower
an arcane, prerational inner strength with the seemingly endless resourcefulness of his intellectual gifts. He is not only determined to take as cruel
a revenge as possible on Thyestes for forcing him out of power and his
furor will help him to do precisely that but also concerned with a rational (if obsessive) doubt which demands to be assuaged, in principle, by
careful investigation, namely whether his children are actually his own or
the illegitimate offspring of Thyestes adulterous relationship with Aerope.
The characterization of Atreus as an expert hunter and decoder of vestigia, I
would argue, is best appreciated in the context of this investigation, and not
only, as the simile suggests, in the context of his ability to deceive Thyestes
in the rest of the play.
Atreus states his concern about the paternity of his children early in the
play (2204):76
fas est in illo quidquid in fratre est nefas.
quid enim reliquit crimine intactum aut ubi
sceleri pepercit? coniugem stupro abstulit
regnumque furto: specimen antiquum imperi
fraude est adeptus, fraude turbavit domum.77
Whatever is wrong to do to a brother is right to do to him. What crime has he left
untouched, or when has he ever recoiled from a sin? My wife he took away with his
75
76
77
On the so called venatic paradigm see Ginzburg (1992) and Cave (1988) 2504; see later, p. 135, for
the important presence of venatic metaphors in Euripides Bacchae.
In later versions of the play the presence of the illegitimate sons becomes a central motif; for a survey,
see Rossi (1989).
See Accius 205 Ribbeck2 = Dangel 33: qui non sat habuit coniugem inlexe in stuprum, with Lana
(195859) 318. Turbare domum suitably recalls Aeschylus
(Supp. 225).
102
debauchery; he stole my kingdom; the ancient token of our dynasty he gained by

fraud, by fraud unsettled our house.
Because of the stuprum, Atreus house has been contaminated, his (positive) certainties shattered: my wife seduced, the solidity of my power is
shattered, my house is polluted, my offspring uncertain nothing is certain save my brothers enmity (23941: corrupta coniunx, imperi quassa est
fides, | domus aegra, dubius sanguis est et certi nihil | nisi frater hostis). The
revenge-plot aimed at punishing Thyestes thus doubles also as a trial which
will try to ascertain the childrens real lineage and soothe Atreus torment
about his dubius sanguis, an expression that condenses a crucial concern of
Roman culture, that of turbatio sanguinis.78 Accius Atreus had expressed
the problem lucidly (2068 Ribbeck2 = 346 Dangel):
quod re in summa summum esse arbitror
periclum matres conquinari regias,
contaminari stirpem, admisceri genus.79
This I believe to be the greatest danger in matters of high state: when royal mothers
are polluted, the family is defiled, the lineage mixed up.
At the conclusion of act 2 Atreus shares with the satelles the plan he has
devised in order to test Agamemnons and Menelaus loyalty and, by implication, their paternity. He intends to make them accomplices in his revenge
plot against Thyestes: a sign of hesitation on their part would reveal that
Thyestes, not Atreus, is in fact their father (Thy. 32530):
consili Agamemnon mei
sciens minister fiat et fratri sciens
Menelaus adsit. prolis incertae fides
ex hoc petatur scelere: si bella abnuunt
et gerere nolunt odia, si patruum vocant,
pater est.
Let Agamemnon be aware of my plot and carry it through, and let Menelaus stand
by his brother, fully aware, too. Let this crime test how true are my uncertain
offspring: if they refuse to fight and dont want to wage the war of hate, if they call
him uncle, he is their father.
In the end, however, Atreus will abandon the plan to make his children
aware of his intentions out of fear that they might unwillingly reveal what
he is plotting (3313). At the end of the play Atreus declares himself satisfied
78
79
The term is used by Ulpian, dig. 3.2.11.1. On dubius sanguis see especially Guastella (1988) 6872.
For Senecas Atreus, the greatest fault vis-à-vis this greatest danger would be the absence or lateness
of a suitable reaction.
A craftier Tereus
103
that his children are really his (1098), as if the vaticinium he has performed
on the corpses of his victims had actually yielded solid results. But there is
also a different aspect worth noticing here: Atreus adherence to logical rules
of enquiry, as highlighted by the Umbrian-dog simile, is always tempered
by his reliance on a form of symmetrical, irrational logic. A trace can be
detected in lines 32930, with their paradoxical statement that si patruum
vocant | pater est (if they call him uncle, he is their father). This kind
of short-circuiting identification returns in a different form at the end of
the play, when Atreus chooses to interpret the death of his nephews as
the rebirth of his own children: since Thyestes pain at the death of his
children proves unequivocally that they really were his (a point which of
course had never been in question), then, symmetrically, it would follow
that Atreus children were not the fruit of Aeropes adulterous liaison with
her brother-in-law.
The physical setting of the vaticinium is extremely important. The darkness of the secret rooms of the royal palace inspires fear and awe (6506),
yet this is precisely the place where the Pelopidai usually seek safe answers
(680: responsa . . . certa) in times of crisis and uncertainty (658: lassis rebus . . .
ac dubiis).
The connection between the horrific appearance of these abodes and
the certainty of the answers that the Pelopidai are able to obtain there is
further strengthened by a reference to the Styx, an archetypal locus horridus,
which is also the source of undoubted fides even for the gods (6667). The
diagnostic examination of the victims entrails will resolve Atreus concerns
over the dubius sanguis (perhaps of the dubiae res of 658) of his progeny
(75560):
erepta vivis exta pectoribus tremunt
spirantque venae corque adhuc pavidum salit;
at ille fibras tractat ac fata inspicit
et adhuc calentes viscerum venas notat.
postquam hostiae placuere, securus vacat
iam fratris epulis.
Torn from the still living breast the vitals quiver; the veins still breathe and the
fluttering heart still beats. But he handles the organs and enquires the fates, and
notes the markings of the still warm veins. When with the victims he has satisfied
himself, he is now free to prepare his brothers banquet.
The vocabulary of enquiry employed here is again reminiscent of the

Umbrian-dog simile: note, for instance, the repetition of different verbal
forms that imply Atreus search with technical precision. Note also the
104
pregnant meaning of placuere and of securus, which I take to designate

that Atreus is finally sure that his suspicions were unfounded, and that
his children are really his. The very act that guarantees his revenge over
Thyestes (the chief goal of his actions) is also the means by which he can
lay his other concerns to rest. His empirical enquiry is successful not in
spite of, but because of, its deep association with the instinctual aspects
of his personality: the furor that inspired his actions thus far is now also
explicitly presented as a viable source of rational understanding.
At the end of the tragedy Atreus revels in his triumph (10969; quoted
above, p. 82):
nunc meas laudo manus,
nunc parta vera est palma. perdideram scelus,
nisi sic doleres. liberos nasci mihi
nunc credo, castis nunc fidem reddi toris.
Now I praise my handiwork; now is the true palm won. I would have wasted my
crime, if you werent suffering this much. Now I am convinced that my children
are my own; now I believe that I can trust again the purity of my marriage-bed.
Atreus notes first that Thyestes grief at the revelation of his childrens
gruesome death ensures that he is in fact their father (11002):
t h. quid liberi meruere? at. quod fuerant tui.
t h. natos parenti at. fateor, et, quod me iuvat,
certos.
t h. What was the childrens sin?
at. They were yours.
t h. Sons to the father
at. Sure. And, I am pleased to say, definitely yours.
Shortly afterwards Atreus answers Thyestes moralizing appeal to the gods

with the retort that the true reason for his despair is in fact quite different
(110610):
fuerat hic animus tibi
instruere similes inscio fratri cibos
et adiuvante liberos matre aggredi
similique leto sternere hoc unum obstitit:
tuos putasti.
This had been your plan, to prepare the same banquet for their unwitting father,
and with the help of their mother attack the children and kill them in identical
fashion. Just one thing stopped you: you thought they were yours.
A craftier Tereus
105
The assumption underpinning Atreus reasoning appears to be that

Thyestes despair at the death of his children would have been more moderate if he had been certain that Agamemnon and Menelaus, too, were his
own offspring. But while he must have suspected that this was the case (or,
Atreus claims, he would have made the first move to punish Atreus), the
following sequence of events has made it clear to both Atreus and Thyestes
that Agamemnon and Menelaus are undoubtedly Atreus children.
Atreus characterizes his victory as a triumph of foreknowledge and
anticipation: Thyestes would have tried to catch him unprepared (1107:
inscio), but his own scientia has been faster, and more effective. Moreover,
Atreus is now confident that the children he has killed are undoubtedly
Thyestes own (1102: certos), and, symmetrically, that Thyestes suspicion
that Agamemnon and Menelaus could also have been his offspring has
been proven false. Thyestes has been prevented from mounting a successful revenge plan because of his unconfirmed opinion (1110: putasti) that
Agamemnon and Menelaus could be his children; Atreus, however, has
acted on his apprehension and searched for the truth. Thyestes chief mistake lies in his inability to understand that fear can be a reliable form of
knowledge. Throughout his anguished canticum (92069), Thyestes comes
tantalizingly close to expressing his subconscious fears and thoughts (his
language, accordingly, appears fractured, hesitant, obscure), yet he is still
unable fully to grasp their significance.
Atreus believes that his fresh realization of paternity, as well as Thyestes
grief, can to a certain extent undo the past: liberos nasci mihi | nunc credo,
castis nunc fidem reddi toris (10989). At the end of her tragedy Medea
reaches a similar conclusion: restored is my kingdom, my ravished virginity
is restored! (Med . 984: rediere regna, rapta virginitas redit). Both Atreus and
Medea, by envisaging their destructive revenge as a means to reshape past
events, display a form of logic which is rooted in the world of unconstrained
and boundless desire, finding a suitable home in the guts of Atreus palace.
ii
Atreus passions are consistently intertwined with a deep understanding of
human psychology, and, in general, with a marked intellectual superiority.
An analysis of several passages will highlight exactly how Atreus displays his
intellectual power, especially his psychological insight, his fiendish ability to
manipulate language in ways which far transcend Thyestes literal-minded
approach, and finally his superior awareness and understanding of a literary
tradition which can provide useful protocols for his behaviour.
106
Atreus lures Thyestes back to Argos because he correctly assumes that

Thyestes will not be able to resist the seductive prospect of a return home.
The whole sequence of events bears out Atreus initial claim that he understands full well the workings of Thyestes mind: I know the untamable
spirit of the man; bent it cannot be but it can be broken (Thy. 199200:
novi ego ingenium viri | indocile: flecti non potest frangi potest).80 The dialogue between Thyestes and Tantalus in which the former elaborates at
length his hesitation as they approach the city can only bolster the audiences
impression that Atreus always knew better. As Tantalus himself points out,
Thyestes doubts are pathetically overdue: it is too late to guard when in the
midst of danger (487: serum est cavendi tempus in mediis malis). This ability
for psychological insight is initially revealed in Atreus discussion with the
counsellor, who doubts that Thyestes fearful as he is of a possible revenge
will accept Atreus invitation (2945):
s a . quis fidem pacis dabit?
cui tanta credet? at. credula est spes improba.
s a . Who will give him confidence in peace? Whom will he trust so much?
at. Wicked hope is credulous.
Shortly thereafter the counsellor offers, in the dogmatic form of a sententia

(one of his favourite forms of expression),81 a commonplace psychological
reason why Thyestes is unlikely to accept his brothers invitation (3025):
at. hinc vetus regni furor,
illinc egestas tristis ac durus labor
quamvis rigentem tot malis subigent virum.
s a . iam tempus illi fecit aerumnas leves.
at. On the one side, his ancient rage for power, on the other, miserable poverty and
harsh toil will tame the man, however much hardened by so many disasters.
s a. By now time has made his troubles light.
Atreus is quick to dismiss the satelles argument with a statement similarly

couched in sententious terms: You are wrong: a sense of wrongs grows day
by day. It is easy to bear misfortune; to keep bearing it is hard (3067: erras:
malorum sensus accrescit die. | leve est miserias ferre, perferre est grave). The
following sequence of events leaves no doubt as to who is right and wrong
in this exchange, but the impression that Atreus actually understands the
80
81
Thyestes himself will admit in due course that Atreus had been right all along, though by not
spelling out any specific detail he continues to dissimulate to a degree (51314): sed fateor, Atreu,
fateor, admisi omnia | quae credidisti.
On sententiae see p. 157.
A craftier Tereus
107
whole situation better than anybody else is also confirmed at a later stage
by an unexpected source Thyestes himself. In the canticum immediately
prior to the final anagnorisis Thyestes expresses his joy at the end of his
long suffering (9224):
fugiat maeror fugiatque pavor,
fugiat trepidi comes exilii
tristis egestas
away with grief, away with terror, away with bitter want, the companion of hunted
exiles.
The literal repetition of Atreus own words at line 303 indirectly reveals
that Atreus evaluation of his brothers feelings had been right all the time,
and that the superficially wise satelles had actually failed to understand an
important aspect of Thyestes personality. Towards the end of the same
section, however, Thyestes mood shifts considerably, as he is suddenly
overcome by an ominous and inexplicable sensation of fear (95764):
mittit luctus signa futuri
mens ante sui praesaga mali:
instat nautis fera tempestas,
cum sine vento tranquilla tument.
quos tibi luctus quosve tumultus
fingis, demens?
credula praesta pectora fratri:
iam, quidquid id est, vel sine causa
vel sero times.
My mind gives warning of imminent grief, presaging evil for itself; when the calm
sea swells without wind, a harsh tempest is upon the sailors. What distresses, what
upheavals do you imagine for yourself, you fool? Let your heart trust your brother:
by now, whatever it is, you worry about it either without reason, or too late.
Credula at line 962 echoes credula at line 295 and confirms that Atreus was
right to assume that Thyestes would not shun his invitation. Here, once
again, Thyestes proves himself an inadequate reader of signs, signs that he
detects but fails to exploit, since he is a defeatist who yields to the force of
events.
While Atreus successfully combines passion and rational knowledge,
exploiting a thorough understanding of the former as a reliable basis for the
latter, Thyestes owes his demise largely to his mistrust of (subconscious)
feelings as cognitive tools. Once he reaches Argos Thyestes has a final,
albeit belated and ineffectual, moment of hesitation. In his exchange with
Tantalus he does have doubts and fears which the play will realize; he
108
suspects that Atreus is plotting his revenge, and insists on turning back:
but now I am returned to my fears; my mind falters and wishes to take my
body back (41820: nunc contra in metus | revolvor: animus haeret ac retro
cupit | corpus referre). The very setting of the scene Thyestes is already
in Argos taints his proclamation with irony, since his wise words on
the potentially deceptive appearance of things are not based on previously
ignored details (416: cum quod datur spectabis, et dantem aspice, when you
look at a gift, check who is giving it, too). Nonetheless Thyestes insists on
his desire to avoid meeting Atreus (4347):
causam timoris ipse quam ignoro exigis.
nihil timendum video, sed timeo tamen.
placet ire, pigris membris sed genibus labant,
alioque quam quo nitor abductus feror.
You ask me the cause of my fear, but myself I do not know it. I see nothing I should
fear, yet I do. I would like to go, but my limbs waver on my shaky knees, and I
feel I am dragged away from where I strive to go.
Thyestes confronts here the same opposition between rational understanding and emotional foreboding that we have encountered before, but he is
ultimately unable to rely on the cognitive force of metus. He falls prey to
Tantalus well intentioned, if somewhat superficial, pleas, and marches
towards his destiny. Thyestes closely follows the words Atreus had used to
describe the state of manic excitement which pre-empted his masterful
creation of the revenge-plot (2602):82
fateor. tumultus pectora attonitus quatit
penitusque volvit; rapior et quo nescio,
sed rapior.
I do confess it. A mindless tumult shakes and churns my breast deep inside. I am
dragged away, I do not know where to, but I am.
Atreus, however, did follow his emotions and was thus able to devise a plan
whose success is now increasingly likely; yet Thyestes experiences a similar
inner tension, but does not listen to his emotions and thus faces a complete
defeat.
Once alerted to the implications of this internal allusion, we will be even
more inclined to receive Thyestes ensuing speech with scepticism, if not
incredulity. Lines 44670 are devoted to a long rhetorical parade, largely
dependent on well-known topoi, in which Thyestes proclaims his preference
82
See p. 51 (with n. 60 on attonitus).
A craftier Tereus
109
for a quiet life removed from the superficial attractions of power. This speech
is often considered to be paradigmatic of the positive ethical values that
are potentially offered by the tragedy as a whole.83 Even if we discount
for the moment the larger, and definitely more complex, ethical frame
that the play elaborately constructs, the epistemological status of Thyestes
considerations renders them unreliable and even ironic, for clearly he does
not practise what he is in the process of preaching. Thyestes concludes his
impassioned tirade with an adynaton which, one expects, should convey an
unshakeable conviction (47682):
amat Thyesten frater? aetherias prius
perfundet Arctos pontus et Siculi rapax
consistet aestus unda et Ionio seges
matura pelago surget et lucem dabit
nox atra terris, ante cum flammis aquae,
cum morte vita, cum mari ventus fidem
foedusque iungent.
His brother loves Thyestes? Sooner the sea will bathe the heavenly Bears, and the
greedy waves of the Sicilian strait will be still; mature crops will grow in the Ionian
sea and dark night will give light to earth; sooner water with fire, life with death,
and wind with sea will join in a trusty pact.
Once again this intimation, which in itself is perfectly justified and expressed in such strong terms, is inexplicably discarded just a few lines later,
when Thyestes reluctantly embraces Tantalus point of view that it is now
too late for fear, and that they should proceed to meet Atreus. In fact, alongside the highly elaborate rhetorical tone of the adynaton, Thyestes brusque
and inconsistent decision looks even more dissonant and inconsequential.
The effect is similar to the one achieved at 53943, when Thyestes rapidly
retreats from his proclaimed determination not to accept the power that
Atreus offers him (540: respuere certum est regna consilium mihi, to refuse
the throne is my fixed intent) and quickly yields to his invitation (542:
accipio, I do accept).84
The second time that Thyestes confronts a reliable insight on the true
state of events, an insight offered not by rational consideration but by pure
emotion, he behaves in exactly the same way. His canticum opens with
an explicit rejection of pavor (922), followed by a reproach of the usual
attitudes of the wretched, who cannot believe their novel good fortune
(93841):
83
See below, pp. 166ff.
84
110

proprium hoc miseros sequitur vitium,
numquam rebus credere laetis:
redeat felix fortuna licet,
tamen afflictos gaudere piget.
This failing is typical of luckless people; they never put trust in their happiness:
even when their good fortune returns, those who have suffered find it hard to
rejoice.
Credere carries obvious ironic overtones that extend to the whole sententious
tone of the phrase: once again Thyestes talks like a wise man, only to find his
words received by an audience which, on the basis of its previous experience,
cannot possibly believe them. Dolor swiftly follows (9424):
quid me revocas festumque vetas
celebrare diem, quid flere iubes,
nulla surgens dolor ex causa?
Why do you restrain me and forbid my celebrating this festive day? why do you
force me to cry, o grief springing up without a cause?
Thyestes repeatedly fails to understand the underlying causes of his feelings

(434, 964, 967) and is thus incapable or unwilling to trust them, when
they could have offered him a means of escape, at least the first time. His
misdirected rationalism has only assisted Atreus ploys and demonstrated
once more his uncanny ability to manipulate knowledge successfully in
order to achieve his goals. Atreus words after the canticum offer final,
triumphant proof of this ability, as he mocks his brother with elaborate lies
about his good intentions (9702; 976):
festum diem, germane, consensu pari
celebremus: hic est, sceptra qui firmet mea
solidamque pacis alliget certae fidem.
...
hic esse natos crede in amplexu patris
My brother, let us celebrate this festive day with mutual accord; this is the day
which will make my sceptre firm and bind tightly the bonds of our assured
peace . . . Be sure that your sons are here in the bosom of their father.
Only when he can no longer forestall the tragic fate of his children does
Thyestes seem capable of borrowing Atreus smart, ironic use of language.
This moment comes in the emotional and expressive centre of the tragedy,
when Atreus unveils (in more senses than one) the severed heads of his
victims before their horrified father. To Atreus mocking question natos
ecquid agnoscis tuos? (do you recognize your sons?, 1005) Thyestes replies
A craftier Tereus
111
without missing a beat: agnosco fratrem (I do recognize my brother, 1006).

This is, we soon realize, a momentary insight, where the truth shines
through and is at last acknowledged even by a reluctant Thyestes. In
this extraordinary moment of primal pain Thyestes faces the raw truth
of the feelings he had previously mistrusted: Atreus could not possibly have
changed for the better. In the logic of anagnorisis, past certainties return
to reclaim their importance. In his retort, Thyestes is able to compete with
his brothers epistemological and emotional self-assurance, to face reality
without the painstaking veneer of pious intentions and illusions. It is, appropriately enough, only a fleeting moment of truth, and hopelessly belated
at that. After his epigrammatic repartee Thyestes can only invoke divine
retribution, a solution which sounds hollow and ineffectual given Atreus
own appropriation of a divine role. Yet even this momentary ability to stare
truth in the face confirms that only emotional awareness can afford such
an epiphany, that one moment of piercing pain, resistant to any verbal
rationalization, can reveal the truth in its vilest upsetting contours.
iii
A canny master of ideas, Atreus is also an exceptional crafter of language.
His power is expressed also through a careful exploitation of double entendres which fly over Thyestes head: the contrast between Thyestes literalmindedness and Atreus sophisticated dissemblance is another aspect of
the epistemological battle between the two brothers. Atreus manipulative
use of language is responsible for the often unsettling curious mixture of
horror and wit which characterizes this tragedy. Once again this feature
finds a pertinent parallel in Shakespeares Titus Andronicus, where Aarons
satanic drollery conceals his savage intentions in the reassuring metaphors
of elevated poetic language (a mirroring of Shakespeares own writing
process).85
Some instances of this phenomenon are particularly noteworthy.86 When
Atreus promises: wear the crown set on your reverend head; I will offer
to the gods the destined victims (5445: imposita capiti vincla venerando
gere; | ego destinatas victimas superis dabo) the reader is aware of the gory
implications of his words, and this awareness creates a complicity central
to the emotional balance of the play. But the ironic overtones of Atreus
double entendres are nowhere more pronounced than in his final meeting
with Thyestes, when the latter is at last dimly conscious that terrible deeds
85
Bate (1995) 11.
86
See Meltzer (1988), a valuable discussion which focuses especially on act 5.
112
have been perpetrated. At Thyestes request to give him back his children,
Atreus responds with a riddle (9978):
t h.
redde iam gnatos mihi!
at. reddam, et tibi illos nullus eripiet dies.
t h. Give back my sons to me!
at. I will give them back, and no day will grab them away from you.
His response to Thyestes subsequent request is no different (102731):

t h.
redde quod cernas statim
uri; nihil te genitor habiturus rogo,
sed perditurus. at. quidquid e natis tuis
superest habes, quodcumque non superest habes.
t h. Give me back what you will see burned at once. As a father, I am not asking
for something to keep, but to lose.
at. Whatever is left of your sons, you have; whatever is not left, you have.
Riddles, puns and double entendres, far from being mere verbal accessories, are an intrinsic part of Atreus primacy over Thyestes. They also
convey the deeper conviction that taking things at face value is a desperately inadequate strategy when confronting unpredictable, cataclysmic and
monstrous deeds. In a cosmos in which even the sun will be forced to alter
its course, it is foolish of Thyestes not to realize that words may not quite
mean what they seem to mean. In this respect Seneca is fully involved in
a reflection on the limits of irony, which is already developed in Ovid and
will become central in Tacitus.87
The different levels of linguistic awareness displayed by Thyestes and
Atreus can be closely charted in a series of utterances centred on the use
of the verb capio and its compounds. Capio is used many times by both
brothers, and it soon establishes itself as a keyword which precipitates many
of the central themes of the play, at least as soon as Atreus offers an interpretation of his brothers behaviour in typically epigrammatic form: to
the satelles objection that Thyestes is not likely to be taken in by the plot
which he is brewing, Atreus points out Thyestes self-defeating inconsistency: non poterat capi, | nisi capere vellet (he could not be caught, were
he not bent on catching, 2889); the ominous connotations of the word
are revealed in the same scene, as Atreus elaborates on various aspects of his
plan and assumes that Thyestes sons will easily be taken in by the illusion
of a return home: if too stubbornly Thyestes spurns my prayers, I will
87
On Ovid see Doblhofer (1960) and Schawaller (1987); on Tacitus see especially Plass (1988), who
also has interesting remarks (928) on Seneca.
A craftier Tereus
113
move his sons with my entreaties: they are inexperienced, weighed down
by grave misfortunes, and easy to trick (299302: si nimis durus preces |
spernet Thyestes, liberos eius rudes | malisque fessos gravibus et faciles capi |
prece commovebo); even more sinister are Thyestes words in his highly
rhetorical praise of a modest life: oh, how good it is not to be an obstacle
to anyone, to eat food without care while lying on the ground! (44951:
o quantum bonum est | obstare nulli, capere securas dapes | humi iacentem!),
which are echoed again in the anagnorisis scene: AT. poculum infuso
cape | gentile Baccho. T H. capio fraternae dapis | donum (AT. Take this
cup, an heirloom, filled with wine. T H. I take this gift of my brothers
feast, 9824).88 At 5201 Thyestes entrusts his children to Atreus obsides
fidei accipe | hos innocentes, frater (as pledge of my faith, brother, take these
innocent boys) who will return them with precisely the same word: iam
accipe hos potius libens | diu expetitos: nulla per fratrem est mora (now, rather,
take these with joy; you have waited for them a long time. Your brother
causes no delay, 10212). And it is finally Thyestes who highlights the dramatic echoes of capio in his last (and involuntary) pun on the word: hoc est
quod avidus capere non potuit pater (this much the father, for all his greed,
could not devour, 1040); Thyestes was unable to understand what lay in
store for him, but was tragically capable of receiving the flesh of his own
children.89
Ironic twists on capio come to symbolize Thyestes intellectual inadequacy and weak resolve. We might apply to him the choruss remark on
Tantalus in the underworld, who is unable to reach the food and drink
laid in front of him time and again: deceptus totiens tangere neglegit (deceived so often, he tries no more to touch, 159). Indeed Tantalus displays
a self-defeating masochism which the chorus captures with epigrammatic
brevity: falli libuit (gladly has he been baffled, 167).
iv
Atreus use of obscure forms of communication as he plots the mise en
scène of act 3 is part of his dissembling character. Atreus rightly identifies dissimulation as an instrumentum regni, and in so doing he problematizes the
contrast between tyrant and king which had been proposed in the second
act. The tyrant can disguise his threats, and plausibly act as a good king
tyranny and dissimulation are closely connected in Greek and Roman
88
89
Note that gentile, too, is ominously ambiguous, since it could suggest a drink consisting of your
gens, with wine poured upon it (Tarrant (1985) 227).
On metaphors of incorporation see Kilgour (1990) and now especially Rimell (2002).
114
thought. The potential ambiguity of dissimulation makes the king, but

especially the tyrant, a difficult text, and it inevitably raises an epistemological as well as a political problem. In Thyestes everybody dissimulates: the
satelles disguises his fear; Atreus cloaks his thirst for revenge; Thyestes covers
up his own worries. The distinction is not between those who dissimulate
and those who do not, but between effective and ineffective dissimulation.
The intellectually superior Atreus is fully aware of Thyestes deception, and
goes on to triumph over him. Thyestes, on the other hand, suspects that his
brother is dissimulating, but fatally he does not act on this intimation.
Power and dissimulation are already linked as anthropological themes;
witness the many stories in which a king seizes power by acting as a harmless fool, for instance Peisistratos and Brutus, or, before them, Odysseus
, who dressed up as a poor beggar in order to regain his
throne and his wife.90 In Roman political discourse dissimulation is a defect traditionally associated with Tiberius, thanks of course to Tacitus and
Suetonius pathological portraits.91 It should not be forgotten, however,
that a form of dissimulation characterizes imperial power from the outset.
Brutus dissimulation marks the end of the monarchy and the beginning of
the Republic, but Augustus own dissimulation allows an essentially monarchical power to be smuggled in as a slightly edited version of the Republican
constitution.
As Torquato Accetto will brilliantly point out centuries later in Della
dissimulazione onesta, dissimulation is a totalizing form of communication
and behaviour, because the discourse of dissimulation must dissimulate,
and also because the only way to reply to those who dissimulate is by
dissimulating in turn. This is why Atreus is afraid that his children may not
be able to dissimulate (315). Dissimulation, in sum, is a weapon of power and
against power, and must be judged according to internal criteria of efficacy
and expediency. Thyestes, technically speaking, is a bad dissimulator, Atreus
an excellent one.
Dissimulation is deeply connected with theatrical fiction, and it is interesting to note that the product of Atreus dissimulation is the mise en scène
of act 3. Dissimulation and deception are principles of artistic creation at
least since Hesiod, well before Accetto will write, famously, that everything
beautiful is nothing but gentle dissimulation. Like Hesiods Muses, tyrants
90
91
On Brutus see Bettini (1987). In Shakespeares tragedy Titus must also resort to dissimulation he
feigns madness in order to accomplish his revenge: I knew them all, though they supposed me
mad, | And will oerreach them in their own devices | A pair of cursed hellhounds and their dam
(5.2.1424).
See Giua (1975); Zecchini (1986); Baar (1990) 14650 (and 517).
A craftier Tereus
115
and poets deceive or reveal the truth on a whim. A constantly dissimulating

tyrant is inevitably a bad dissimulator (as Tiberius is, in fact, according to
Accetto). The discourse of power, like the discourse of poetry, is very much
exposed to the deconstructive force of dissimulation. Already Odysseus,
as he lies while maintaining that he is speaking truthfully (Od . 14.192),
comes dangerously close to the Cretan paradox. In his Panegyricus Pliny will
state that sincerity, or rather the appearance of sincerity, can be obtained
by emulating those forms of spontaneous expression that it would take too
long to falsify.92 Centuries later Baltasar Gracian will argue that after all
sincerity itself is a lie.93 (Modern literary theory would indeed agree that
the reality effect intensifies the fictional status of a narrative.)
v
As an authorial figure, Atreus is fully aware of the intertextual inspiration
of his actions, and this knowledge of precedents and models will give
him a decisive advantage at crucial junctures. Atreus explicitly displays
his knowledge of the Ovidian story of Procne and Tereus, the single most
important source of inspiration which he invokes in his very first appearance
on the stage. Thus, as we have seen, when we hear Thyestes declare, in the
emotionally charged meeting with his brother, that lacrimis agendum est
(517), we can only suspect that he is simply ignoring Procnes more resolute
words in Ovids Metamorphoses: non est lacrimis hoc inquit agendum. 94
This is an oversight that Atreus would certainly have avoided, but it is
not simply a matter of academic competence.95 The important point is
that while Atreus is following a masterplot which guarantees him useful
material for his revenge, Thyestes confesses his ignorance of that model,
and fails to foresee the fatal danger that awaits him.
No less ironic is the effect resulting from Thyestes inept appropriation
of Virgil. In his canticum at the beginning of act 5 Thyestes begins to be
dimly aware of the tragedy awaiting revelation: my mind gives warning of
imminent grief, presaging evil for itself (9578: mittit luctus signa futuri |
mens ante sui praesaga mali). Virgils Mezentius had been able to realize even
before the procession arrived that the corpse carried back to the camp was
92
93
95
See Pan. 3.1 and 3.4, whose contorted logic reveals a very interesting cognitive quandary. Note that
similar concerns emerge already in Republican times, as the fractiousness of political life destroys
deep-seated beliefs in the certainty of the meaning of key political terms: vera vocabula rerum
amisimus, as Catilina points out (Sall. Cat. 52.11, with Canfora (1991)).
94 See p. 77.
Oraculo Manual y Arte de Prudencia, para. 13.
We should perhaps remember Ps.-Longinus observation that in fact one finds low emotions distinct
from the sublime, like pity, grief, fear (Subl. 8.2).
116
that of his son Lausus: agnovit longe gemitum praesaga mali mens (Mezentius
had a presentiment of evil. He heard the wailing in the distance and knew
the truth, Aen. 10.843). The almost verbatim repetition highlights the sharp
contrast between Thyestes and Mezentius, the former unable to decode the
ominous signs that surround him just as the latter swiftly jumps to the right
conclusion.96
A similar instance of Thyestes insensitivity to literary models can be
found in another elaborate intertextual connection which I have already
touched upon.97 Finally back in Argos, Thyestes recalls with barely restrained emotion his youthful victories in the races, as he sees the racecourse thronged with youth, where more than once, lifted to fame, have I
in my fathers chariot won the palm (40910: celebrata iuveni stadia, per
quae nobilis | palmam paterno non semel curru tuli). As we have seen, these
lines echo two important programmatic passages, Horaces first ode (Carm.
1.1.36) and the proem to the third book of Virgils Georgics (3.1020).
Atreus, too, will have a chance to reactivate the audiences memory of these
models by picking up the keyword palma almost at the very end of the play:
nunc meas laudo manus, | nunc parta vera est palma. perdideram scelus, |
nisi sic doleres (10968). His palma, to be sure, has nothing to do with
Thyestes racing exploits; it has been warranted by the astute manipulation
of reality on which his revenge has been predicated all along, by his ability
to produce a spectacle (Thyestes pained by harrowing grief ) which constitutes his own literary masterpiece.98 In his first ode Horace had singled
out racing victories as the first item in a long list of lesser pursuits which
he shuns for the glory of poetry, as he declares at the end of the poem,
addressed to Maecenas: but if you include me among lyric bards, I will hit
the stars with my exalted head (quodsi me lyricis vatibus inseres, | sublimi
feriam sidera vertice).99 Atreus, by contrasting his vera palma with Thyestes
pointless evocation of past sporting achievements, shares his awareness of
Virgils and Horaces line of thought, which, evidently, Thyestes either did
not know or did not share.
96
97
98
99
The instance is analysed by Tarrant (1985) 225. Note that, significantly, Atreus is similar to Mezentius;
see below, pp. 1256 (on the lion simile).
Ch. 2, p. 59.
Actors may have fought to conquer a palma already in Plautus time; see Plaut. Amph. 69, Poen. 37,
Trin. 706 and Ter. Phorm. 1617 with Duckworth (1952) 78. For authors see Cic. Phil. 1.36.
Hor. Carm. 1.1.356. A parallel could also be drawn between the imagery of 8856 (aequalis astris
gradior et cunctos super | altum superbo vertice attingens polum) and that of Hor. Carm. 2.20.14 (non
usitata nec tenui ferar | penna biformis per liquidum aethera | vates, neque in terris morabor | longius . . .).
2.20 concludes the second book of the Odes, and aequalis astris . . . of Atreus (885) signals a similar
moment of completion (cf. 8889: summa votorum attigi. | bene est, abunde est, iam sat est etiam
mihi).
A craftier Tereus
117
By displaying his intimate, active knowledge of the literary tradition,

Atreus further boosts his privileged relationship with the audience, which is
invited to share Atreus literary awareness, and thus to side against Thyestes
unattractive literalness, his deadly lack of literary competence.
per f e c t i o n , o f a k i n d
Perfection, of a kind, was what he was after,
And the poetry he invented was easy to understand;
He knew human folly like the back of his hand,
And was greatly interested in armies and fleets;
When he laughed, respectable senators burst with laughter,
And when he cried the little children died in the streets.
(W. S. Auden, Epitaph on a Tyrant)
i
Through a powerful combination of qualities passion and reason, tragic
violence and tragi-comic irony and through his ability to exceed the
expected and the acceptable, Atreus embodies in the play the limitless
energy that Tantalus had tried in vain to keep in check in the prologue to
Thyestes.100 Atreus power is doubly lethal, because it not only makes room
for nefas, but also gives it an unquestionable aesthetic attractiveness. Atreus
the poet is cunning, funny, articulate, simply irresistible. The destruction
of any boundary to nefas and decorum is thus inextricably linked to his
creative power, and we, the audience, must admit that one cannot exist
without the other.
Thyestes discourages, we have seen, clear-cut definitions of characters and
(even more) their hasty promotion to ethical types. Atreus cannot be reduced to a furious monster, Thyestes to a Stoic sapiens more or less close to
possessing a bona mens. Positing a stark contrast between an unreasonable
tyrant and a (potentially) good king would be equally unreliable, if for no
other reason than that tyrant and rex iustus are not ontologically opposite
types. As Cicero, for instance, points out, the just king and the tyrant are
different points in a continuum. Acting like a tyrant can be a momentary
madness or a lifelong pattern, since it depends on a more or less successful control over passions, but one is not born a tyrant, and the struggle
is never won once and for all. A tyrannus can always be lurking behind
the comforting image of the rex, and, as Cicero claims, there is very little
100
See above, ch. 2, passim.
118
distance between the two: when the king begins to act unjustly . . . he himself is a tyrant, the worst type, and the closest one to the best.101
The mastertext of this juxtaposition between king and tyrant is to be
found in Platos Republic.102 The tyrannical man comes about through a
degeneration of the democratic man, because in the former remain stronger
and more numerous (571b) the illicit, indeed terrible and savage (572b)
desires common to all men, but in some individuals . . . repressed by laws
and better instincts can be totally extirpated or lessened and weakened
(571b). These desires (571cd):
are awakened in sleep when the rest of the soul, the rational ( ), gentle
and dominant part, slumbers, but the beastly and savage ( )
part, replete with food and wine, gambols and, repelling sleep, endeavours to sally
forth and satisfy its own instincts. You are aware that in such case there is nothing it
will not venture to undertake as being released from all sense of shame ( )
and all reason ( ). It does not shrink from attempting to lie with a
mother in fancy, or with anyone else, man, god or brute. It is ready for any foul
deed of blood; it abstains from no food, and, in a word, falls short of no extreme
of folly (
) and shamelessness (
).
This passage, whose wider significance will not escape Sigmund Freud,103
posits a connection between psychology and politics which will be at work
more or less explicitly in most of the Hellenistic and Roman reflection on
the good king, and definitely in Senecas own De clementia. The sleep of
reason, we might well say, creates tyrants; or, to put it another way, the
tyrant is he who never controls or represses his instincts but gives them
immediate and complete satisfaction. We know from a great wealth of
anthropological and literary material that all rulers, in more or less mediated
or terrifying ways, are characterized as men who regularly break or trespass
all sorts of boundaries. Rulers (especially tyrants) are all-powerful since their
superhuman power makes them more similar to terrifying animals such as
lions and leopards than to mere mortals.104
The literature of Imperial Rome focuses insistently on this conceptual
knot. One may turn to Suetonius Lives of the Twelve Caesars, a fascinating
101
103
104
102 See Lanza (1977) esp. 6594.

Cic. Rep. 1.65.
First of all in a 1914 addition to The Interpretation of Dreams (Freud (1900) = SE i v.67): Plato,
on the contrary, thought that the best men are those who only dream what other men do in their
waking life. The remark is echoed almost verbatim in Introductory Lectures on Psycho-Analysis (Freud
(191617) = SE xv .146). There is no mention of Plato in Creative Writers and Day-Dreaming
(Der Dichter und das Phantasieren (1907): SE i x.14353), where Freud analyses the relationship
between (day)-dreams, fantasy and poetic creation and effectively, if implicitly, appropriates for
psychoanalysis Platos seminal observation. It is a well-known limitation of this and other Freudian
writings on art that they focus more on the subject matter than the signifying practices shared by
art and the unconscious.
On the excesses of Greek tyrants see Catenacci (1996), especially 14270 on erotic ones.
A craftier Tereus
119
document about the perception of power in the first century, to understand

how the contemporary imagination lived with the presence, high on the
Palatine, of rulers increasingly free from meaningful checks and balances.
Politics must needs turn into psychology, because it is the individual rulers
psychology, that of the rex-tyrannus, which must be controlled, reined in,
bettered. The whole of De clementia, the most important work for understanding Senecas and some of his contemporaries political vision, is simply
an attempt to persuade Nero that it is in his moral and practical interest to
acknowledge those limits which no outside force is any longer capable of
imposing.
Platos intuition that the tyrant is a man who gives free rein in his life
to alogon, the violation of rational and ethical norms, and accomplishes
what moral self-repression or external laws keep out of the reach of normal
people is rich in theoretical implications. If in the tyrant there is at work a
form of extreme violence akin to the violence of unrepressed desires, those
which get free rein in dreams unless proper rational control is exercised,
then a dispassionate reflection on the tyrants potential emotional appeal
as a literary character is in order, especially as he can be at the same time
terrifying and magnificent, attractive and repulsive.
In book 10 Plato deals for the second and last time with the issue of poetry
and its dangers. Poetry leads people astray for at least two reasons: because
poets tend to imitate in their work the worse instincts of the soul, not the
better ones (603c605c), and because poetry incites the audience to privilege
the parts of the soul which are best kept under control (605c607a). Poetry
is equivalent to loosening inhibition, to yielding to alogon, which in political
terms is embodied by the tyrannical character, linked as he is to irrational
and uncontrollable forces. Poetry has no citizenship in a well-regulated polis,
because it escapes the control of reason. The notion that poetic inspiration is
connected with divine elements and contains something inexplicable is prePlatonic (Democritus), but it is Platos specific contribution to regard this
inspiration as irrational, even Bacchic (533e534e).105 Poetry often arrives in
dreams, is inspired by supernatural sources whose epistemological status is
frequently debated in Augustan poetry. The rich tradition of the inspiring
dream codifies in Greek and Roman literature the positive side of the
relationship between poetry and dreams. But another side is not entirely
forgotten, namely the awareness that an excess of poetic irrationality can be
105
See Murray (1996) 79. Note in this context Zeitlins remarks on the gender assumptions which
underlie Platos rejection of poetry as a dangerous female mimesis (1996) 36774 (on the connection
between women and mimesis Zeitlin (1996) 375416 is crucial). Indeed the tyrant, a theatrical man
who is a slave of his passions, resembles a woman: Resp. 577b and 579bc with Zeitlin (1996) 371
and n. 56.
120
compared to the disturbed and unreliable dreams of a sick man. Consider

the opening of Horaces Ars poetica (69a):
credite, Pisones, isti tabulae fore librum
persimilem cuius, velut aegri somnia, vanae
fingentur species, ut nec pes nec caput uni
reddatur formae.
Believe me, Pisones, the book will be very similar to this picture, if idle fancies
(vanae . . . species) are shaped in it as in the dreams of feverish people, so that neither
head nor foot can be assigned to a single shape.
Horace contrasts the folly of this limitless imagination with the reliable
rules of good judgement (9b-13):
pictoribus atque poetis
quidlibet audendi semper fuit aequa potestas
scimus, et hanc veniam petimusque damusque vicissim;
sed non ut placidis coeant immitia, non ut
serpentes avibus geminentur, tigribus agni.
But painters and poets have always enjoyed a full right to dare whatever they
fancied. True, this is a licence which we poets request and concede in turn; but
not to the extent that savage animals should lie down with domestic ones, or snakes
should mate with birds, or lambs with tigers.
The licence (licentia) which Horace grants to poets106 is an enlightened

absolutism of sorts: there is a lot they are free to do, but they should not
overturn the foundations of human nature and society. Light-hearted and
full of grotesque imagery as these lines may sound, their seriousness is not to
be underestimated, especially if we consider that the Horatian examples of
adynata, of impossible conjunctions, recall a very important section of De
rerum natura book 5 (lines 878924). There Lucretius argues that the first
living creatures created by Mother Earth must surely have been imperfect,
even monstrous to our mind, but they could not defy the basic rules of
atomic aggregation which forbid the union of different species.
Let me briefly restate the crux of my argument: it is possible to argue, at both
a contextual and theoretical level, that poetry is the sphere of human activity
where the kinds of thoughts, feelings and images which reason would rather
keep under control and even silence are expressed and communicated.
Furthermore, one might propose that the poets violation of this censorship
is homologous to the tyrants transgression of behavioural norms. The poet,
106
Cf. Mayor (1879), and Brink ad loc.
A craftier Tereus
121
that is, is like the tyrant, because just like him he ignores the boundaries set
by logos and nomos. Not all poets, and not all the time, of course. We should
adopt in this case as well an articulation of the concept that is parallel to the
one suggested above between rex iustus and tyrant, an articulation which
Augustan and post-Augustan poetry and poetics encapsulate in the related
but distinct concepts of poeta and vates.107 We could therefore complete the
theoretical proposition by positing an analogy, on the one hand, between
poeta and rex iustus, and, on the other, between vates and tyrant, the former
two champions of moderation and self-restraint, the latter closer to sublime
forces of Bacchic enthousiasmos or Apollinean inspiration.
The parallelism that I posit does not exhaust the exegetical dividends
afforded by comparing the political and the poetic. Indeed, the theoretical
argument developed so far can be put to further use once we formulate a
final corollary. I would like to argue that there exists between the tyrant
and the enthusiastic vates a latent solidarity based on a basic homology.
In first-century literature the tyrant is attractive because of the similarities
between those who exercise political power and the power of the poets.108
A vates will invariably be a subject of power in his sphere of activity, and an
object of power in the political domain: hence the powerful tensions and
contradictions we find in the relationship between poets and rulers in the
Rome of Augustus and Nero (and beyond).
ii
The tyrants attractiveness is rooted in the characteristics he shares with the
vates. Poets and tyrants are similar, first of all, because they both claim for
themselves the right to act supra . . . fines moris humani (268). They both
are authors, auctores, creators and innovators of reality, masters of life and
death, of creation and destruction.109 Atreus power is explicitly connected,
in the tragedy, with his ability to create a compelling mise en scène. The
poet is like the tyrant, and the tyrant can dress up as a poet in order to fulfil
his goals. His weapons are exactly the same: creativity, dissimulation, irony,
107
108
109
For a full picture of the emergence and development of the concept of vates in the first century see
Newman (1967). One feature appears to underline the new meaning of the word after Virgil (and
Varro, who, according to Isidorus, Orig. 8.7.3, suggested among others the interesting etymology
of vates a vi mentis): the term vates made the poet a being with more than ordinary powers
(Newman (1967) 100). See Jocelyn (1995).
A similarity that famously becomes, in the case of Nero, an identification; Bartsch (1994) 3662.
One might recall the debate between Socrates and Polus on whether rhetors have real power in the
polis. According to Polus they do, because like tyrants, they can kill whoever they want to, deprive
anyone of his property and expel him from their cities as they think fit (Pl. Grg. 466c).
122
double entendres, knowledge of the literary tradition. Atreus superiority

over Thyestes in words, logic and deeds is reflected in his strong impact on
the audience. If nothing else, there can be no doubt where our aesthetic
allegiances lie:110 with Atreus energetic poiesis, his mastery of words and
puns, his ruthless determination to plot, stage and act his revenge.
One way to gauge the potential effect of Atreus on the audience is to
look at the reactions of the characters who watch him within the play, from
the docility of the counsellor in act 2 to the messengers horror as he recalls
the sacrificial slaughter in act 4. In both cases we face the impotent awe
of human beings confronted with behaviour that goes well beyond their
normal horizon of expectations.
The messenger provides the most articulate analysis of the reactions that
Atreus inspires, and conveys them not only in his moral judgement, but
also in the elaborate similes he uses in his gripping portrait of the king.
Atreus is first compared to a tigress in his uncertainty over the order of the
sacrifice (70714):
ieiuna silvis qualis in Gangeticis
inter iuvencos tigris erravit duos,
utriusque praedae cupida quo primum ferat
incerta morsus (flectit hoc rictus suos,
illo reflectit et famem dubiam tenet),
sic dirus Atreus capita devota impiae
speculatur irae. quem prius mactet sibi
dubitat, secunda deinde quem caede immolet.
As in the jungles by the Ganges a hungry tigress wavers between two calves, eager
for both prey, uncertain where she should bite first (to the one she turns her jaws,
then turns to the other, and keeps her hunger waiting), so does cruel Atreus scan
the heads destined to his cruel rage, and wonders whom he should first sacrifice to
himself, whom he should slaughter second.
Shortly afterwards, as the chorus enquires about the fate of Thyestes

younger child, the messenger engages in a new comparison of a similar
type (73241):
silva iubatus qualis Armenia leo
in caede multa victor armento incubat
(cruore rictus madidus et pulsa fame
non ponit iras: hinc et hinc tauros premens
vitulis minatur dente iam lasso inpiger),
non aliter Atreus saevit atque ira tumet,
110
For the notion of aesthetic allegiance see Orlando (1971) passim.
A craftier Tereus
123
ferrumque gemina caede perfusum tenens,

oblitus in quem fureret, infesta manu
exegit ultra corpus, ac pueri statim
pectore receptus ensis in tergo exstitit.
As in the Armenian woods a maned lion, victorious after much slaughter, lies down
amidst the herd (his jaws reek with gore, but even after he has quelled his hunger he
rages on: now here, now there attacking the bulls, he threatens the calves, tireless
even as his jaws are tired) not otherwise Atreus raves and swells with anger and
holding the knife drenched with double slaughter, forgetting whom he is attacking,
with deadly hand he drives it through the body, and the sword enters the boys
breast and stands out upon his back.
While the first simile focuses on Atreus procedural doubt, the tigress
image highlighting the combination of rational and bestial, the second
develops at length an aspect of the lions behaviour emphasized in previous
texts, namely the animals indulgence in violence well beyond the practical
impulse to kill its prey.111 It is not simply hunger that drives the lion, but
an instinctual passion for violence which is partly pursued for its own sake.
Both similes concentrate on Atreus animal-like behaviour, a notion which
should not immediately and inevitably translate into a moral judgement,
but which does introduce a key element of his characterization. As we have
already noticed, Atreus bestial nature plays an important role when the
tragedy comes to terms with the relative positioning of, and transactions
among, men, gods and animals, as articulated in sacrifice.
The immediate antecedent of both similes can be found (unsurprisingly)
in Ovids Metamorphoses. It is especially interesting to note that Procne is
compared to a Gangetica tigris112 just as she sets out to murder her son; the
detail of the beasts hesitation when confronted with two victims further
echoes Perseus own uncertainty at Metamorphoses 5.1649. Neither passage, however, mentions lions, and while yet another Ovidian line might
have suggested the choice of a different animal,113 there is much to be
gained by expanding the possible implications of Senecas departure from
111
112
113
Note the change in gender from a male to a female animal between the two similes. The detail may
be significant in the light of the Dionysiac aspects of Atreus personality which I discuss below.
Dionysus was, notoriously, a sexually ambivalent god, described as having feminine traits, especially
from the fifth century onwards; see Dodds (1960) 1334; Detienne (1979) 2052. The tradition is
present in Rome as well, see Naev. 57 Ribbeck2 (from the Lycurgus). Note also the traditional
association of Dionysus with a lion that goes back to the Homeric Hymns 7.44 (Dodds (1960) xvii
with n. 6), and present in Sen. Oed. 4246, 4578.
Met. 6.6367: veluti Gangetica . . . | . . . tigris, to compare with Sen. Oed . 458: tigris . . . Gangetica,
on which passage see the preceding note. See p. 80.
According to Tarrant (1985) 195, who refers to Met. 15.86: Armeniae tigres iracundique leones. Note
the insistence on ira at Thy. 735 and 737, above.
124
his model. By comparing Atreus to a lion Seneca does more than reiterate Atreus beastly violence: he invites reflection on important thematic
and metaliterary affiliations. Chronology notwithstanding, it is probably
best to turn first to Lucans Caesar, who is characterized as a lion in the
first, extended simile which the Bellum Civile devotes to its main character
(20412):
inde moras solvit belli tumidumque per amnem
signa tulit propere: sicut squalentibus arvis
aestiferae Libyes viso leo comminus hoste
subsedit dubius, totam dum colligit iram;
mox, ubi se saevae stimulavit verbere caudae
erexitque iubam et vasto grave murmur hiatu
infremuit, tum torta levis si lancea Mauri
haereat aut latum subeant venabula pectus,
per ferrum tanti securus volneris exit.
Then he broke the barriers of war and through the swollen river quickly took his
standards. Just so in torrid Libyas barren fields the lion, on seeing his enemy at
hand, crouches in hesitation till he has concentrated all his anger; next he goads
himself with fiercely lashing tail, his mane is bristling, from his massive jaws deep
he roars then if a lance, hurled by a swift Moor, or hunting-spears pierce and
stick in his broad chest, ignoring such a terrible wound he rushes onward, driving
the weapon deeper.
Lucan insists on the lions ira (a traditional detail),114 but also stresses the
beasts almost heroic defiance in the face of the enemy. Lucans Caesar, of
course, consistently proves to be a character whose unrestrainable proclivity to nefas and violence constitutes the emotional and narrative focus of
the poem, an attractive, if fearful, mixture of defiance and ruthlessness,
epic grandeur and impious heroism. In this respect Caesar follows in a
distinguished line of (anti)heroes whose most immediate and influential
model can be traced to two important Virgilian characters, Turnus and
Mezentius. They, too, are repeatedly compared to a lion, and Virgil elicits
from his Homeric model a consistent series of connotations.115 In Aeneid 9
114
115
See Sen. De ira. 2.16.1: iracundia leones adiuvat; Hor. Carm. 1.16.15 with Nisbet-Hubbard (1970)
211. It is important to note that the passage refers to Thyestes ira, a somewhat difficult notion
given Atreus traditional association with revenge. Among the possible explanations that Nisbet
and Hubbard advance, it is worth reporting that of Vollmer (on Stat. Silv. 5.1.57), who suggested
that Varius might have introduced a variation of the legend in which Thyestes anger played a more
significant part.
The most relevant Homeric similes are to be found in Il. 3.236 and 17.5402, which insist on
the lions anticipated joy at the carnage; 12.292300 is also interesting for Sarpedons somewhat
excessive behaviour (like a lion, he has a ! ). Il. 17.617 foregrounds the abundant
blood in which the animal revels. Also, Odysseus victory is compared to the behaviour of a lion,
see Od . 4.3359 ( = 17.12630), with Moulton (1977) 139 and 123 (with further references).
A craftier Tereus
125
Turnus is described as a lion standing unperturbed in front of his enemies

(9.7928):116
ceu saevum turba leonem
cum telis premit infensis; at territus ille,
asper, acerba tuens, retro redit et neque terga
ira dare aut virtus patitur, nec tendere contra
ille quidem hoc cupiens potis est per tela virosque.
haud aliter retro dubius vestigia Turnus
improperata refert et mens exaestuat ira.
. . . crowding him like a pack of huntsmen with levelled spears pressing hard on a
savage lion; the lion is afraid and gives ground, but he is still dangerous, still glaring
at his attackers; his anger and his courage forbid him to turn tail, and though he
would dearly love to, he cannot charge through the wall of steel and the press of
men just so did Turnus give ground, uncertain but unhurried, and his mind was
boiling with rage.
This description focuses on the strength and defiance of the animal, its
bloodthirstiness and grandiosity. Even when wounded, the lion will rejoice
in the forthcoming slaughter and continue to display its determination
(12.49):
Poenorum qualis in arvis
saucius ille gravi venantum vulnere pectus
tum demum movet arma leo, gaudetque comantis
excutiens cervice toros fixumque latronis
impavidus frangit telum et fremit ore cruento:
haud secus accenso gliscit violentia Turno.
Just as a lion in the fields round Carthage, who does not move into battle till he
has received a great wound in his chest from the hunters, and then revels in it,
shaking out the thick mane on his neck; fearlessly he snaps off the shaft left in his
body by the ruffian that threw it, and opens his gory jaws to roar just so did the
violent passion rise in Turnus.
It is precisely the pleasure gained from the anticipated slaughter that Virgil
highlights in the simile devoted to Mezentius later in book 10 (7239):
impastus stabula alta leo ceu saepe peragrans
(suadet enim vesana fames), si forte fugacem
conspexit capream aut surgentem in cornua cervum,
gaudet hians immane comasque arrexit et haeret
visceribus super incumbens; lavit improba taeter
ora cruor
sic ruit in densos alacer Mezentius hostis.
116
A similar, briefer description is to be found in Aen. 10.4546. On these and other animal similes
applied to Turnus see Traina (1990) 327, with further bibliography.
126
Just as a ravening lion scouring the deep lairs of wild beasts, driven mad by the
pangs of hunger, if he sights a frightened she-goat, or sees a stags antlers rising, he
opens his great jaws in delight, his mane bristles and he battens on the flesh with
foul gore washing his pitiless mouth just so did Mezentius charge hot-haste into
the thick of the enemy . . .
By comparing Atreus to a lion Seneca thus places him in a genealogy of

characters who display a powerful passion for nefas and inspire the awed
attention of the audience. Heroic in his evil, Atreus, like Mezentius or
Caesar, shows that the more overt layer of moral condemnation offered
by the poems structuring ideology can be at odds with the inner tensions
and deeper emotions evoked by the text. Turnus, like Mezentius, attains
grandiosity by heeding his passion (his furor) well beyond the normal
bounds of human behaviour.117
The analogy with man-eating lions and tigers indirectly highlights
Atreus involvement with cannibalism. Here again Senecas strategy is rich
and sophisticated. The association with wild beasts and the elaborate cooking scene reported by the messenger all conjure up the image of a cannibalistic Atreus, who would thus join the series of tyrants (especially Eastern
ones) who did not shrink from eating human flesh, sometimes specifically
as a form of punishment.118 In this case, however, Atreus (and Senecas)
masterstroke consists in shifting the blame, indeed the praecipuum . . . nefas (Thy. 285),119 onto Thyestes. Atreus is, for all purposes, the cannibal
of the two, yet he cunningly manages to involve his brother in this peculiarly tyrannical nefas while ostensibly refraining from it himself. This
can be seen, on one level, as the pinnacle of dissimulation, a strategy at
which Atreus excels.120 On the other, however, it proves a central tenet of
Atreus philosophy, that one cannot rely too much on intrinsic differences
between ethical types. Atreus shows how flimsy the divide between man
and animal can be (hence the lion similes and the cannibalism), but shows
furthermore that even among men unforeseen turns of events can result in
a blurring of ethical categories. Again, who is the real animal, Atreus, who
117
118
119
This conclusion is not weakened by the fact that the lion simile is first encountered in connection
with Nisus (9.33941). Nisus is hardly a Mezentius-like character, but it is important to remember
his initial words: Is it the gods who put this ardour into our minds, or does every mans irresistible
desire become his god? (1834: dine hunc ardorem mentibus addunt, | Euryale, an sua cuique deus
fit dira cupido? ).
Again a feature of Platos tyrant, Resp. 565d-566a. An analogy can also be drawn between cannibalism
and incest, both forms of unnatural appropriation of a body; see Parker (1983) 98, 326 and above,
p. 94. Thus, in a sense, as he eats his children Thyestes is also perversely repeating the crime that
led to his punishment. On the Greeks view of cannibalism see Detienne (1979) 5367; on similar
charges directed against a tyrannical Mark Antony see Leigh (1996).
120 On dissimulation see pp. 113ff.
See below, p. 144.
A craftier Tereus
127
plots the banquet, or Thyestes, who actually eats the flesh of his children?
If Atreus is a god, or at least, certainly, plays god with Thyestes (thus
also blurring the boundaries between gods and men), the latter cannot be
absolved of his actions simply because he did not know what he was doing
and because a superior power put him in harms way. Oedipus, of course,
would be able to disabuse Thyestes of any such ill-conceived notion of
innocence.
It is equally important to frame the lion simile in the context of an
explicit reflection on literature proposed by Seneca in Letters to Lucilius 41.
This letter elaborates a defence of the sublime aesthetic appeal of terrifying
images, such as a dark grove, a deep grotto (41.3) or a lion which is speciosus
ex horrido and cannot be watched without intense fear (41.6: non sine timore
aspici): its decor lies in fact in this very quality. Letter 41 does not answer
in full the moral issues raised by such a vocal defence of beauty arising
from fear speciosum ex horrido, as the phenomenon could be defined
but, interestingly, the example reinforces the intimation that one should
live according to ones nature: a tamed, dressed up lion would be a pitiful
spectacle (41.6). This identification of naturalness and aesthetic appeal paves
the way for a full artistic exploitation of the psychagogic and aesthetic
potential of negative characters, and, of course, of such a distinctly Senecan
feature as the locus horridus.121 Atreus is not artistically appealing in spite
of his cruel, negative nature, but precisely because his nature, as we see in
the messengers simile, is not in the least bridled or tamed.
Atreus can thus be seen to embody a form of artistic and behavioural
sublimity which transcends humanity and attracts the audience beyond
and even against the purview of their ethical beliefs. What is sublime can
in fact overcome the distinction between beautiful and ugly, and aims
instead at offering powerful, uncontrollable emotions:
For grandeur (" #$) has the effect of transporting its audience rather
than persuading it; and anything amazing and astonishing (% &' . . . !
( ) always prevails over what is merely persuasive and pleasant. The fact
is that persuasion is generally something we can resist, whereas these other effects
exert an irresistible power and force (
)
) and overcome
every hearer.
This incisive analysis of the effects of sublimity is developed in the tract

On the Sublime (* +, ) which, not long before Seneca composed
his tragedies (or perhaps even at the same time), dealt with the notion
of sublimity in a particularly influential fashion.122 There is little to be
121
In general see Schiesaro (1985), with further bibliography.
122
Subl. 1.4.
128
gained in exploiting On the Sublime as a source for Senecas conception

of tragedy, and not just because its elusive chronology would make such a
strategy risky, at best.123 But it is surely fruitful to turn to this work in search
of a contemporary analysis of the sublime, and to refer to it in the present
attempt to ground in specific forms of behaviour and expression Atreus
nature as a sublime character endowed with a consummate excellence of
language (1.3: &' ). The comparison with On
the Sublime, of course, is made all the more pertinent by Atreus distinctive
metadramatic role: as a poet on stage, obsessed with the plotting and mise
en scène of his own play, Atreus develops a coherent and articulate poetics,
one which centres on the unrestrained power of poetry over its creator and
its audience alike.124
If the messengers simile between Atreus and a lion thus acquires an
intriguing metadramatic overtone, a deeper connection can also be established between Atreus artistic project as a whole and the intrinsic nature
of sublime poetry as articulated in On the Sublime. Among the natural
(8.1: -.) sources of sublimity, Ps.-Longinus lists first the power
of great thoughts (8.1: ! " /) ), and second strong and inspired emotion (8.1: ! ! & !
( ). Indeed, nothing contributes more to greatness of expression than
authentic emotion at the right moment, as if some frenzy or divine inspiration animated the words (8.4: #!
!
& & ), filling them, as it were, with the divine breath of
Phoebus (8.4: )(0 ). The sublime is, in sum, the echo of a great soul
(9.2: ). I recall here my observations about
in chapter one, especially in connection with Cleanthes
aspirations to a form of poetic expression that could aptly convey divine
greatness (. ).125 There I tried to show that any attempt to rein
in the potentially disruptive force of poetic enthousiasmos and reconcile it
safely with the Stoics stated goal of a morally instructive poetry is intrinsically doomed to failure. Atreus can now offer a case study of a sublime poet
in action, one who allows us to glimpse not only, as it were, the finished
product, but also the creative stages that bring it to life.
Atreus passion is intense, grandiose and, to borrow from Ps.-Longinus,
distinctly ) (10.6). He explicitly declares his intention to scare
off a very special sector of his audience, the gods themselves, with his
123
124
On the sublime in Seneca valuable general indications are offered by Michel (1969). I am not sure
that the connection is invalid just because in his prose Seneca would not admit that he is trying to
move his readers to ekstasis (Subl. 1.4, cf. 15.9), as Traina (1987) 123 argues.
125 SVF 1.486. Cf. Mazzoli (1970) 47. See above, p. 23.
See above, ch. 1.
A craftier Tereus
129
extraordinary nefas126 not to mention the fact that the direct and indirect witnesses of his sacrifice (the messenger, the chorus, Thyestes himself )
react to his actions with unrestrained terror.127 An element of the intense
emotions that Atreus is capable of stirring is undoubtedly connected with
the very violence of his own passions. Ps.-Longinus offers, in connection
with Sapphos fragment 31 Voigt, a compelling analysis of the impact of
unrestrained passions, especially of those that catalyze different emotions
at once: Are you not astonished at the way she summons up all together
mind and body, hearing, tongue, sight, colour, as though they were separate elements external to her, and feels contradictory sensations, freezes and
burns, raves and reasons (after all, she is terrified or even on the point of
death), as if she wanted to display not one single emotion, but a complex
of emotions (Subl. 10.3).
Ps.-Longinus connects sublimity with a specific attitude entertained by
the poet about his past and his future. The poet who aspires to sublimity
is characterized by his deeply agonistic relationship with his models, which
he should imitate and emulate (13.2:
and 0) as he constantly
concerns himself with the judgement of posterity: If I write this, how might
posterity judge it? (14.3). Atreus entertains precisely the same concern and
has a trenchant answer ready: age, anime, fac quod nulla posteritas probet, |
sed nulla taceat (up, my soul, do what no coming age shall approve, but
none forget, Thy. 1923). Atreus programme of poetic imitation is indeed
predicated on
and 0 vis-à-vis his models. Not only is he
fully aware of the pertinent models for his own endeavour, and explicitly
turns to them for inspiration (Subl. 13.2), but he also insists on competing
with them, on trying to surpass their evil with an increasingly original
nefas of his own. According to Ps.-Longinus striking image this inspiration
drawn from past models impregnates the poet just as the divine wind
penetrates the Pythia: many [poets] are possessed by a spirit not their own,
just as (so the story runs) the Pythia at Delphi sits on her tripod near a
cleft in a ground which (so they say) breathes out a divine vapour, and is
thereby made pregnant (& ) by the supernatural power and is at
once inspired to prophecy. Likewise, from the genius of the old [writers]
a kind of effluence ( ) from those holy mouths flows into the
souls of their imitators (13.2). As vates, the sublime poet operates at the
critical juncture between overwhelming inspiration, prophecy and poetic
creativity. To achieve these heights of inspiration and poetry he must go
126
127
2656: fiat hoc, fiat nefas | quod, di, timetis.

The messenger at 6348; the chorus at 744, 789884; Thyestes at 92069 (before the revelation).
130
beyond the closed boundaries of his masculine self, and be pervaded by an

irresistible outside force precisely as Atreus does: tumultus pectora attonitus
quatit | penitusque voluit; rapior et quo nescio, | sed rapior (Thy. 2602).128
Atreus obsession with maius nefas is directly connected with his agonistic attitude towards tradition, and is but one aspect of his sublime nature,
always in search of higher pursuits and stronger emotions, constantly obsessed with excess, with what is more and bigger. The maius-motif 129
pervades his reflections in act 2; to him nullum [sc. facinus] est satis (no
crime is enough, 256); his animus pushes him to accomplish nescioquid . . .
maius et solito amplius (something greater, larger than normal, 267); pain
forces him to devise a revenge bigger (maius, again) than the one meted out
to Tereus (2725). His never-ending search for maius is consistent with the
ideology of tyranny. The tyrant constantly hungers to escape limitations, to
ignore sufficiency and moderation. A relevant statement on the subject can
be found in one of the most interesting literary debates on tyranny, the exchange between Jocasta and Eteocles in Euripides Phoenissae. A distraught
Jocasta is firm in her belief that to the wise man what is adequate is always
enough (554: & ( 1 21 3 " . 4 ), and that the
search for ! (advantage, literally more) is the pursuit of a mere
name (553). But Eteocles had already made it clear that striving after more
is a given which does not require (nor indeed allow) any explanation: it is
not manly (
) to lose more and settle for less, being happy with
5 (less) when it is possible to have more, ! (50910).
The search for ! , maius, can never cease; it is the prime motivator
and ultimate goal of the 6 of the ethics and aesthetics of
tyranny.130
On the Sublime does not confine its analysis to the psychological tension
underlying the poetics of sublimity, but takes into account a number of
particularly representative techniques of expression that are related to it.
A sublime style reveals itself both in specific arrangements of the subject
matter and in a series of rhetorical tropes. Among the former, particular
consideration should be devoted to 5', the ability to gather a number
of details and present them as a compelling whole (Subl. 12.2). In his
speech in act 2, for instance, Atreus overwhelms the counsellor thanks
also to the elaborate accumulation of details which reinforce the vividness
of his plot. Phantasia, too, is a principle that Atreus-the-author would
readily embrace. Phantasia, Ps.-Longinus explains, occurs when moved by
128
130
129 See p. 31, n. 16.

See above, pp. 51ff.
See Mastronarde (1994) 303 on the discussion in Platos Gorgias and Republic about the tyrant as
the supreme example of the , which is presupposed by these lines.
A craftier Tereus
131
passionate emotion, you seem to see the things of which you speak, and
place them before the eyes of your audience (15.1). Atreus is fully aware of
the effectiveness of this process even as he is still in the planning phase of
his revenge. The strength of his inspiration and the vividness of his project
are such that a full picture of the imminent slaughter is already available
to him: already before my eyes flits the whole picture of the slaughter; his
lost children heaped up before their fathers face (Thy. 2813: tota iam ante
oculos meos | imago caedis errat, ingesta orbitas | in ora patris).
The reader of On the Sublime need perhaps go no further than the first
few lines of Atreus initial monologue (17680) in search of distinctive
features of the sublime:
ignave, iners, enervis et (quod maximum
probrum tyranno rebus in summis reor)
inulte, post tot scelera, post fratris dolos
iratus Atreus?
Undaring, indolent, nerveless, and, what in important matters I consider a kings
worst reproach, unavenged, after so many crimes, after a brothers treacheries, and
breaking every law, you are busy with idle complaints is this Atreus in a rage?
This period is strongly marked by asyndeton, a stylistic device which raises

the emotional pitch of the sentence and, as Ps.-Longinus explicitly indicates,
is one of the hallmarks of the sublime, just like interrogations and selfinterrogations, which are another device frequently employed by Atreus.131
Significantly, the whole structure of this period an extended self-addressed
question falls within the technai of the sublime listed by Ps.-Longinus.132
The three adjectives that Atreus uses at line 176 to describe his behaviour
so far are all found in rhetorical and literary contexts. Ignavus, as we glean
from Horaces Epistles,133 can be used of an indolent, slow style, similar
to the type of compositio which Quintilian will define as tarda et supina
(slow and languid, 9.4.137). Its Greek counterpart, argos, plays a significant
131
132
133
On asyndeton in Senecan tragedy: Canter (1925) 169ff. and Billerbeck (1988) 1223. The Auctor
ad Herennium (4.41) provides an apt description of the devices effects: this figure (asyndeton) has
animation and very great force, and is suited to concision (hoc genus [sc. dissolutum] et acrimoniam
habet in se et vehementissimum est et ad brevitatem adcommodatum). See also Quint. 9.3.54 and
Calboli (1993) 3702, esp. n. 178. Self-apostrophe occurs at 192 (age, anime).
Subl. 18.12: the impassioned rapidity of question and answer and the technique of making an
objection to oneself make the passage, by virtue of its figurative form, not only more sublime,
but more convincing. For emotion carries us away more readily when it seems to be generated by
the moment rather than deliberately assumed by the speaker, and the self-directed question and
answer represent the momentary quality of emotion. On the technique see Canter (1925) 140ff.
and Billerbeck (1988) 123.
Epist. 2.1.67, with Brink ad loc. for further references.
132
role in Ps.-Longinus comparison between Hyperides and Demosthenes

(Subl. 34.4):
Yet Hyperides beauties, numerous as they are, are without grandeur: inert (()
in the heart of a sober man, they leave the audience at peace. Nobody is afraid
when he reads Hyperides. But as soon as Demosthenes begins to speak, he gathers
to himself the faculties of true genius in their highest form the intensity of
lofty speech (#,
), vital emotion, abundance, variety, speed where
it matters, all his unapproachable force ( ) and power ( ).
Horace will call versus . . . inertes (Ars P. 445), lifeless lines, lines virtute
carentia (Epist. 2.2.123) as boring as the carmen iners with which, Calpurnius
Lycidas will complain, a rival has inexplicably wooed his beloved Phyllis
(Ecl. 3.5960). Enervis (feeble) belongs to the vast repertoire of anatomic
and physiological metaphors we find in Latin literary terminology; Cicero,
for instance, relates enervis to mollis, weak (Tusc. 4.38); Quintilian relates
it to effeminatus, unmanly (9.4.142).134
Note, however, Tantalus description of his father as Thyestes enters the
stage for the first time in act 3 (Thy. 4212):
pigro (quid hoc est?) genitor incessu stupet
vultumque versat seque in incerto tenet.
My father (what is it?) moves with slow step as if in a daze, keeps turning his face,
and is mired in doubt.
The ensuing dialogue repeatedly contrasts the huge difference between

Atreus determined enthousiasmos and his brothers uncertainty (tinged with
hypocrisy) as he extols the virtues of measure, exile and modest living
vis-à-vis the false wealth of power.
Atreus will promptly overcome his initial weakness, and go on to embody
a stylistic and behavioural model grounded in energy, speed and determination. It is a model which Cicero already recognized in Accius Atreus,
and one he recommends himself to the speaker who wants to convey vis
with words.135
134
135
Enervis also occurs in Tac. Dial. 18.5, on which see Gudeman (1914) 31819. More generally, Bramble
(1974) 358 (esp. 36, n. 3), with further bibliography.
De or. 3.219 energy (must take) another kind of tone, intense, vehement, eager with a sort of
impressive urgency (aliud [sc. vocis genus . . . sumat] vis, contentum, vehemens, imminens quadam
incitatione gravitatis). There follows a quotation of Accius Atreus 198201 Ribbeck2 = Dangel
2932 (quoted below, p. 142), where the anaphora of iterum and the polyptoton maior/maius clearly
contribute to the stylistic strength Cicero wants to exemplify.
A craftier Tereus
133
iii
In his polymorphous manifestations Atreus tramples the boundaries between different realms. As author, director and spectator, he occupies the
entire scenic space, with all its possible functions and points of view. As
both victim and executioner, he lays claim to the audiences sympathy for
the wrongs he has suffered while also eliciting their horror at the intensity
of his revenge. As the protagonist of the play, he parades an unrivalled
capacity to use language creatively and metaphorically to ensnare his victims. He enjoys the prerogatives of masculine political power, but he does
not hesitate to rely on his feminine inner self as he yields to passions and
allows himself to be carried away by inspiration, as he deceives and entraps
his enemy. His knowledge of the literary tradition establishes beyond any
doubt his metadramatic credentials, just as his insistence on the sacrificial
nature of the slaughter that he performs seems to secure his divine status
as a man who behaves like a wild animal and shines like a god.136
What surfaces from a synoptic analysis of all these tightly interconnected
aspects is not simply Atreus extraordinary complexity nor even his ability to
unify the tensions between drama and metadrama, between the illusional
power of the poetic word and the harsh impact of the reality principle
realized in his all-too-real revenge. Atreus, all-encompassing, all-powerful,
ecstatic, embodies the sublime in its ultimate, Dionysiac incarnation. If
Dionysus Euripides Dionysus above all is the god of theatre, Atreus
can be considered his (super)human counterpart, endowed with many of
the same alluring ambiguities and irresistible attractions.
The analogy extends to fundamental aspects of Atreus persona. As gods
of theatrical mimesis both Atreus and the protagonist of Bacchae act as
playwrights on the stage and control the unfolding of the dramatic action.
Both Atreus and Dionysus import into the tragedy comic elements which
not only enrich their expressive repertoire, but also prove invaluable in the
battle against less articulate opponents such as Thyestes and Pentheus.
Atreus Dionysiac overtones are established early in the play as an effect of
the intertextual connection with Ovids Procne, whose revenge takes place
in the frenzy of trieterica Bacchi.137 Appropriating the wounded persona of
the betrayed queen, Atreus also shares in the violent revenge that can be
136
137
Boyle (1983b) 212.

Ov. Met. 6.587. Ovid follows Accius less usual version of the myth, which places the story in a
Thrace converted to the cult of Dionysus; Dangel (1995) 3467. The setting may well go back
to Sophocles; Ciappi (1998) 439. On the connections between the sacrifice of Thyestes and the
Dionysiac dimension of the Procne story see Burkert (1983) 1812.
134
understood only in the context of such wild rites. Similarly, when Atreus
confesses to the unexplained and overwhelming power of the inspiration
that has taken him over,138 he transcends the normal limits of human action
and partakes of the irrational excess of the god of furor, Dionysus-Bacchus
himself. Thyestes, like the Euripidean archetype of Dionysiac tragedy,
Bacchae, is (among other things) a tragedy of revenge and of familial bonds
ignored, overturned and destroyed. Thyestes may not be aware of what he
is doing, but Atreus certainly is when he specifies that the most hideous
aspect of his revenge be carried out by him. Like Agave, Thyestes is effectively blinded (wine fuddles his mind) and forced to become (at least
symbolically) the killer of his own offspring. Finally, the crime committed by Atreus is similarly coloured by the religious overtones of a rite, of
a novel ceremony in honour of a novel god, Atreus himself, who is even
more controversial than his Greek counterpart.
Atreus, like Dionysus, plays a crucial metadramatic role, appearing as
the consummate manipulator of words, knowledge and emotions, and
overpowering all others. Much of Atreus and Dionysus power resides
in their ability to introduce into the tragic text a dimension of skilled
irony, manipulation and disguise which other characters are unequipped to
understand, and by which they are inevitably trapped. Pentheus is baffled
by this different form of communication and is consistently taken in by it.139
In the second episode of Bacchae, in particular, Dionysus double entendres
exploit the same linguistic ambiguity that will serve Atreus so well in his
successful attempt to tease and deceive his brother.140 The god, we are
reminded, is a sophos (Bacch. 656), whose knowledge far exceeds that of
uninspired mortals such as the king of Thebes. We realize now that the
comic elements play a very significant role in the articulation of the plays
meaning, because they strike at the core of what sophia really is. One of
the recurrent themes of Bacchae is precisely that those who appear to be
foolish (Dionysus and his followers) are actually wise, and the wise are
ultimately devoid of sense.141 Puns and double entendres bring home this
fundamental opposition and its paradoxical resolution.
138
139
140
141
Lines 26778, on which see above, p. 130. It may be worth pointing out the comparable structure of
the exchange between Cadmus and Agave at Bacch. 12812 (KA. -!
(. | AG. 7 . . .) with Thy. 10056: AT. natos ecquid agnoscis tuos? | TH. agnosco fratrem.
Segal (1982) 230 aptly labels him an authoritarian literalist.
See for instance the gods remark on Pentheus significant name at Bacch. 508. Pentheus, however,
does recognize Dionysus linguistic prowess: 8 % 7 )( - (491).
See 47980.
The topic is particularly prominent in the first episode; see especially 196, with 269, 326, 332,
6559.
A craftier Tereus
135
While the comparison between Thyestes and Bacchae is especially significant at a symbolic and functional level, some further thematic affinities
are worth mentioning. Hunting imagery plays an important role in both
plays. Agave and her fellow Bacchants literally hunt down Pentheus until
he is ripped limb from limb (7313, 977, 118991). The outcome of Atreus
hunting will be no less devastating for being almost entirely psychological:
the traps he has deployed against Thyestes will indeed yield the desired,
bloody result: the beast is caught in the nets I placed; I see both him and,
joined together with him, the offspring of the hated race I see (Thy. 4913:
plagis tenetur clausa dispositis fera: | et ipsum et una generis invisi indolem |
iunctam parenti cerno) parallels Dionysus reference to Pentheus at Bacchae
848, the man is falling within the cast of the net (/6 & )

).142 While in Seneca Thyestes definition as a fera is without direct consequences, in Euripides the Bacchants attack Pentheus because they
mistakenly believe that he is a lion (98991). The lion, however, is one of
the animals traditionally associated with Dionysus143 and indeed, when he
is captured by Pentheus soldiers, he is presented as a wild beast (436: ).
Dionysus thus shifts this animal quality onto Pentheus, using him as a
scapegoat, a process which is parallel to the one whereby Atreus, the really
feral cannibal, ultimately casts Thyestes in the role of a bestial man-eater.
Also, the belated anagnorisis of Thyestes can be compared with Pentheus
equally ineffectual anagnorisis in Bacchae. There the king finally acknowledges his past errors (11201), and realizes that his end is close (1113), but
to no avail: the divine force of Dionysus should have been recognized and
obeyed earlier, just as Thyestes ultimate understanding of Atreus character
and intentions agnosco fratrem only underscores his previous intellectual failure. Agaves recognition of her own deeds, too, is tragically belated
(1345).
The knowing smile of Dionysus is an apt emblem for Atreus, too, as
he contemplates from a superior vantage point the extent of his success.
Equating himself with the gods, Atreus becomes a veritable god of tragedy,
the presiding icon of the metadramatic manipulation staged in Thyestes. As
Dionysus precipitates Pentheus death in a sort of play within the play144
which he himself has authored, so does Atreus plot and enact his revenge
over Thyestes.
Common to all aspects of Atreus superiority over Thyestes is his ability
to play at the same time from different scores, to undermine Thyestes
142
143
144
On hunting and nets see also 231, 451, 1021 and passim.
Bacch. 1019, with Dodds (1960) 205 and xviii.
Foley (1980) 109. I am indebted to Foley for many insights on Bacchae.
136
certainties, and to assuage his latent fears by switching unpredictably

between codes which would normally be considered mutually exclusive. Alongside Thyestes utterly tragic, and fatally doomed, monodimensionality,145 Atreus displays a huge range of behaviour, which is ultimately
the key to outmanoeuvring his brother.
At times Atreus takes issue with the boundaries of generic affiliation,
and infiltrates into the tragedy a distinctly comic tone.146 We have already
insisted upon Atreus double role as author and actor. Although Atreus,
needless to say, is not another Plautine servus currens,147 as he directly addresses the audience in impassioned asides148 he comes very close to the
conspiratorial attitude that several Plautine protagonists assume vis-à-vis
their public.149 The crossing of boundaries thus accomplished is at least
twofold, since what is at stake is not only the generic categorization of the
play, but also the social status of the protagonist a king who abandons all
sense of propriety and whose behaviour on stage recalls, of all things, that
of cunning, comic slaves. Pace Cicero, who decreed that comic elements in
tragedy, and tragic in comedy, are inappropriate, 150 comic elements can become striking signifying strategies in tragedy, highlighting with their ironic
contrappunto the fatal ignorance of certain characters, and creating opportunities for emotional release which bond the audience with the characters
who control irony (a strategy famously not lost on Shakespeare).
A similar manipulation of genre-specific codes underlies the final scene of
the play. Setting the tragedys denouement at a banquet precipitates a generic
short-circuit, which further destabilizes the audiences expectations.151
Banquets and food play a prominent role in comedy, and the texts attention to Thyestes bodily functions (his untragic burping, 911)152 activates the
145
146
147
148
149
150
151
152
I borrow the term from Foley (1980) 122.

On comic and tragic see Silk (2000) 5297; on Euripides and the comic see briefly Silk (2000) 51;
Seidensticker (1978); Gredley (1996); and Taplin (1996). Specifically on Bacchae, and the liminality
of genre of Dionysus in the play, see Segal (1982) 2546.
Ulixes language in Tro. 61314 is indeed reminiscent of clever comic slaves; see Boyle (1994) 190.
Such servi could in turn evoke lofty mythological models for their enterprises; see Plaut. Bacch.
925; Pseud. 1063, 1244 with Fraenkel (1960) 912.
On asides see Tarrant (1978) 237, who points out that they seem to belong to fourth-century tragedy
as much as to comic theatre. On the importance of asides in the latter see Duckworth (1952) 10914.
In turn, clever slaves in Plautus are eager to appropriate tragic or epic models for their exploits. See
again Chrysalus canticum at Bacch. 92578, modelled on Iliou persis, but with likely borrowings
from tragic language; Fraenkel (1960) 5763, with Norden (1927) 370. See also Pseud. 1063, 12434
(all prepared by 524, 584).
De optimo genere oratorum 1.
A deipnon concludes Plautus Stichus (73972), as well as Aristophanes Lysistrata, Acharnians and
Birds. See Duckworth (1952) 380.
Meltzer (1988) 315 with Dover (1968) 70. On burping in Plautus see Pseud. 1295, 1301. See also
Manilius 5.462 ructantemque patrem natos (referring to Thyestes), with Aesch. Ag. 1598601.
A craftier Tereus
137
comic intertext. At the same time, of course, we are bound to perceive the
banquet as the inevitable last step in the elaborate sacrifical ritual which had
structured Atreus killing of Thyestes children.153 The sudden appearance
of a drunken Thyestes on the stage catalyzes the tragic irony of the dramas
final moments, as it displays a character desperately unaware of the looming disaster. Moreover, the comedy implicit in this presentation ultimately
denies Thyestes the compassion that such a character could otherwise elicit.
The generic boundary-crossing promoted by Atreus is yet another tool in
his unequal fight against Thyestes, because he has been able to manipulate the literary code, once again, to his exclusive advantage. He has taken
up the ethos of a comic hero, better to deceive his brother, and has organized a banquet which takes the normal comic obsession with food and
warps it into its most gruesome and painful opposite. Thyestes, unable as
usual to comprehend his brothers ingenuity, is completely deceived, and
his belated, rather weak intimations of uneasiness are drowned out by the
loud incongruity of the scene. By evoking a comic intertext which Thyestes
has failed to suspect, Atreus effectively invites the audience to relinquish
empathy for Thyestes, and to replace these tragic emotions with a sense
of physical disgust and moral detachment which strips Thyestes of any
remaining sense of dignity.
Equally interesting in this context is Atreus apparently friendly request
that his brother change the exiles shabby attire for robes worthy of his newly
regained royal status: take off your foul garments, spare my eyes, and put
on ornaments equal to mine (5246: squalidam vestem exue, | oculisque
nostris parce, et ornatus cape | pares meis).154 Changing clothes, a frequent
event in comedy,155 only ends in disaster for Senecan tragic characters,
especially when they trade upwards.156 Agamemnon relinquishes his military garments and accepts Clytemnestras robe just before he is murdered
(Ag. 8814).157 In Troades, Helen treacherously persuades Polyxena to dress
for her wedding while she is actually being driven to her death (8835).158 In
Thyestes we can observe the same interplay of irony and doom, as the final
touches to Thyestes new outfit pave the way for one of Atreus most chilling
double entendres: wear the crown set on your reverend head; I will offer to
the gods the destined victims (5445: imposita capiti vincla venerando gere; |
153
154
155
156
157
On the sequence sacrificeextispiciumbanquet Tarrant (1985) 198 compares Ov. Met. 15.1309.
The importance of this aspect has been pointed out by Erika Thorgerson in an unpublished seminar
paper (Princeton, 1994).
On the metadramatic implications of robing see Segal (1982) 223.
See Tro. 883, Ag. 8813, with Tarrant (1985) 165.
158 See Fantham (1982) 341.
See Tarrant (1976) 339.
138
ego destinatas victimas superis dabo). There can be little doubt that the
text orients the audience towards an ironic evaluation of these details,
if only because of the explicit caveats offered by the chorus in the ode
preceding this scene, which Thyestes is conspicuously failing to heed:
a king is not made by riches, nor by the colour of a Tyrian garment,
nor by the royal mark of honour on his head, regem non faciunt opes, |
non vestis Tyriae color, | non frontis nota regia (3446). In act 5, as a drunken
Thyestes is suddenly overcome by anguish, his royal garments now dishevelled, the possibility that the audience may share his emotions is further
reduced (90910, 947). Bacchae, too, offers a striking example of this fatal
connection: when, at line 842, Pentheus unwittingly agrees to dress as a
woman, he not only elicits an ironic reaction of which he is utterly unaware, but also takes a decisive step towards his own demise.
Atreus shares with Dionysus the superior power that derives from their
being in control of the dramatic strategies enacted on stage. Unlike Thyestes
and Pentheus, they control events because they devised the plot and set it
in motion; they are not only passive actors, but also crafty authors.159 In
the manipulation and transgression of boundaries that shape human society
and literary expression, Bacchae and Thyestes reveal both the artificiality and
the strength of those delimitations.160 Both plays force their audiences into
a complex negotiation of conflictual emotions, offering them the vision of
an exhilarating freedom and at the same time of the horrific extremes that
freedom could provoke.161
159
160
161
Foley (1980).
A tentative connection could be established between Thy. 1034 (sentit introitus tuos | domus et
nefando tota contactu horruit) and the earthquake that shakes Pentheus palace at Bacch. 58692.
The contexts are clearly different, yet the notion that the royal palace metaphorically shatters when
(Bacchic) furor enters could perhaps be related.
A tension poignantly captured in Bacch. 861, where Dionysus is called , 4 1
94 .
chapter 4
Atreus rex
non quis, sed uter

dignum est Thyeste facinus, et dignum Atreo
i
Despite its title, Thyestes is of course a play about Atreus, whose fundamental role in articulating the plot is matched by his consistently overpowering presence on stage. The designation of Atreus counsellor as satelles
is metaphorically most fitting: other characters revolve around the largerthan-life royal protagonist with the limited, virtually non-existent autonomy of satellites locked in a gravitational field that they cannot control.1
The counsellor voices his feeble resistance as Atreus plot is already marching along briskly; the chorus is feeble and unable to affect, at times even to
understand, the irresistible progress of the revenge.2 And Thyestes, too, for
all his aspirations, most often appears to be the necessary but hardly selfdetermined complement to his brother. After all, he is lured into a carefully
organized trap, and his every reaction, practical as well as psychological,
has been successfully gauged and pre-analysed by Atreus.
The unquenchable enmity between the two brothers only casts their
blood-bond into sharper relief. Indeed, Atreus conjures up an image of his
brother that virtually mirrors himself an image that the chorus finds plausible. This elusive yet powerful bond adds significantly to the disturbing
appeal of the play: because they know each other so deeply, and because we
can only glimpse the nature and depth of their relationship, both Atreus
1
Note that the word, even in its basic meaning of escort or attendant does often carry a rather
negative connotation (OLD s.v. 1), and can also be used in the fully negative sense of accomplice in
crime (with gen.: OLD s.v. 2).
For more details on the choruss attitude see below, pp. 16476.
139
140
and Thyestes are never polar opposites, representing two well-defined sides
of an ethical debate. The more we delve into the details of Senecas characterization, the more we are able to appreciate the complex, nuanced and
often intrinsically contradictory personalities of both brothers.
As it consistently tries to sustain the contrast between them, the play
devotes remarkable attention to the potential deconstruction of this opposition, a process that should come as no surprise after what we have
repeatedly observed about the complex structure of the tragedy, and the
ethical implications of that structure. Seneca offers a far from univocal image of Thyestes: he is a character whose loudly proclaimed moral aspirations
fail to assert themselves with the required degree of conviction, a character
on whose conduct past and present the play casts heavy shadows of doubt
and uncertainty. But even more pervasive, and certainly more disruptive in
the linear development of the play, is the suggestion that, in a sense, the
roles of the brothers could have been interchangeable, that had it been
his turn Thyestes revenge could have been every bit as gory as the one
that Atreus happens to be plotting; that finally their different roles in
the tragedy are predicated on a specific series of actions and counteractions,
but not on an essential moral difference. I will insist on this last aspect first.
The audience is immediately informed of Thyestes crime by Atreus at the
beginning of his speech in act 2 (17680; quoted above, p. 131):
ignave, iners, enervis et (quod maximum
probrum tyranno rebus in summis reor)
inulte, post tot scelera, post fratris dolos
iratus Atreus?
Undaring, indolent, nerveless, and, what in important matters I consider a kings
worst reproach, unavenged, after so many crimes, after a brothers treacheries, and
breaking every law, you are busy with idle complaints is this Atreus in a rage?
I will try to show later what these scelera actually are; what matters at this
point is that after the prologues announcing of the triumph of furor and
nefas in the house of the Pelopidai, it is Thyestes scelera, dolos and fas
ruptum that are foregrounded. The fact that an obviously partisan source
voices these accusations does not detract from their impact. Atreus is clearly
not trying to justify his behaviour; if anything, he is complaining about
his tardiness and lack of resolve. Even more importantly, nowhere in the
rest of the play are these accusations rebuked: they stand unchallenged,
notwithstanding Atreus display of cruelty. Far from setting a supposedly
Atreus rex
141
moral Thyestes against his monstrous brother, the tragedy constantly insists, through Atreus viewpoint, on the moral affinity, even the potential
equivalence, between the two men.
The complex family history of the Pelopidai gave a certain prominence
to Thyestes crimes. He had seduced Atreus wife, Aerope, and possibly
polluted his brothers family line.3 Atreus, at the beginning of the play, is
determined to take revenge for such crimes, and he repeatedly hints at the
fact that he suspects Thyestes of preparing a revenge as cruel as the one he
has in mind (1936):
aliquod audendum est nefas
atrox, cruentum, tale quod frater meus
suum esse mallet scelera non ulcisceris,
nisi vincis.
I must dare some atrocious, bloody crime, such that my brother would have wished
it to be his own you dont avenge crimes if you dont surpass them.
By insisting on the notion of revenge, the play underlines the circular,

repetitive nature of the conflict between brothers and of the play which
portrays it. What we are about to see is merely another round in an endless
cycle of Aeschylean revenge and counter-revenge.4
In fact Atreus is constantly worried at the thought that his brother might
strike first with comparable cruelty (2014):
proinde antequam se firmat aut vires parat,
petatur ultro, ne quiescentem petat.
aut perdet aut peribit: in medio est scelus
positum occupanti
Therefore, before he grows in strength and readiness, let him be attacked first, lest
he attack me while I am at peace. He will either kill me or die; the crime is there,
ready for him who will seize it first.
The awareness that a similar crime exceptional as it appears could have

been devised, or is being devised by Thyestes is crucial to understanding
Atreus psychology (31416):
istud quod vocas saevum asperum,
agique dure credis et nimium impie,
fortasse et illic agitur.
3
The motif is well developed in Greek tragedy: Aesch. Ag. 11923; Eur. frs. 4669 Nauck2 , El. 699728.
For possible reconstructions of Ennius Thyestes and bibliography on earlier treatments of the myth
see Jocelyn (1967) 41219. On earlier treatment of the mythical plot see Marchesi (1908); Lesky
(192223); Lana (195859); Lefèvre (1976) 227.
See now esp. Kerrigan (1996).
142
What you call cruel and harsh, and think is being done ruthlessly, regardless of
right or wrong, perhaps is being done there too.
Such a reassured awareness of the interchangeability of their roles and

their reactions seems to justify Atreus self-description as (so far) innocens
(27981):
bene est, abunde est: hic placet poenae modus
tantisper. ubinam est? tam diu cur innocens
servatur Atreus?
This is good, plentifully so. I like this way of punishing him, at least for now.
Where is he? Why does Atreus maintain his innocence for so long?
It is not easy to determine to what extent previous Roman tragedies dwelt

upon Thyestes own faults, yet it is clear that two traditions concerning his
return to Mycenae coexisted:5 according to Accius,6 who is here following
Aeschylus,7 Thyestes returns of his own accord in order to take revenge on
his brother:
iterum Thyestes Atreum adtractatum advenit
iterum iam adgreditur me et quietum exsuscitat:
maior mihi moles, maius miscendum est malum,
qui illius acerbum cor contundam et comprimam
Once again Thyestes comes to attack Atreus; once again he approaches to rouse
me from my calm. Bigger is the danger, bigger the evil I must stir up to crush and
crunch his cruel heart.
From a verse by Varius we can glean a similar scenario:8

iam fero infandissima,
iam facere cogor.
now I suffer unspeakable evils; now I am forced to commit them.
While Seneca ostensibly espouses an alternative version of the mythical plot

and makes Thyestes return contingent on Atreus deceitful invitation, he
transforms Accius (and presumably Varius) version into a powerful subplot
which substantially affects our perception of the events. His Atreus emerges
as an unusually nuanced and composite character, who is neither the quiet,
5
6
7
8
Jocelyn (1967) 414; Tarrant (1985) 120.

198201 Ribbeck2 = 2932 Dangel. On this passage and its connection with Senecas Thyestes see
Marchesi (1908) 86; Lana (195859) 31617; Cipriani (1978); De Rosalia (1981) 2256; Dangel (1995)
277.
Ag. 15878, see Jocelyn (1967) 414, n. 1.
Fr. 1 Ribbeck2 . Note that Quintilian introduces Varius lines as an example of audacious conduct
spurred by indignatio rather than malitia (3.8.45). On the reconstruction of Varius play see Lefèvre
(1976) and Leigh (1996).
Atreus rex
143
potentially passive target of Accius nor an irrational or deranged iratus. Or,

more precisely: the intertextual memory of Thyestes criminal actions and
intentions forces us to wonder whether Atreus relentless obsessions are not
after all justified. The alluring force of the play resides precisely here: Atreus
reasons according to a symmetrical logic akin to the epistemic protocols
of the unconscious when he identifies his brother with himself, but his
intimations, far from being dismissed, are actually corroborated by the
chorus.9 Indeed Atreus sees the outside world as a projection of his own
self he is above anyone else: cunctos super (885), and does not recognize
the independent will of others: quod nolunt velint (what they dont want,
let them want it, 212).
The implications of Senecas strategy are lit up by the fact that Accius
fragment is emphatically recalled not only at Thyestes 202 (quiescentem),
but also in the insistent maius-motif 10 which lies at the core of Atreus
programmatic statements11 and had already been introduced by the Fury in
the prologue (Thracium fiat nefas | maiore numero).12 The allusive gesture
paradoxically highlights Senecas departure from Accius: this time the Fury
is responsible for inspiring Atreus plans, which are no longer exclusively
(or even primarily) contingent on Thyestes own intentions.
Atreus depiction of his motives is essential to our understanding of the
play. We cannot read Atreus as a larger-than-life monster without paying
at least some attention to his claims that his cruelty was provoked, and that
he is taking revenge for heinous crimes. Thyestes crime polluted Atreus
blood, and he is fittingly punished by drinking his childrens blood. There
are no predetermined roles or certainties allotted. (In his own philosophical
writings, Seneca observes that slaves should be treated decently because,
among other factors, a sudden and unexpected turn of events can easily
turn free men into slaves.13 )
An equally upsetting sense of shifting boundaries casts its shadow over
the actions of Thyestes. Depending on unpredictable events, Atreus could
end up as a quiet victim, viciously wronged and subsequently destroyed
by his evil brother, or as a man whose justified awareness of the injuries
9
10
11
12
13
Symmetrical logic, or bi-logic, is a concept formulated by Matte Blanco (1975) and (1988). On
projective identification, a concept originally developed by Melanie Klein, see Matte Blanco (1988)
1035 and passim. Note that Atreus himself cannot escape the deflagration of the boundaries of the
self which he advocates: he fills up Thyestes with the body of his children (8901: implebo patrem |
funere suorum), but he is himself filled up by the external force of divine inspiration (2534: impleri
iuvat | maiore monstro).
A term introduced by Seidensticker (1985).
See 254 (maiore monstro), 267 (nescioquid . . . maius) and 2745 (maius hoc aliquid dolor | inveniat).
Lines 567. See above, p. 27, on the programmatic importance of iterum.
Letters to Lucilius 47.10.
144
he has endured pushes him towards a revenge which must necessarily be

exaggerated and perverse.14
In the second part of the tragedy, once Atreus plans have been meticulously and successfully realized, he turns his suspicions into a mocking
reproach to his brother, who, given the opportunity, would have done exactly the same. This satisfying thought is first voiced in the monologue in
which Atreus contemplates the completion of his scheme (91718: mixtum
suorum sanguinem genitor bibat: | meum bibisset, his sons mingled blood let
the father drink; he would have drunk mine), and is then communicated
to Thyestes himself at the very end of the play (110410):
scio quid queraris: scelere praerepto doles;
nec quod nefandas hauseris angit dapes:
quod non pararis! fuerat hic animus tibi
instruere similes inscio fratri cibos
et adiuvante liberos matre aggredi
similique leto sternere hoc unum obstitit:
tuos putasti.
I know what you complain of: you are sorry that this crime has been pre-empted;
nor do you grieve that you have swallowed unspeakable foods: just that you have
not prepared them yourself for me. This had been your plan, to prepare the same
banquet for their unwitting father, and with the help of their mother attack the
children and kill them in identical fashion. Just one thing stopped you: you thought
they were yours.
This declaration echoes with perfect symmetry Atreus remarks in his initial monologue, and, once again, it should be noticed that neither Thyestes
nor, for instance, the chorus makes any attempt at rebuking it, preferring
instead to expand on the monstrosity of Atreus deeds.
While Atreus is perfectly aware of the moral implications of his plans,
of his new scelera, he also knows that the revenge he has plotted will taint
Thyestes even more than himself: what is the crowning outrage in this
crime he himself will do (2856: quod est in isto scelere praecipuum nefas, |
hoc ipse faciet).15 Atreus observation is not groundless: the explicit ritual
overtones of the actual murder connect the central episode of the tragedy to
a well-known mythical background; Thyestes inexcusable contamination
thus increases the immorality of his character.
14
15
Atreus states explicitly that the revenge is to be disproportionate to the crime: scelera non ulcisceris, |
nisi vincis (1956).
See later in this chapter, pp. 1456.
Atreus rex
145
The chorus appears to lend credibility to Atreus words as it voices its

conviction that both brothers are to be blamed for the endless succession
of evils in the family, and that they are only taking turns (340: alternis)16 in
their folly (33941):
quis vos exagitat furor,
alternis dare sanguinem
et sceptrum scelere aggredi?
What rage drives you to shed by turns each others blood, to seize the throne
through crime?
But further interventions of the chorus on the same topic do not raise
comparable doubts. In a renewed outburst of optimism the chorus rejoices again, at 54676, at the reconciliation of the two brothers, but grants
that Atreus had substantial reasons for his rift with Thyestes: his ira was
indeed provoked by great causes (552). The culmination of this attitude
(perhaps the only point on which the chorus is not severely out of step with
reality) comes as a reaction to the messengers distraught appearance on the
stage. The messengers first words announce that a terrible nefas (624) has
been perpetrated, but specify no author; the choruss question at 63840 is
revealing:
animos gravius incertos tenes.
quid sit quod horres ede et auctorem indica:
non quaero quis sit, sed uter. effare ocius.
You keep our minds in doubt too painfully. Tell what it is that makes you shudder,
and point out its author. I am not asking who it is, but which of the two. Speak
out quickly.
Much as it occupies the moral high ground, the chorus is not necessarily
bound to side with Thyestes from the beginning. Its tormented question at
line 640 opens a dramatic vista, once more, on what could have happened:
what Thyestes himself, that is, could have accomplished against his brother.
To the growing number of characters who are ready to believe in the moral
equivalence of Atreus and Thyestes (first Atreus, then the chorus), we must
finally add Thyestes himself, who, after discovering what he has done to his
children, appropriates the very word with which the chorus had indicated
his potential responsibility: uter. Thyestes immediately, if unknowingly,
sides with Atreus contention that the monstrous banquet will make him as
guilty as the brother who has devised it. Now, the meal over, Thyestes admits
16
Statius will make this hallmark of fratricidal strife the opening statement of the Thebaid: fraternas
acies alternaque regna profanis | decertata odiis (Theb. 1.12).
146
he should be punished along with Atreus (101112: stare circa Tantalum |

uterque iam debuimus, we should both of us long since have stood alongside
Tantalus), and he personally invokes Jupiters punishment on both of them
(10858):
vindica amissum diem,
iaculare flammas, lumen ereptum polo
fulminibus exple. causa, ne dubites diu,
utriusque mala sit; si minus, mala sit mea.
Avenge the day which has been lost, throw your flames, banish light from the sky
and fill it with your thunder. Let the cause of both of us do not wait in doubt
be equally evil; if not, let mine be evil.
These words echo Medeas invocation to Jupiter in Senecas eponymous

tragedy: whether his thunderbolt strikes herself or Jason, it will always
punish a crime: whichever of us falls will perish guilty; against us your bolt
can make no error (5357: quisquis e nobis cadet | nocens peribit, non potest
in nos tuum | errare fulmen). But the comparison is not innocent either.
In Medea, Jasons faults play a well-defined and explicit role, incomparable
with the hushed references to Thyestes previous crimes which surface at
crucial points in Thyestes (and, for this reason, are usually played down by
critics). By equating himself with Jason, Thyestes finally acknowledges that
his brothers suspicions against him were not, after all, misplaced.
No amount of implicit or explicit accusations levelled against Thyestes
past behaviour, however, can make up for the extraordinary emotional impact that Atreus machinations must have on the audience. In foregrounding relative differences of behaviour, not an ontological opposition, Seneca
implicitly builds on a view of civil strife which has trouble establishing
a reliable hierarchy of responsibility. Ovid had already rewritten the confrontation between Aeneas and Turnus in relativistic terms, and Lucans
civil war is mired in the confusion which inevitably arises as like fights
against like.17 The irrationality of the civil war goes hand in hand with
the intolerance of distinctions, boundaries and clear-cut oppositions which
mark the logic of the unconscious.
Atreus and Thyestes are waging their own civil war. They are brothers,
but their moral outlook is also similar: it is difficult to point to hard and fast
hierarchies, to a well-defined sense of right and wrong. As we turn to the
analysis of the brothers characterization, we would do well to keep in mind
that the tragedy offers very little in the way of solid and incontrovertible
17
Ov. Met. 14.56872, with Hardie (1993) 245. See also (as Hardie does) Sil. 9.4025, who will remark
explicitly on the interchangeability of Scipio and Hannibal.
Atreus rex
147
moral assessment, and that even the most explicit indications, such as some
of the choruss statements, are unsteadied by context.
ii
The potential equivalence repeatedly suggested in the text between the
character, motives and intentions of the two brothers is matched by an image
of Thyestes which no amount of goodwill can restore to bona fide Stoic
credibility. Against Atreus resplendent violence, his sublimity of words and
thoughts, Thyestes can only muster ambiguous gestures towards Stoicizing
wisdom. His manner is tentative, however, his conviction faltering. The
audiences moral endorsement of Thyestes as a victim, problematized by
the language of doubling and interchangeability as well as by the sustained
connection between Atreus and poetic pleasure is made even more difficult
by the portrayal of Thyestes actual behaviour in the tragedy.
A long-standing tradition, which in modern times can be traced back to
a 1938 article by Olof Gigon,18 has chosen to recognize in Thyestes a Stoic
sage, a man who is seriously trying to live his life according to high moral
principles and refuses to hate even after enduring the worst of revenges.
Atreus disturbing claims that Thyestes is guilty, developed at 2214 and
answered by Thyestes at 51214, would provide an example of dramatic
discontinuity,19 just as the contrast between those charges and Thyestes
overall characterization should in the end be attributed to the imperfect
amalgamation of multiple sources.20 Along similar lines, E. Lefèvre has
more recently interpreted Thyestes as a proficiens who still hesitates on the
right road, a man whose behaviour would correspond to the second type
described in Letters to Lucilius 75.13, those who have shed the most serious
diseases and inclinations of the mind, but in such a way that they do not
yet have guaranteed possession of their freedom from danger: they can still
relapse (et maxima animi mala et adfectus deposuerunt, sed ita ut non sit illis
securitatis suae certa possessio; possunt enim in eadem relabi).21
Such readings of Thyestes strive to preserve a measure of coherence between Senecas tragedies and his prose works. Yet almost without fail they
18
19
20
21
Gigon (1938).
Thyestes statements at 51214 are interpreted by Gigon ((1938) 182) in reference to a general consciousness that human beings are all guilty.
Gigon (1938) 182.
See Lefèvre (1985). Marti (1945b) offers a similar interpretation of Lucans Pompey (see Marti (1945a)
for the thesis that Senecas tragedies represent a coherent programme of Stoic instruction to Nero,
with each play illustrating a Stoic idea in dramatic form).
148
stumble on a crucial methodological issue, inasmuch as they try to extrapolate from the fabric of the play a unified image of Thyestes without paying
enough attention to how his words and his actions are actually presented
in the plot, and how the complex interplay of different points of view
affects our impression of Thyestes character. Even more fundamentally,
such an essentialist reading of Thyestes (or indeed of any other Senecan
play) is rooted in the attempt to preserve a diametrical opposition between
good (perhaps even Stoic) and bad characters, which in turn would
ensure the viability of a didactic reading of the tragedies: the representation
of unmitigated evil could then be seen to act as a deterrent. Good and
bad qualities, of course, would also be bound to determine the publics
reactions: approval and disgust, both ethical and aesthetic, would unify the
moral and artistic dimensions of the plays.
The moral aspirations of Thyestes and the chorus, however, are criticized even as they are apparently endorsed. The choruss excessive readiness
to believe the unbelievable (such as, for instance, the sudden quiet between
the brothers), its loose connection with the dramatic events at large, belie
its philosophical credentials and lend its vaguely stoicizing feelings a lack of
conviction which deprives the audience of a solid point of emotional identification against Atreus. Thyestes is similarly denied his own authoritative
voice, since his prototypical role that of the good king, the Stoic sage is
weakened by his uncertainties and ambiguities, his sudden contradictions,
his unclear resolve. What is more, his potential function as the positive
emotional pole of the tragedy is undermined by the moral parallelism that
is voiced by Atreus and validated by the chorus with its persistent, if feeble,
doubts.
Thyestes begins, quite literally, with a false step. His appearance on the
stage, back at Argos after a long and painful exile, is marked by a strong
sense of joy and relief (4047):
optata patriae tecta et Argolicas opes
miserisque summum ac maximum exulibus bonum,
tractum soli natalis et patrios deos
(si sunt tamen di) cerno . . .
The longed-for homes of my fatherland and the wealth of Argolis I see at last,
and, greatest and best of goods to a wretched exile, a stretch of native soil and the
ancestral gods (if gods do exist after all) . . .
Stoic sages, notoriously, should avoid hopes and fears alike, and the emphatic presence of optata at the very beginning of the speech can hardly be
regarded as innocent.22 Thus, if we allow the participle to contradict Stoic
22
Tarrant (1985) 149.
Atreus rex
149
orthodoxy, we cannot accept Thyestes impassioned lines at face value: his

orthodox strictures against power, and his desire to leave Argos (41220)
are overshadowed by the suspicion that he is appropriating the rhetorical
stance of a Stoic sage without real conviction.
Equally troubling is the conclusion of his exchange with Tantalus. After
a dialogue which sets his desire to avoid Argos against Tantalus insistence
that he go ahead, Thyestes yields to his son with a contorted disclaimer
(4879):
serum est cavendi tempus in mediis malis.
eatur. unum genitor hoc testor tamen:
ego vos sequor, non duco.
It is too late to take caution when one is deep in troubles; let us go. As your father,
though, this one thing I declare: I do not not lead you; I follow.
Sequor at line 489 recalls the conclusion of the prologue, where Tantalus
ghost, tortured by the Fury, finally submits to her with that very word:
sequor (100). The analogy highlights the difference. Tantalus was tortured
and in fact had no means to escape from an overwhelming supernatural
entity. Thyestes has been engaged in a dialectic exchange with his son which
would easily leave room for retreating. The tragic-sounding acceptance of
what appears to be an inevitable destiny is totally disproportionate to the
situation.
When we look at Thyestes reactions to his brothers proposal in act 3,
we are confronted with a puzzling succession of emotional states. Atreus
initial aside (491507) is likely to reinforce in the audience the notion that
he enjoys a superior awareness of events, and stresses again the fictionality
that will dominate the whole scene: his final words praestetur fides (507)
raise the curtain on a new play. Thyestes reaction to Atreus deceptively
friendly welcome strikes an odd note (51216):
diluere possem cuncta, nisi talis fores.
sed fateor, Atreu, fateor, admisi omnia
quae credidisti. pessimam causam meam
hodierna pietas fecit. est prorsus nocens
quicumque visus tam bono fratri est nocens.
I could excuse all I have done if you were not like this. But I confess it, Atreus,
I do confess it: I have done everything you believed me to have done. The brotherly
feelings you display today make my case as bad as possible. The man who has
appeared guilty to so good a brother is guilty indeed.
Recalling Thyestes impassioned demonstration to Tantalus that Atreus

could only be lying and trying to deceive him (4238, 4349), this
150
unrequested admission of guilt, though couched in somewhat ambiguous

terms (note especially visus in line 516), comes as a surprise, or as an unexpected confirmation of Atreus own charges against his brother. Thyestes
spontaneous offer of his children as obsides seems to underline his decision,
already voiced at 489 (sequor), to follow the flow of events with no further
hesitation, whether this is going to hasten a tragic ending which he dimly
foresees, or because he has after all accepted Tantalus arguments. It is in
any event a serious, tragic Thyestes who enters this uneven dialogue with
Atreus, and it is all the more surprising to see precisely how he turns his
initial refusal into an acceptance of Atreus fake offer. After two longer
interventions, the brothers engage in a stychomythic exchange (53343):
t h.
. . . liceat in media mihi
latere turba. at. recipit hoc regnum duos.
t h. meum esse credo quidquid est, frater, tuum.
at. quis influentis dona fortunae abnuit?
t h. expertus est quicumque quam facile effluant.
at. fratrem potiri gloria ingenti vetas?
t h. tua iam peracta gloria est, restat mea:
respuere certum est regna consilium mihi.
at. meam relinquam, nisi tuam partem accipis.
t h. accipio: regni nomen impositi feram,
sed iura et arma servient mecum tibi.
t h. . . . allow me to hide in the crowd.
at. This kingdom can take the two of us.
t h. Brother, whatever is yours I believe to be mine too.
at. Who can turn down the gifts flowing in from good Fortune?
t h. Whoever knows from experience how easily they can flow away.
at. Do you want to prevent your brother from achieving such great glory?
t h. Your glory has already been accomplished; mine has yet to be: it is my firm
determination to turn down the kingdom.
at. If you do not accept your share, I will relinquish mine.
t h. I accept it; I will take the name of king which is thrust upon me; but laws and
armies and I too will obey you.
Note here Thyestes sudden reversal: it takes him all of one line to shift from
a determination to shy away from the throne, to accepting the substance
of his brothers offer.23 It is true that in the tragedies Seneca often portrays
decisions precisely in this light, as sudden, abrupt deviations from a path
23
Thyestes specific acceptance of the regni nomen sounds particularly ironic in the light of his earlier
strictures: falsis magna nominibus placent (446). Accipio is used by Thyestes with ominous connotations at 5201 (obsides fidei accipe | hos innocentes, frater), by Atreus at 10212 (iam accipe hos potius
libens | diu expetitos).
Atreus rex
151
which had hitherto been carefully followed. More than the acceptance
per se, this extreme abruptness underlines the fundamental inconsistency of
Thyestes character;24 the audience, relying on their privileged knowledge of
the implications of Atreus offer, are bound to be surprised at this transition.
The whole structure of the dialogue accentuates Thyestes unreliability and
inevitably affects the substance of what he has been saying so far: his lofty
Stoic sententiae are ensnared in the grotesque demise of their spokesmans
reputation. The very nature of phonai,

cut and dried maxims supposed to
be of use in determining ones everyday conduct, only heightens the sense
of ridicule.
A sceptical evaluation of Thyestes character should not necessarily entail an automatic endorsement of Atreus. This is not the point. What we
should keep in mind, rather, is that Thyestes is no moral testing stone, and
that Atreus violence finds in the play no immediately convincing alternative. Even more upsetting is how far Atreus point of view can influence
the choruss and the audiences perception of his adversary: in Atreus perversely symmetrical logic, Thyestes is his criminal doppelganger, ontologically identical and only momentarily, by chance, cast in a subordinate role.
The play is unable to counter the force of this logic and its primal appeal,
a logic cherished and reviled in equal measure. The poignancy of Thyestes
is precisely that it leaves us to contemplate the contrast between a ruthless
but aesthetically and emotionally appealing murderer and a hypocritical,
faltering and charmless adulterer.
de clementia
quod nolunt velint
Atreus is at the same time the sacerdos and the addressee of the sacrifice, as
he makes clear that he is offering the victims to himself.25 By collapsing two
roles which normal religious practice obviously kept apart, Atreus signals a
more general subversion of rules. Sacrifice traditionally articulates a welldefined hierarchy between gods, men and animals. Thus a god is always
the recipient of the sacrifice, a man the performer, an animal its object.
24
25
One may compare Medeas behaviour in Ovids Met. 7, when she stresses the importance of following
virtue (6971), but quickly changes her mind at the sight of Jason (767). More poignant is the
parallel with Agamemnon in Aesch. Ag. 9445 (but note how forcefully Clytemnestra argues for her
request before Agamemnon yields), and Pentheus in Eur. Bacch. 8456: in both cases the change of
mind has tragic consequences.
On Atreus as a god see Hine (1981) 266.
152
As a man becomes god, others (Thyestes children) tragically fulfil the role
of animal victims. Gods are dragged down from the sky, or, in turn, man
can raise himself to the stars. Either way the foundations of religious and
political order are fatally challenged.
Atreus appropriation of a divine persona exacerbates the gods problematic status in the play and the world it pictures. Thyestes and the chorus
display a firm belief (at least, a firm discursive belief ) in their existence and
importance, while Atreus effectively disposes of them, and replaces them
with himself. His main target, clearly, is the upper gods, the superi. It is to
them that the scorning dismissal of line 888 I release the gods: I have
attained the utmost of my prayers (dimitto superos: summa votorum attigi)
is referred, and it is they who are said to have fled at line 1021. In this particular removal of upper deities, and the concurrent promotion of violent
characters connected with the underworld, Atreus fulfils Junos programmatic and metaliterary statement in the Aeneid: if I cannot prevail upon
the gods above, I shall move hell (7.312: flectere si nequeo superos, Acheronta
movebo). Even more so, of course: Acheron is not simply exploited, but
brought to earth.
Thyestes own statements about the gods are often contradictory. In
his opening speech he voices scepticism about their existence: tractum soli
natalis et patrios deos | (si sunt tamen di) cerno (4067). Later, when Atreus
has revealed his actions, Thyestes reacts with utter despair: the gods have
fled away (1021: fugere superi). But suddenly, at line 1077, Thyestes embarks
on an impassioned prayer to the summus caeli rector (exalted ruler of the
sky), who is asked to avenge Atreus crime. The rhetorical elaboration of
can only sound hollow after the inconclusiveness of the previous
the klesis
statements, and provides the background for Atreus repartee at 11023,
when he responds to Thyestes emphatic I call on the gods who guard
the innocent ( piorum praesides testor deos) with a chilling reference to his
past misconduct: why not the marriage-gods? (quid? coniugales?). The
tragedy ends by reaffirming Atreus and Thyestes opposing views about
the role of gods and men in shaping events (111012): TH. The gods
will be my avengers; my prayers deliver you to them for punishment. AT.
For punishment, I deliver you to your children (TH. vindices aderunt
dei; | his puniendum vota te tradunt mea. | AT. te puniendum liberis trado
tuis).
Atreus constantly refers to the level of human interaction, and his selffashioning as a god is in fact consistent with the notion that gods do not
have any impact on human activities. The standard criteria for divinity are
wrecked: each man is his own god provided that he can muster sufficient
Atreus rex
153
power. This ambivalent affirmation of divinity in a godless world communicates one of the most chilling messages of the play. The triumph on stage
of a king-priest-god is, of course, an explosive combination in the context
of Imperial Romes structuring ideology. Under Augustus the princeps26
had acquired a distinctive characterization as a sacerdos,27 prominently involved ex officio in a number of sacrificial rituals even as his divine attributes
increase dramatically.28 In a world where human actions have effectively
appropriated divine power, that boundless power is concentrated in new
and novel gods.
Such a radical redefinition of the relationship between human and divine
is central to the conceptualization of power in Thyestes, a play whose political
implications have often been at the heart of interpreters concerns. The
issue is made more, not less, complicated by the fact that the play offers
the superficially attractive possibility of a reading a` clef .29 Atreus could be
Nero; the hapless satelles, a Seneca-like counsellor, though noticeably less
successful; Thyestes, a victim, the Victim of the tyrants cruelty; the chorus,
a righteous but powerless senatus. Clearly, if we accept these equations we
are bound to steer our reading of the play in one specific direction. If
Atreus is (like) Nero, then he must be wrong, because he impersonates
the ultimately cruel and irrational tyrant Suetonius unadorned words
project here their powerful shadow. Even if we accept that Thyestes blatant
inconsistencies hardly make him a convincing hero, we will inevitably tend
to cast the chorus in the ambitious role of the external moral adjudicator
who comments philosophically on the follies of human power.
The main difficulty with such a reading is that it works only by taking messages out of context, most notably the second chorus on kingship
and the satelles or Thyestes sententious statements.30 None of these lofty
26
27
28
29
30
The very etymology and concept of princeps (he who takes the first portion) may be traced back
to sacrificial ritual; Scheid (1988) 273.
On the importance of sacrifice in the articulation of Roman imperial cult see Price (1980); specifically
on the Ara pacis and its interpretation in this context Zanker (1988) 118 and passim.
It is interesting to recall a passage from De clementia in which the characterization of the princeps as sacerdos and the imagery of murder-as-sacrifice are combined. Augustus debates whether
he should punish L. Cinna, who attempted to murder him: ergo non dabit poenas, qui . . . non
occidere constituat, sed immolare? (nam sacrificantem placuerat adoriri) ( shall he not pay the penalty
who . . . is determining not to murder but to immolate me? (for the plan was to attack him while
offering sacrifice), 1.9.4).
Thyestes teases the audience in this direction, thanks to a number of anachronistic references to
Roman reality, especially the incongruous mention of Quirites at 396; see Picone (1976). Direct
references to contemporary events in the plays are advocated by Lefèvre (1985) and (1990), but see
Calder (197677). See above, p. 16.
Note, however, that in spite of Suetonius attractive definition of Nero as religionum . . . contemptor
(Nero 56.1), he was not the only emperor criticized in this way; see Suet. Tib. 44 and Cal. 32.3, with
Barton (1994) 53.
154
propositions, however, exists in a vacuum, and their meaning shifts considerably once they are assessed within the complex dynamics of the play.
The politics of Thyestes, I will argue, are located not so much in what the
characters say as in how, when and to whom they say it. Atreus may be a
monster, and the chorus a safe-haven of restraint and reason, but what is
particularly upsetting in the scenes that I want to analyse in this context
is that they engage extensively with the reality outside the play, with other
texts and with actual political positions. Once reality is drawn into the
world of Thyestes, it is less easy to dismiss the whole of the play as a nightmare: perhaps it is as real as the solid political tracts it brings into the fray,
and perhaps the Realpolitik it advocates in both theory and practice is not
just a frenzied fiction.
The dialogue between Atreus and his satelles, as we have already noticed, reiterates and expands the complex interaction between layers of
dramatic plots and substantiates the connection between Atreus mastery
of his vengeance plot and the creative force of poetry.31 It is now important
to consider more systematically the role played by the satelles in that dialogue, and especially the protocols of political behaviour to which both he
and Atreus refer.32
The first point to note is the disproportionate stage presence given to each
of the two characters. This is hardly surprising given the respective roles they
play, but this inequality is not a mere reflection of this hierarchy: in principle
nothing prevents the satelles from playing a much more prominent role, and,
conversely, his complete subordination can only affect the ideas he voices.
The scene opens with a long monologue by Atreus (176204), who fails to
acknowledge the satelles presence. The latters opening words come in the
form of a question at 2045: doesnt public disapproval deter you? (fama
te populi nihil | adversa terret?). The questioning mode often conveys the
emotional and cognitive weakness of a character, as in the case of Tantalus
in the opening scene of the play, or, as we will see shortly, of the chorus
in many other instances. Here the satelles avoids both assured statements
and emotional outbursts, but chooses to couch his first intervention in the
respectful and hesitant mode of a question. He does not pass any general
moral comment on the plans that his master is devising, nor does he try to
stop him; indeed, just like the chorus, his question raises a (minor) possible
objection which Atreus has already anticipated and dismissed, since he
has already stated that a negative reputation is precisely what he strives for:
31
32
See ch. 2, p. 46.

Tame and submissive the satelles might be, but it would be wrong to deny him a distinct personality, as
Knoche ((1941) 70) does by suggesting that he simply embodies another aspect of Atreus irrationality.
Atreus rex
155
Up, my soul, do what no coming age shall approve, but none forget (1923:
age, anime, fac quod nulla posteritas probet, | sed nulla taceat).
Atreus answer to this first question a pragmatic apology for the need
of sheer force in the exercise of power prompts the satelles to explain
his own political philosophy more assertively. Note that at this point the
satelles does not suggest an alternative to Atreus exclusive focus on the best
means to preserve power: the satelles even at his most eloquent makes no
attempt to raise the discussion to a higher plane of abstract moral principles;
in practice he accepts Atreus agenda in full, and thus poises himself for
inevitable defeat. Even his belated praise of honestum (213), sanctitas, pietas
and fides (216) is subordinated to the question of how to retain power and
popularity, and thus fails to turn the dialogue into a loftier debate about
absolute good and evil. The satelles exhortation comes in the apparently
confident form of a sententia (213: rex velit honesta: nemo non eadem volet,
let the king want what is right; no one will want anything else), but even
so the clumsiness of the litotes betrays the weakness of the statement.
As in the dialogue between Thyestes and Atreus, sententiae are exploited in order to highlight, in fact, the moral inconsistency and potential
hypocrisy of those who utter them. Compared to Atreus articulate, eloquent and passionate oratory, the satelles staccato statements sound like
perfunctory attempts to remind his master of a hardly compelling point of
view. Just as Thyestes turns in the space of two lines from a determined
refusal to accept Atreus invitation (540) to a thinly qualified assent (5423),
the satelles moves from the firm principle stated at line 219 nefas nocere
vel malo fratri puta (count it wrong to harm even a wicked brother)
to the astonishing exhortation with which he responds to Atreus lengthy
remarks: slain by the sword, let him spew forth his hateful soul (245: ferro
peremptus spiritum inimicum expuat).
From this point onwards (we are not yet at the middle of the scene) the
satelles forgoing any attempt to keep the moral high ground is effectively
an accomplice in the elaboration of Atreus revenge plot.33 His renewed
mention of Pietas after his initial acquiescence can only be understood as a
reference to the way in which Atreus will carry out his vengeance: having
endorsed the idea of murdering Thyestes (245), the satelles mentions Pietas
when Atreus confesses that a simple killing is not enough (246: de fine poenae
loqueris, ego poenam volo, you speak of the punishments completion; I
desire punishment itself ). This scaling down of the role of Pietas, which
33
See La Penna (1979) 138. Note also La Pennas important suggestion (1334) that Accius Atreus, by
way of contrast, could have staged direct and violent resistance against a tyrant.
156
is consistent with the disappearance of sanctitas and fides, reflects back on

the initial reference to it at line 217, and emphasizes the weakness of that
first appeal.
Now the satelles can only question the details of Atreus plot, suggesting
better alternatives or pointing out dangerous loopholes in the planning.
The rapid exchange at 2558, for instance, underlines the closeness of the
two mens cooperation and the satelles active advisory role:
at. nil quod doloris capiat assueti modum;
nullum relinquam facinus et nullum est satis.
s a. ferrum? at. parum est. s a . quid ignis? at. etiamnunc parum est.
s a. quonam ergo telo tantus utetur dolor?
at. Nothing confined within the limits of ordinary anger; I will not overlook any
crime, and none is enough.
s a. The sword?
at. Too little.
s a. Fire, then?
at. Still too little.
s a. What weapon, then, will such grief use?
Once again, Atreus impassioned speech at 26786 has an amazing effect

on his advisor, who immediately erases any remaining difference between
himself and his master: the trap devised for Thyestes is now ours (287:
nostros . . . laqueos). In the final part of the scene, moreover, the satelles, by
asking several questions on points of detail, shows his complete acquiescence in the revenge plot, his desire to avoid possible mistakes, and his
total dependence on Atreus sharper intuitions. Ethical and intellectual
weaknesses are thus joined together. From hesitation to tacit agreement
to active, if rather useless, participation, the satelles moral stance has now
been completely invalidated.
After Atreus plot has been sketched out, the satelles devotes his questions
to analysing Thyestes possible reactions (2869 and 2945):
s a . sed quibus captus dolis
nostros dabit perductus in laqueos pedem?
inimica credit cuncta. at. non poterat capi,
nisi capere vellet.
...
s a . quis fidem pacis dabit?
cui tanta credet? at. credula est spes improba.
s a . But caught by what wiles will he be led to set foot in our traps? He sees enemies
everywhere.
at. He could not be caught if he didnt himself want to catch . . .
Atreus rex
157
s a. Who will give him confidence in peace? Whom will he trust so much?
at. Wicked hope is credulous.
In this context the satelles usual inquisitive mode foregrounds his subordination to Atreus as far as a proper understanding of psychological reactions
is concerned. It is now Atreus turn to utter sententious statements non
poterat capi, | nisi capere vellet (2889) and credula est spes improba (295)
but these seem to be predicated more on solid practical experience than
on the rather cold moralistic tone of the satelles own pronouncements at
the beginning of the scene. Atreus statements, moreover, will prove reliable and true in the rest of the play. This contrast between the satelles
generic, textbook sententiae and Atreus authoritative, experienced beliefs
is highlighted in the exchange at 3057:
s a . iam tempus illi fecit aerumnas leves.
at. erras: malorum sensus accrescit die.
leve est miserias ferre, perferre est grave.
s a. By now time has alleviated his troubles.
at. You are wrong: a sense of wrongs grows day by day. It is easy to bear misfortune;
to keep on bearing it is hard.
Similarly, it is Atreus who carefully considers the reliability of his own

sons as possible accomplices in the execution of the plot, and bases his
judgement, once again, on actual experience: the art of silence is taught by
lifes many ills (318: tacere multis discitur vitae malis).
In the closing lines of the scene, at the masters injunction to keep their
deliberations secret (333: nostra tu coepta occule, as for you, conceal my
plans), the satelles replies that he needs no warning: no need to admonish me; both fear and loyalty shall shut them in my heart, but rather
loyalty (3345: haud sum monendus: ista nostro in pectore | fides timorque,
sed magis claudet fides).34 This distinction between timor and fides must
be read in conjunction with the previous discussion on the role of fear
in the exercise of power (20510); there Atreus had boldly rejected the
satelles apology of sincere popular favour for the tyrant by pointing out
that truly unfettered power consists not so much in convincing ones subjects, but forcing them to want even what they do not want: quod nolunt
velint (212). Power consists in replacing psychological and moral truth with
factual superiority, which forces a reliable consent: the satelles final words
provide direct proof of the fact that fides can indeed be attained not by
34
Calder ((197677) 9) suggests that a pause before fides would convey very effectively the truth,
namely, that timor will guarantee the satelles silence.
158
proposing honesta, but by creating a system whereby superior power cannot

be resisted.
I now focus again briefly on the overall structure of the dialogue. In
his transition from dissent to complicity the satelles can be unflatteringly
compared with Phaedras nutrix. In Phaedra the first dialogue between the
queen and her nurse occurs in the same structural position as the dialogue
in Thyestes, namely in the first act. There, again, the nurse represents the
paradigm of ratio and moderation as opposed to Phaedras insane passion,
but she is much more active and articulate than the satelles: not only does
she speak more extensively, thus acquiring substantial stage-presence, but
she is also capable of putting forth an insistent, fiery rhetoric which the
satelles repeatedly fails to achieve. Not unlike the satelles, the nurse shifts her
position in the course of the dialogue; however, this structural similarity
foregrounds the satelles own weakness. First of all the nurse starts changing
her mind only when faced with Phaedras resolve to commit suicide, and
her words indicate that her softening to Phaedras wishes is a direct result
of her motherly affection for the woman: and so should my old age allow
you to rush headlong to your death? Stop your frenzied impulse (2623: sic
te senectus nostra praecipiti sinat | perire leto? siste furibundum impetum). The
satelles swiftly became Atreus accomplice without any direct menace, either
against himself or others. Secondly, the degree of the shift is remarkable.
At the end of the dialogue the nurse does agree to sound out Hippolytus
on Phaedras behalf, but again explains her choice on affective grounds,
and her words convey the sense that Hippolytus is being unduly harsh to
Phaedra (26773):
solamen annis unicum fessis, era,
si tam protervus incubat menti furor,
contemne famam: fama vix vero favet,
peius merenti melior et peior bono.
temptemus animum tristem et intractabilem.
meus iste labor est aggredi iuvenem ferum
mentemque saevam flectere immitis viri.
My mistress, only comfort to my tired old age, if so unruly is the frenzy that seizes
your mind, disregard fame: fame is not partial to truth, but better to those who
deserve worse and worse to the good. Let me sound out his sad and intractable
heart. Leave this difficult task to me to approach this fierce youth and bend his
cruel mind, harsh as he is.
After the choral ode, the nurse takes the stage again to describe Phaedras
awful state of mind and physical distress (36086), and before addressing
Hippolytus she invokes divine protection for her deeds (40630). These
Atreus rex
159
moments of hesitation and deferral stand in sharp contrast to the satelles

rapid shift from a moral high ground to the level of Atreus plotting, and
especially to his final, wholehearted subscription to his masters criminal
plans in the name of timor rather than fides (335).
Before the satelles disavows any attempt to influence Atreus, he insists
on four closely connected topics: the rulers necessity to acquire a good
fama (2045); the ambivalent effects of metus (20710); the rulers duty to
pursue honesta (213) and the necessity to uphold pudor, cura iuris, sanctitas,
pietas, fides (21517). With their laconic, often formulaic tone, many of
the satelles words have a rather textbookish ring to them, and in fact they
reflect widespread Hellenistic and Roman ideas on how a prototypical
good king should behave. Given this background, it is not surprising
that they are often close to Senecas own words in De clementia, but this
intertextual relationship should not be explained away. It would be pointless
to evaluate the satelles words without considering that they reflect if, as
seems probable, Thyestes was written later than De clementia what Seneca
elsewhere considered to be the right prescription for a rulers considerate
behaviour.35
The satelles very first statement concerns fama: fama te populi nihil |
adversa terret? (2045). Principes multa debent etiam famae dare (princes
are bound to give much heed even to report) was the advice Seneca had
offered Nero in De clementia (1.15.5), but the satelles use of the verb to
terrify (terret) is strikingly incongruous after Atreus bold and self-confident
opening monologue at 176204. At 20710 the satelles tries to impress upon
his master the inherent dangers of a power based on fear:
quos cogit metus
laudare, eosdem reddit inimicos metus.
at qui favoris gloriam veri petit,
animo magis quam voce laudari volet.
Whom fear compels to praise, them, too, fear makes into enemies; but he who
seeks the glory of true favour will wish heart rather than voice to sing his praise.
Compare De clementia 1.12.34:

interim, hoc quod dicebam clementia efficit, ut magnum inter regem tyrannumque
discrimen sit, uterque licet non minus armis valletur; sed alter arma habet, quibus
in munimentum pacis utitur, alter, ut magno timore magna odia compescat, nec
illas ipsas manus, quibus se commisit, securus aspicit. (4) contrariis in contraria
agitur: nam cum invisus sit, quia timetur, timeri vult, quia invisus est, et illo
execrabili versu, qui multos praecipites dedit, utitur: Oderint, dum metuant.
35
On all the passages from De clementia see now Malaspinas rich commentary (Malaspina (2001)).
160
Meanwhile, as I was saying, it is mercy that makes the distinction between a king
and a tyrant as great as it is, though both are equally fenced about with arms; but
the one uses the arms which he has to fortify good-will, the other to curb great
hatred by great fear, and yet the very hands to which he has entrusted himself he
cannot view without concern. Conflicting causes force him to conflicting courses:
for since he is hated because he is feared, he wishes to be feared because he is hated,
and not knowing what frenzy is engendered when hatred grows too great, he takes
as a motto that accursed verse which has driven many to their fall: Let them hate,
if only they fear.
The debate reaches its climax at 21517, where the satelles sums up the
essential qualities of a stable kingdom: where there is no shame, no care
for right, no honour, righteousness, faith there sovereignty is unstable
(ubi non est pudor | nec cura iuris sanctitas pietas fides, | instabile regnum
est). Several passages in De clementia deal with these concepts, and similar
lists of positive qualities can be found, for instance, at 1.19.8: who would
dare to devise any danger for such a man? Who would not wish to shield
him, if he could, even from the chance of ill him beneath whose sway
justice, peace, chastity, security and honour flourish? (quis huic audeat
struere aliquod periculum? quis ab hoc non, si possit, fortunam quoque avertere
velit, sub quo iustitia, pax, pudicitia, securitas, dignitas florent . . . ?). Compare
also 2.1.4:
nunc profecto consentire decebat ad aequum bonumque expulsa alieni cupidine,
ex qua omne animi malum oritur, pietatem integritatemque cum fide ac modestia
resurgere et vitia diuturno abusa regno dare tandem felici ac puro saeculo locum.
Now assuredly it were fitting that men, thrusting out covetousness from which
springs every evil of the heart, should conspire for righteousness and goodness,
that piety and uprightness along with honour and temperance should rise again,
and that vice, having misused its long reign, should at length give place to an age
of happiness and purity.
Finally, cura iuris refers at the same time to respect for judicial procedure,
and to the all-important Hellenistic notion that the good ruler should
constantly consider himself subjected to the rule of the Law. Indeed this
is one of the notions that Seneca literally puts in Neros mouth in the
reported speech which opens De clementia: I so hold guard over myself
as though I were about to render an account to those laws which I have
summoned from decay and darkness into the light of day (sic me custodio,
tamquam legibus, quas ex situ ac tenebris in lucem evocavi, rationem redditurus
sim, 1.1.4).
Against this careful mix of respectable Stoic concepts, Atreus behaviour
is all the more remarkable. The disproportion in the quantity and quality
Atreus rex
161
of words he can muster vis-à-vis the satelles is part of a dramatic truth

which the presumed moral superiority of the satelles can hardly efface.
This was, of course, what the public expected of him, and a repenting
Atreus would not have been credible. As Cicero remarks in De officiis,
the famous motto oderint dum metuant can stir up an applause precisely
because it fits the speaker so well: When Atreus speaks in this manner, he
elicits an applause, because his words are worthy of his character (Atreo
dicente plausus excitantur; est enim digna persona oratio, 1.28.97).36 But just
as it would be dramatically inappropriate to overestimate the satelles moral
high ground, it would also be wrong to dismiss Atreus without scrutinizing
his words more closely, as if they simply represented the furious outpouring
of a demonic character.
Atreus displays a coherent vision of power which is not simply based on
ira, and thus disqualifiable as irrational, but rather on a keen perception
grounded in experience of the realities of human interaction. Atreus
philosophy of power consists in the exclusive focus on praxis; the exercise
of power should not be predicated on anything but the most effective ways
to preserve and further it, and every strategy should be evaluated according
to this perspective, leaving aside irrelevant considerations about morality,
divine power or the search for popular favour.
The contrast between inessential and essential components of power is
foregrounded in the very first words that Atreus and the satelles exchange:
at the latters enquiry about fama | . . . adversa (2045), Atreus replies by
focusing on facta (206). The satelles question is predicated on the contrast
between words and deeds, and on the assumption that external power could
force the latter but not control the former, that an obedient population
might still retain the psychological freedom to bestow on its ruler a negative
fama. Atreus replies that the tyrant can force thoughts as well as deeds: the
greatest advantage of royal power is this, that the people are compelled
as well to bear as to praise their masters deeds (2057: maximum hoc
regni bonum est, | quod facta domini cogitur populus sui | tam ferre quam
laudare). There is a sense of proud and almost joyful subversion of the
satelles credibility in Atreus words which makes his behaviour look more
convincing and consistent by comparison. Similarly, when the satelles replies
that true praise only reaches those who covet favoris gloriam veri (209),
Atreus quickly points out that he is not interested in a reward which even
a humble subject could attain: laus vera et humili saepe contingit viro, |
non nisi potenti falsa. quod nolunt velint (21112). Atreus is here applying a
36
Contrastingly, in De ira 1.20.4 Seneca expands on the immanitas of such a dictum.
162
logic which Seneca had tried to turn to very different ends in De clementia,
where he repeatedly points out that the most authentic sign of distinction
for a ruler lies in his ability to choose not to do something that any one of
his subjects could also do: anybody can kill in revenge, but only the ruler
can decide to refrain from revenge and spare a life (1.5.4). Atreus bases his
decision to spurn the reward of genuine praise on precisely the same sense
of distinction and uniqueness. Later in the play, Thyestes portrays himself
as a ruler who supposedly enjoys the peoples favour and expects a warm
welcome on his return home: Argos will come to meet me, a great crowd
will come (411: occurret Argos, populus occurret frequens). Yet the play does
not provide any indication that this actually occurs, and Thyestes words are
left without any visible sign of fulfilment, thus encouraging us to suppose
that Atreus cynical lack of illusions is more realistic than his brothers faith
in popular favour.
While the motto quod nolunt velint is a reminder of the famous oderint
dum metuant, it is also significantly different, and a great deal more demanding. Atreus aspires to complete control over his peoples reactions,
and is aware that force can turn dissent into consent, that his superior will
can bend an initially uncooperative mind. Always an innovator, Atreus also
injects into his political philosophy a sense of bold novelty, which parallels
his search for private vengeance. Like the Fury of the prologue, he curtly
dismisses the notion that he should look for consent and approval: like
Tantalus shadow, his subjects will be forced to forget their own wishes and
conform to his. Thyestes provides in fact a dramatic confirmation of Atreus
insight when in the following act he yields to his sons persuasive advice
and enters the city unwillingly: my mind falters and wishes to take my
body back; my steps are unwilling (41920: animus haeret ac retro cupit |
corpus referre, moveo nolentem gradum).
In his assertive defence of the tyrannical ethos, Atreus is more than a
predictable stereotype of a tragic tyrant. His words are forceful, original,
impressive, and his deeds cohere with his policy statements. The satelles
(and, as far as he is later concerned, Thyestes) can only voice a perfunctory
array of well-meaning but ineffectual commonplaces, which the tragedy
itself reveals by the way they are uttered, and the defeat they encounter
to be hardly worth serious consideration. The contrast between the two
characters is illuminating, once again not because of the practical results
of the confrontation the unsurprising fact that Atreus has his way and
carries out his revenge plan undisturbed but because Seneca chooses to
represent the opposition to Atreus plans and thoughts in an extremely
ambivalent and unappealing form. The satelles did not have to turn into
Atreus rex
163
an Antigone to make his lofty thoughts more believable. The fact that he
is not essential, in the rest of the play, for the actual realization of Atreus
revenge, further characterizes his acquiescence as an act of moral weakness,
of almost willing submission to Atreus vision. Of course Atreus held him
hostage and could have easily punished any resistance; but the play does
not give any indication that this might happen, and portrays instead the
satelles parabolic descent from resistance to complicity.
Although it may be superficially appealing to identify Seneca himself in
the satelles, a fit complement for a Caligula or Nero in the role of Atreus, the
debate between Atreus and his minion is more than a reflection of a specific
incident in the history of Roman political life. It is in fact a dramatized
contrast between two different conceptions of power, a losing and a winning
one. It matters little which one holds the higher moral stature on paper:
what really matters (on the stage, and, in fact, in the palace) is how they
deal with each other, and what degree of credibility they are able to instil in
the audience. The dialogue juxtaposes a truth and a fiction: the fiction of
half-hearted resistance versus the matter-of-fact truth of what power really
is and how it functions. In this respect Atreus embodies a view of power
which in practice, if not in theory, is truly in keeping with the reality of
Roman imperial rule.
The chorus may well appear to be a more reliable candidate for moral
guidance than the satelles or, for that matter, Thyestes himself.37 Indeed it
is tempting to control the disruptive force of the play by locating a reliable
moral message in the choruss lofty interventions, or at least by arguing
that the chorus enjoys a relatively detached position apart from the moral
turmoil experienced by the main characters.38 True, much of what the
chorus observes in various moments of the tragic action can be connected
with Stoic concepts as they are explained and advocated by other sources.39
The chorus, however, is, willingly or not, completely enmeshed in the
vicissitudes of the play, both because its comments fit in the linear sequence
of events and often explicitly refer to very recent developments, and because
the chorus, for all its idiosyncratic behaviour, is yet another character on
the stage. The choruss actions and reactions cannot be evaluated in the
abstract, as a moral commentary which can be taken at face-value, but must
37
38
39
Among recent work on Senecan choruses in general see Tarrant (1978) 2218; Mazzoli (198687);
Davis (1993); and Castagna (1996). Hiltbrunner (1985) 98991 offers an annotated bibliography.
Tarrant (1985) 137.
Tarrant remarks ((1985) 45) that the breadth of its [the choruss] perspective and the dignity of
its ultimate response give the play its only moments of moral sanity. That may well be true; it is
arguable, however, whether this sanity can muster enough credibility to be counted as a realistic
alternative.
164
be evaluated dialectically, as another, often dissonant, voice in the conflict

of contrasting points of view staged in the play.
The chorus first appears on stage after the metadramatic prologue, and
delivers an impassioned appeal for an end to the chain of horrors which has
so far besieged the Pelopidai. The very beginning of the ode is characterized
by a vein of hesitation and uncertainty40 which is evident in the repetition
of hypothetical statements (1226):
Argos de superis si quis Achaicum
Pisaeasque domos curribus inclitas,
Isthmi si quis amat regna Corinthii,
et portus geminos et mare dissidens,
si quis Taygeti conspicuas nives . . .
If any of the gods loves Achean Argos and Pisa famous for its chariots, if any loves
the kingdom of Corinthian Isthmus, its twin harbours, and the sea divided, if any
loves the far-seen snows of Taygetus . . .
In these lines the chorus displays not only scepticism on the existence of
a divine protection for Argos in particular, but also doubts about the possibility of knowing even very basic facts such as the existence of protecting
gods.
This cognitive inadequacy acquires a further dimension in the following
lines, as the chorus deals with Tantalus past and thus invites a close comparison with the prologue. Lines 1226 list several locations close to Argos,
and the description of the last one, the snow-covered peaks of Taygetus,
ends in a brief digression on the alternating forces of winter and summer,
which provoke and dissolve the snow (1279). Against this background of
natural alternation by definition constant and unstoppable the chorus
voices the desire that no new misdeeds plague the royal household, and
specifically that there be a conclusion to the alternae . . . vices (133)41 which
have so far besieged it (1325):
advertat placidum numen et arceat,
alternae scelerum ne redeant vices
nec succedat avo deterior nepos
et maior placeat culpa minoribus.
Let divine power look to us peacefully and forbid that crimes in alternate sequence
return, that a worse grandson succeed his grandfather, or a greater crime please the
new generations.
40
41
Tarrant (1985) 106. In this passage Seneca goes well beyond the traditional si = siquidem in prayers
and conveys hesitation and scepticism.
See Ag. 778: quas non arces scelus alternum | dedit in praeceps?, with Tarrants remarks ((1976) 182)
on Greek parallels and his note on Ag. 44: sanguine alterno ((1976) 179).
Atreus rex
165
The chorus speaks here without any knowledge of the prologues events,
and possibly in a temporal sequence which is parallel rather than subsequent to those events.42 But from a dramatic point of view these hopes
are unequivocally voiced after the prologue has already shown that they
can no longer be nurtured, that repetition constitutes the plays moral and
dramatic dynamic: a number of specific verbal parallels underlines the stridency of this contrast.43 The chorus sides here with forces which have tried
to prevent the unfolding of dramatic events, and, just like Tantalus shadow
or the satelles, it is doomed to failure. Indeed, the following emphatic statement: peccatum satis est (138) sounds particularly ironic, since the immediate
context suggests, rather, a relentless pattern of return.
Inextricably linked to the prologue, the first choral song looks forward
to future developments in the plot. This prefiguration of events is not
presented, however, as a form of predictive ability which would bestow
on the chorus a claim to higher knowledge, but is rather perceived in
this context as involuntary, and therefore tragically ironic. The chorus
describes Myrtilus and Tantalus crimes while hoping that they will never
be repeated (138: peccatum satis est), but in fact they ostensibly prefigure
the monstrous deeds that the tragedy will once again evoke. Deception
is the element stressed most in this part of the song, thus hinting at its
importance in the subsequent unfolding of the plot;44 especially deceptus
totiens (so often deceived, 159) and falli libuit (gladly has been baffled,
167), both referring to Tantalus, aptly describe Thyestes own behaviour
at a later stage and reinforce the structural parallelism which links the two
characters throughout the play. But precisely because of the context, this
insistence on deception reflects also on the choruss own tendency to be
deceived, its inability to grasp events effectively.
Even the richly detailed description of Tantalus punishment is consistent with the choruss display of inadequate knowledge or sheer wishful
thinking. Contrary to the choruss opinion that Tantalus has been subjected to the most appropriate form of retribution (1501: nec dapibus feris |
decerni potuit poena decentior, nor could a more fitting punishment have
been decreed for such barbarous food), the prologue has already shown
that a worse torture has in fact been devised for him (701; 823). Similarly, the final image of the ode describes, with a strong sense of closure,
the apparently eternal situation of Tantalus in the underworld, where he is
42
43
44
See ch. 2 and ch. 5, passim.

See 25: alterna vice; 289: rabies parentum duret et longum nefas | eat in nepotes; 8990: ducam in
horrendum nefas | avus nepotes? (Tarrant (1985) 108 on 1335); line 135 recalls one of the Furys most
striking mottos: Thracium fiat nefas | maiore numero (567).
Tarrant ((1985) 110) notes the occurrence of this theme in the first part of the play.
166
forced to drink dirty shallow water after all the enticing goods presented to
him have disappeared (1745). Again, the prologue has shown that closure
and conclusion are far from guaranteed, since Tantalus has already been
summoned to earth and forced to provoke a new turn in the terrible series
of the catastrophes suffered by the Pelopidai.
Because of the choruss tragic inability to understand the situation unfolding on the stage, the absence of precise indications of who exactly its
members are45 acquires particular relevance. It enhances the choruss shadowy appearance, its rarified, almost inactive existence,46 suspended in the
netherworld between powerful dramatic actions. Its utterances of hope, its
calls for restraint, are neutralized by the weakness of its interpretative tools.
The first song makes it clear that the chorus has no higher claim to the
truth, and does not offer to the public a secluded island of moral certainty
amidst the turmoil of the tragedy.
Subsequent choral odes strengthen these intimations of tragic irony and
doomed inefficacy, and further reduce the plausibility of retrieving a deeper
and truer meaning of the play from the choruss words. Credat hoc quisquam?
The opening of the third choral ode once again foregrounds the issue
of knowledge and belief. This second interlude between actions, namely
between the meeting of the brothers and the appearance on scene of the
messenger, centres on the choruss inability to see beyond the deceptive
surface of Atreus actions: once again, the chorus and Thyestes are paired as
victims of deception, and the choruss reiterated expressions of fear remain
a mute and ineffective counterbalancing element. Since the ode directly
follows one of Atreus most ominous double entendres (545: ego destinatas
victimas superis dabo), the opening words are even more loaded with tragic
irony (54651):
credat hoc quisquam? ferus ille et acer
nec potens mentis truculentus Atreus
fratris aspectu stupefactus haesit.
nulla vis maior pietate vera est:
iurgia externis inimica durant,
quos amor verus tenuit, tenebit.
45
46
Tarrant (1976) 180 on the impersonal quality of Senecan choruses; see Tarrant (1985) 106, n. 1, with
Leo (1897) 51013. On the notion, however, of the chorus as a collective character see Grimal (1975)
265, with Picone (1976) 64. There are useful remarks on the choruss personality also in Rozelaar
(1976) 56163 and a nuanced discussion in Davis (1993) 3963. Zwierlein ((1966) 746) sees in this
impersonal quality yet another sign that the plays were not performable.
It is worth noting that in both of the parodoi that Tarrant ((1985) 106) compares with this ode,
Aesch. Sept. 87170 and Soph. OT 158215, the chorus is seen reacting to events with a fear inspired
by its perception of the disasters to come.
Atreus rex
167
Will anyone believe this? Atreus, that cruel, harsh man, out of control, bloodsoaked, stopped still at sight of his brother. No power is stronger than true fraternal
love; angry strife with strangers lasts, but those whom true love has bonded together
will continue to bond.
This insistence on the veracity of brotherly love is particularly striking since

the chorus completely misrepresents the conflict between appearance and
reality in the dealings between them: far from being moved by Thyestes
appearance, it is in fact Atreus who is able to deceive him with a false
aspectus.
Tarrant rightly compares this song to several Sophoclean choruses which
express hope immediately before a catastrophe,47 and the messenger will
shortly provide ample evidence of such a metabole. Note, however, the
peculiar status of the choruss remarks. After the dialogue of act 3 it is clear,
for instance, that the audience enjoys a superior degree of knowledge, which
connects it to Atreus and differentiates it from Thyestes and the chorus.
Thus the choruss belated mixture of belief, hope and fear must strike the
audience as a tragically pointless stance. The enormous disparity in levels
of knowledge, I would suggest, makes it difficult for us to identify with the
choruss feeling.
This choral ode, not unlike the first one, contains a number of potentially
contradictory or at least puzzling statements. In the first stanza, for instance,
it is remarkable that the actions that the chorus attributes to Pietas the
love which supposedly re-links Atreus and Thyestes hands are themselves
violent and cast an ominous light on the success of the reconciliation: Love
stays the steel, and joining their hands leads men, even against their will,
to Peace (5589: opprimit ferrum manibusque iunctis | ducit ad pacem Pietas
negantes). Negantes squarely emphasizes the brothers unwillingness (once
again mutual) to yield to peace, even if the chorus would lend it a concessive
force.
The second stanza displays again a sceptical inclination: this sudden lull
out of so great uproar, what god has wrought? (5601: otium tanto subitum
e tumultu | quis deus fecit?) a question which has a rather obvious answer
for the audience lends itself to irony. Appropriately, the ensuing lines indulge in a detailed, obsessed description of the dangers now apparently past,
and not of the new, apparently positive situation. As it suddenly shifts its
thoughts backwards, the chorus itself falls prey to the irresistible attraction
of painful reiteration. The wound of the recent bellum civile (562) cannot
be forgotten even in this apparent lull, and the chorus perceptively lingers
on the expectation of war rather than on actual fights, which have not
47
Aj. 693717; Trach. 63362; Ant. 111554; OT 1086109.
168
occurred so far. The concluding remark, that the fear of war is worse than
war itself (572: peior est bello timor ipse belli, worse than war is the very fear
of war) represents one of the few immediately authentic and believable
statements uttered by the chorus. Of course, it does mean rather more than
the chorus is aware of, since timor is still very much a present factor; but it
is important to recognize that this moment of authenticity and credibility
comes precisely when the chorus abandons its analysis of what it perceives
to be the present reality (which it is utterly inadequate to comprehend) and
privileges instead the self-evident emotional reality of fear. In other words:
when the chorus finally yields to the same force which has already subjugated the shadow of Tantalus and Thyestes. The force of remembering,
repeating and re-enacting is stronger than the peace supposedly at hand.
This lack of balance between past and present, and between fear and
solace, largely holds true for the following stanza, which contains an elaborate simile referring to natural forces, preceded by a positive statement on
the newly acquired peace (5736):
iam minae saevi cecidere ferri,
iam silet murmur grave classicorum,
iam tacet stridor litui strepentis:
alta pax urbi revocata laetae est.
now the swords dire threats have ceased; now the deep trumpet-blare is silent; now
the shrill of the clarions blast is quiet; deep peace has been restored to the happy
city.
Out of nineteen lines which form the simile, only four depict the sea finally
at rest; the bulk of it is occupied with a vivid account of the tempest which
precedes the calm. Even as this positive side is finally introduced, at line
588, it is with a hypothetical si rather than a temporal cum (58895):
si suae ventis cecidere vires,
mitius stagno pelagus recumbit;
alta, quae navis timuit secare
hinc et hinc fusis speciosa velis,
strata ludenti patuere cumbae,
et vacat mersos numerare pisces
hic ubi ingenti modo sub procella
Cyclades pontum timuere motae.
If the winds strength has failed, the sea sinks back calmer than a pool; the deep
waters which even a ship adorned with fully spread sails on both sides had feared
to cleave, now lie open even to a small pleasure-boat; and now one can count the
fish under water, right where a moment ago the Cyclades, shaken by a huge storm,
feared the sea.
Atreus rex
169
None of the positive statements stands without an immediate balance in the

opposite direction provided by a return to the negative past, accomplished
by quae at line 590 and hic ubi at line 594. This is yet another instance of the
choruss propensity to indulge more in the recollection of past fears than
in the enjoyment of the present. By its very tortuous, revolving nature, it
underscores the real lack of definitive closure which is inherent in the simile
and is then made explicit in the gnomic statement of 5967: no lot endures
long; pain and pleasure, each in turn, give place more quickly, pleasure
(nulla sors longa est: dolor ac voluptas | invicem cedunt; brevior voluptas).
Indeed, just as calm succeeds tempest, tempests come back over and over
again.48
The real significance of this section of the ode lies, then, more in the
implications suggested by its structure than in the often less than compelling
lexical choices, and this is consistent with the general interpretation of the
choruss attitude that I have outlined so far: the chorus does say several
useful and perceptive things, but almost invariably malgre soi, between the
lines rather than explicitly, with tragic irony rather than full awareness. For
the chorus, the storm followed by a period of calm is an apt analogy for
Atreus conversion, and on the basis of this false assumption it expands with
utter seriousness on the larger moral framework governing human actions:
fortune is mutable, powerful men should beware of sudden reversals, and
remember that they, too, are subject to divine punishment (596622). The
statements which hold true do so, however, in a very different sense to that
envisaged by the chorus itself, and many are simply reversed. The elaborate
discussion of divine retribution at 60714, for instance, culminates in the
certainty that superbi will be punished (61314: quem dies vidit veniens
superbum, | hunc dies vidit fugiens iacentem, whom the rising sun has seen
high in pride, him the setting sun has seen laid low), and the chronological
framework invites us to read this as a reflection among other things on
tragic time; yet, of course, the ultimate superbus, Atreus, will be far from
destroyed at the end of the tragedy.
The discrepancy between the choruss actual grip on events and its prominent presence in the play is never more evident than in its dialogue with
the messenger immediately after the third ode. The messengers arrival on
the scene provides powerful confirmation of the cyclical and yet largely unpredictable nature of events which the chorus had endorsed in the second
48
A topical image, to be sure, but it is worth comparing Soph. Aj. 66976, especially 6745:

|
, in the context of Ajaxs misleading acceptance
of the status quo.
170
part of the ode, and the messengers first line (623: quis me per auras turbo
praecipitem vehet | . . . ?, what whirlwind will drag me headlong through
the air?) is pointedly linked to the choruss last utterance (6212: res deus
nostras celeri citatas | turbine versat, god shakes our affairs in a swift whirlwind). In the scene that follows, the chorus is directly involved in the tragic
action, since it interrogates the messenger and reacts to his narrative as it
unfolds. Yet this active role is matched by the superficiality of the choruss
reactions. After insisting that the messenger deliver his news quickly (626,
633, 63840) and being informed of Atreus deeds, the chorus resumes its
dialogic function by asking a series of questions on points of detail; all the
questions are formulated along the same syntactical pattern, and all with
the partial exception of the third display little emotional involvement:
quis manum ferro admovet? (who lays his hand on the sword?, 690); quem
tamen ferro occupat? (whom, for all that, does he first attack with the steel?,
716); quo iuvenis animo, quo tulit vultu necem? (with what spirit, with what
countenance did the youth bear his death?, 719); quid deinde gemina caede
perfunctus facit? | puerone parcit, an scelus sceleri ingerit? (what did he then
do after the double murder? Did he spare one boy, or did he heap crime on
crime?, 7301).
These interventions do nothing to alter the assumption that the chorus is
tragically ignorant and superficial, not just unable to modify events (something which it is not expected to do), but also clumsy and unfocused in its
reactions, predictable and formulaic in its expressions of horror. In fact even
these emotional outcries are emphatically delayed, and the questions are,
until after the better part of the rhesis, the only reaction we hear. It is only
at line 743 when almost all of Atreus crimes have been exposed that the
chorus gives voice to its own emotions, with an exclamation whose stylistic banality underlines the inadequacy of the response: o saevum scelus!
(oh, savage crime!). Subsequently, it is still questions barely more
coloured with emotion which we hear from the chorus: an ultra maius
aut atrocius | natura recipit? (does nature admit anything greater or more
atrocious?, 7456) and quid ultra potuit? obiecit feris | lanianda forsan corpora atque igne arcuit? (what more did he manage to do? Did he perhaps
throw the bodies for wild beasts to tear apart, or refuse them fire?, 7478)
sound an almost ironic note because of the predictability of the answers
they will receive.
Shocked by the sudden disappearance of the sun, anguished by fear and
despair, in the fourth and final song the chorus acquires a more powerful
and convincing dramatic status. For the chorus this is already the end of
the play; Atreus crimes have been told in detail, and the exchange with the
Atreus rex
171
messenger has been a true anagnorisis: now the chorus knows how wrong it
had been in its optimistic assumptions about Atreus conversion. It is at this
point, therefore, that the chorus finally seems to come into its own as far as
both emotion and cognition are concerned: it knows all there is to know
and has a chance to react not simply with hope or fear both, in a sense, are
preliminary reactions but with a more forceful set of emotions. Yet again,
Senecas handling of the ode highlights precisely opposite implications, and
the choruss final appearance ultimately leaves in its trail more questions
than answers.
This ode, to begin with, is emphatically not what it could be expected to
be, namely a reaction to the rhesis. The chorus is laconic and inexpressive
when it hears the messengers chilling narrative. Even its questions, which
could after all reflect a desire to learn as many details as possible before
attempting a deliberate response, evidently fail to stir new emotions. The
ode limits its focus to the very last fact the messenger had told, the sudden
disappearance of the sun, which is described as a divine reaction of disgust
at Atreus crimes. This novel event startles the chorus and leads it to new
worries and new doubts: rather than concentrating on a set of events which
had just been narrated in detail and whose causes are by now clear, the
chorus wonders at length about a consequence which the messenger had
already briefly but definitively explained at 7835.
Thus it is under the usual light of detachment and bewilderment that
the chorus enters on stage once more, and, again, its first words are uttered
as questions (78993):
quo terrarum superumque parens,
cuius ad ortus noctis opacae
decus omne fugit, quo vertis iter
medioque diem perdis Olympo?
cur, Phoebe, tuos rapis aspectus?
Father of lands and heavens, at whose rising every star of the dark night flees away,
where, oh where do you turn your orbit and destroy the day in the middle of its
path? Why, Phoebus, do you snatch away your sight?
Tantalus attitude had been similar (14):

quis inferorum sede ab infausta extrahit
avido fugaces ore captantem cibos?
quis male deorum Tantalo visas domos
ostendit iterum?
Who drags me forth from the accursed abode of the dead, where I snatch at food
ever-fleeing from my hungry lips? What god shows Tantalus again the homes he
saw to his ruin?
172
The Fury, on the other hand, knew from the beginning that the sun might
well disappear: Look! Titan himself is in doubt whether to order the day to
follow on, and with his reins to force it towards its destruction (1201: en
ipse Titan dubitat an iubeat sequi | cogatque habenis ire periturum diem), and
Atreus is not worried, but pleased, that the earth is shrouded in darkness: so
that shame should not impede me, the day has retreated. Go ahead, while
the sky is empty! (8912: ne quid obstaret pudor, | dies recessit: perge dum
caelum vacat). Once again the contrasting camps of characters are opposed
to each other by (among other things) a different level of knowledge and a
different capacity to react effectively to events: in this case, moreover, the
contrast is sharpened by the otherwise dramatically inexplicable fact that
the chorus goes on wondering why the sun disappeared even after it has
been told.
The surprising way in which the chorus frames its intervention after the
rhesis makes the hypotheses that it formulates as it tries to understand the
causa (803) of such a novel event even more puzzling (80314). The emotional tone of these lines switches from the familiar examples mentioned
before (the sound of the bucina, the bewilderment of the farmer) to a much
loftier catalogue of possible mythical explanations, all centred around the
fight between Zeus and the Giants the archetypal exemplum of subversive
violence directed against a superior power. This, in turn, leads the chorus to surmise that far more wide-ranging consequences have to be feared,
and the next two stanzas alternate between the certainty of an unbearable
catastrophe and the suspicion that the world has indeed come to an end.
Thus the chorus is ready to consider the disappearance of the sun as a fatal
blow to the regular alternation of cosmic rhythm, and a sense of permanent
damage prevails: have the usual movements of the world come to an end?
will there be no more sunsets, no more sunrises? (solitae mundi periere
vices? | nihil occasus, nihil ortus erit? (81314)). After a graphic description
of the novel meeting between a bewildered Aurora and the sun insueto
novus hospitio (startled at such unwonted welcoming, 821) the chorus
returns to doubt and fear (82732):
sed quidquid id est, utinam nox sit!
trepidant, trepidant pectora magno
percussa metu,
ne fatali cuncta ruina
quassata labent iterumque deos
hominesque premat deforme chaos . . .
But whatever this may be, if only it were night! Our hearts tremble, tremble
shaken by great fear, lest all things fall down shattered by fated ruin, and once
again shapeless chaos weigh upon gods and men.
Atreus rex
173
as
But doubt and fear quickly turn into the certainty of a final ekpyrosis,
the subjunctive is followed by a long string of future indicatives (starting
with dabit at line 837) which culminate in the final ruin of the Chariot at
line 874 (ruet). In these thirty-seven lines over a third of the total ode
the detailed insistence on a series of specific astronomical disasters lends
support to the choruss belief that the end of the universe is imminent.49
This belief is yet again stated in more doubtful terms at first (87581), but
is finally acknowledged in the drastic tone of a gnome (87584):50
nos e tanto visi populo
digni, premeret quos everso
cardine mundus?
in nos aetas ultima venit?
o nos dura sorte creatos,
seu perdidimus solem miseri,
sive expulimus!
abeant questus, discede, timor:
vitae est avidus quisquis non vult
mundo secum pereunte mori.
Of many generations, is it ours which has been deemed worthy to be overwhelmed
by the sky, its axis upturned? Has the last day come in our time? Alas for us,
begotten with cruel lot, whether we have lost the sun or banished it! Away with
lamenting, go away, Fear: he who does not want to die when the world is dying
with him is too greedy for life.
Both tone and contents send strong signals of closure, and in fact this
is the choruss last appearance on stage (there is no exodos, as we will see
shortly); yet it is precisely the lack of closure that is highlighted in the rest of
the play. Not only have the mundi vices not been permanently altered and
the world has not come to an end, but the next two hundred lines will in fact
repeat from a different narrative point of view the final part of the events that
the rhesis had announced: in dramatic terms, that is, the banquet unfolds all
over again as Atreus, instead of taking the sudden darkness as an indication
of divine disgust, exploits it to further the completion of his plans. In
pointed contrast to the choruss insistence on closure, the play ends with
the promise of future retributions: TH. vindices aderunt dei; | his puniendum
vota te tradunt mea. | AT. te puniendum liberis trado tuis (TH. The gods
will be my avengers; my prayers deliver you to them for punishment.
AT. For punishment, I deliver you to your children, 111012). Use of the
49
50
Note that even as the chorus is ready to admit that a final end has arrived, it is still in doubt as to
why exactly that happens.
A gnomic closure is common in Greek tragedy: Kremer (1971) 11721, and, for further references,
Roberts, Dunn and Fowler (1997) 276.
174
future tense, together with the symmetry between Thyestes and Atreus
final lines, ensures that no final word has been spoken, and that the vices
of their enmity still have a long way to go: whether gods or humans will
enforce it, vengeance is the only prospect on which both characters agree.
Once again we see that the choruss sustained tone is undermined by
failure to really grasp what is happening. More pointedly, the description
of the conflagration of the universe, coherent as it is with Stoic belief,
conspicuously fails to represent the last word and again makes the chorus
look misguided in its moral loftiness and basic lack of understanding. In
this final song the chorus had sided with Tantalus and the satelles in its wish
that no further progress be made, that an end be put to the domino-effect
of nefas; that since the Furys and Atreus drive to repetition had already
been fulfilled, this at least be matched by a final repetition of a different
nature (note iterum at 831 and 833), the end of a cosmic cycle and the return
to chaos, and silence. None of these wishes comes true: the world goes
on, Atreus and Thyestes brace themselves for future reprisals, and the play
chooses to continue, and even to repeat the nefas it sings.
Although it has often been thought to embody the tragedys real message, the second ode does not cast the chorus in a different light. A more
optimistic interpretation, which sees in it the plays closest approach to a
positive statement of values whose beauty . . . remains deeply satisfying51
is possible only if we agree to see in a positive light the choruss detachment from the surrounding events. This would be an awkward line to
pursue, however, because Senecas choral odes in general, and those of
Thyestes in particular, cannot really be considered self-standing interludes
(embolima).52 Successfully or not, they attempt to establish a connection
with reality, and consequently display, rather than emotional autarkeia, joy
and fear, worry and hope. Thus the choruss weakness appears functionally
similar to the satelles unexcited and quickly dismissed attempt to restrain
Atreus, or to Thyestes own ambivalence in the choice between a properly
Stoic behaviour and the reality of his wishes and fears.
As it elaborates an ideal model of royalty just after the audience has witnessed the depth of Atreus deception, the choral ode emerges as a triumph
of wishful thinking over reality.53 Atreus and his satelles have painstakingly
51
52
53
Tarrant (1985) 138; see Davis (1993) 172: [Odes 2 and 3] establish a philosophic standard by which
the plays central characters, Atreus and Thyestes can be judged.
Pace Zwierlein (1966) 7680.
On contrastività as a fundamental feature of Senecan choruses see the important remarks by Mazzoli
(198687).
Atreus rex
175
mapped the future course of actions, and there can be no further doubt
that the Furys goals will be attained. In particular, the final part of the
dialogue focuses on deception and secrecy, as Atreus and the satelles consider Thyestes reactions to the invitation to return home, and Menelaus
and Agamemnons involvement in the ploy. In this context many keywords
already present in the first ode return to the forefront: fraus (312, 316),
dolus (318), fallere (320, 321). In fact the very last exchange between the two
characters on stage elaborates on the importance of keeping their secrets
well hidden, even from Atreus own sons (3323), and, we must infer, from
anybody else (3345). The chorus appears to be still hoping for a future
which is no longer attainable.54
A further unsettling implication emerges from a detailed comparison of
the standards advocated by the chorus and the behaviour of Atreus and
Thyestes.55 In its eloquent rehearsal of traditional Stoic topoi the chorus
predictably aligns itself with Thyestes own moralizing, unreliable as this
may be shown to be.56 But on closer inspection the choruss autarchic
view of power is paradoxically more in tune with Atreus than Thyestes
behaviour. It is Atreus, after all, who truly despises the fickleness of popular
favour and the external signs of royal status.57 Were it not for his ambition
(350: ambitio) he would be an almost perfect embodiment of the qualities
extolled in the ode. Thus, by opening itself to a paradoxical interpretation
which is clearly at odds with the choruss presumable authorial intention,
the song does more than reinforce the feeling that interpretation is fraught
with ambiguities; it also involuntarily sanctions, in one of the most eloquent
parts of the tragedy, the notion that real power can invariably turn language
to its own advantage, and that even the most hallowed of Stoic precepts are
not safe from tendentious exploitations a` la Atreus.
The choruss reaction to and understanding of events is perhaps the most
important aspect of its characterization. The choruss lack of defining traits,
its constant mood of doubt and uncertainty, its tendency to misinterpret,
54
55
57
The contrast with the preceding scene is even greater if lines 3368 are retained (tandem regia nobilis,
| antiqui genus Inachi, | fratrum composuit minas); they are deleted by Richter (1902) (followed by
Sutton (1986) 401) because they are inconsistent with the preceding scene and anticipate 54651,
but are retained by Zwierlein and Tarrant. Calder (1989) suggests that they are spoken by the chorus
as it comes back on stage probably at 330 and utters its real beliefs (from 339) only after Atreus
and the satelles have left. This explanation presupposes that the chorus already knows that Atreus
plans to pretend to welcome back Thyestes, which is unlikely given that the plan has just been
hatched.
56 See above, pp. 141ff.
I am indebted here to Davis (1993) 1768.
Lines 3447 and 3537.
176
the feebleness of its emotional responses all these factors conspire to

subtract a great deal of dramatic and moral appeal from its noble-sounding
and apparently inspired ethical considerations. Thus the solutions it offers
on a variety of levels the true nature of power or the preference for
a retired, ataraxic life, its faith in the gods presence and providence, the
certainty of retribution are all fatefully undermined.
It is interesting to note that Senecan plays never give the chorus the last
word,58 thus depriving it of any opportunity to deliver a final evaluation
of the events. The choruss feelings and thoughts are only voiced between
events, whether or not these have been directly witnessed or even correctly
understood. Not even in structural terms does the chorus enjoy a privileged
platform. It is small wonder, then, that its recipes for a better life appear in
the end to be more of an exercise in abstract morality than a compelling
indication of viable options.
58
The only exceptions are represented by the certainly un-Senecan Octavia, and the dubious Hercules
Oetaeus; Leo (1897) 512. See p. 68.
chapter 5
fata se vertunt retro

(Seneca, Agamemnon 758)
i
The time has almost come to bring the reading of Thyestes, if not yet
to a conclusion, at least to a point where we can take a comprehensive
view of the main issues that I have analysed in previous chapters. Before
doing so, I would first like to explore a set of related topics which play an
important role not just in Thyestes but in several other Senecan tragedies.
Accordingly, in this chapter I will largely move away from Thyestes and offer
a thematic reading focusing mostly on other plays: Hercules furens, Troades,
Agamemnon, Medea and Oedipus. In the next chapter I will extend the
argument developed here, in order to situate my final analysis of Thyestes
within a broader context.
I propose to look first at certain peculiarities of Senecas treatment of
dramatic time. There seems to exist a broad consensus, implicit or explicit,
that many of the plays temporal structures display markedly idiosyncratic
features. Unfortunately, this is often taken as further evidence for the theory
that, by abandoning the conventions of Attic drama, Seneca has irretrievably
adulterated the pure forms of tragedy, so that his treatment of time, like
many other aspects of his dramatic technique, testifies to a decadence in
the evolution of tragedy. It is well established that Seneca breaks away
from many fifth-century conventions:1 the unity of time and space, the
rigid delimitation of the time allotted to the tragic action, the coherent
succession of scenes in an undisturbed temporal continuum these are
all rules to which Seneca finds surprising (and, I think, highly effective)
alternatives. The notion that many of his plays offer detached tableaux
1
The classic treatment of this topic is Tarrant (1978).
177
178
rather than an organic plot is predicated precisely on the sometimes loose,

unconventional, often puzzling temporal connections established between
acts. It is incumbent upon us to read these disjunctions and discrepancies
as markers of meaning. I will take my cue from some peculiar treatments
of dramatic time to investigate their semiotic and thematic significance:
what does it mean for Seneca to subvert linear chronology, and what role
does time play in the configuration of Senecas tragic thought and tragic
writing?
ii
At the end of her harsh exhortation to Tantalus in the prologue, the Fury
orders him to remain on earth and to watch the doomed banquet which will
eventually conclude the tragedy: go, fill up your fasting; let blood mixed
with wine be drunk before your eyes; I have found foods which even you
would want to flee stop, where are you rushing headlong? (657: ieiunia
exple, mixtus in Bacchum cruor | spectante te potetur; inveni dapes | quas ipse
fugeres siste, quo praeceps ruis?).2 Shortly thereafter, however, she dismisses
the ghost and sends him back to the underworld: go to the caves of the
underworld and your familiar river (1056: gradere ad infernos specus |
amnemque notum). Between these two seemingly contradictory orders3
stands Tantalus short-lived and ultimately ineffectual rebellion (6883a,
86b-101), as he is overcome by the Furys excruciating tortures (96101). It
has been suggested4 that at 1056 the Fury is relenting precisely because
of Tantalus emphatic reaction: first he must wreak havoc in the house of
the Pelopidai (83: ante perturba domum); only afterwards will he be allowed
to leave.5 Difficult as it is to imagine the Fury suddenly surrendering to
Tantalus complaints, this is not the only problem raised by such a conciliatory explanation. In the lines that follow Tantalus unwilling agreement,
and conclude the prologue, the Fury vividly describes the consequences of
his pollution. Verbal repetitions convey her excitement6 (1014):
2
3
5
6
See pp. 45ff.

The inconsistency was first noticed by Lesky ((192223) 533), who explained it as being a result of
Senecas imperfect adaptation of his Euripidean model. Hine ((1981) 268) believes that the Furys
order at 1056 is ironic, and is uttered by the Fury alone on stage, after Tantalus has in fact already
entered the house and polluted it. See Picone (1984) 28, n. 48.
Steidle (194344) 257 and Picone (1984) 28. Lesky ((192223) 5337) puts the discrepancy down to
Senecas departure from his Euripidean model: in Euripides Thyestes, in fact, the ghost would have
remained inside the house for the duration of the whole play. See Anliker (1960) 278.
On this point see Shelton (1975) 2589.
See Tarrant (1985) 103 for a discussion of Calders suggestion (1984) that the repetition marks the
strokes of whip which the Fury inflicts upon Tantalus. The suggestion would probably have come
from Virgils Dido (Aen. 4.660), who stabs herself with the words sic, sic iuvat ire sub umbras.
179
hunc, hunc furorem divide in totam domum.

sic, sic ferantur et suum infensi invicem
sitiant cruorem. sentit introitus tuos
domus et nefando tota contactu horruit.
This, this rage distribute throughout the house! Thus, thus let them be dragged,
and in turn, as enemies, let them thirst after each others blood. Your house feels
your entering, and has recoiled in horror from your unutterable contagion.
It is at this point that the Fury is finally satisfied (1057):

actum est abunde. gradere ad infernos specus
amnemque notum; iam tuum maestae pedem
terrae gravantur;
Enough! More than enough! Go to the caves of the underworld and your familiar
river; already your step falls heavy on the saddened earth.
While actum est abunde reinforces the Furys affiliation to her Virgilian
model,7 it also establishes a meaningful point of reference within the play.
In act 2, after selecting the best form of revenge against his brother, Atreus
expresses his contentment in much the same words (27980):
bene est, abunde est: hic placet poenae modus
tantisper.8
This is good, more than enough. I like this way of punishing him for the moment.
The parallelism between these two scenes goes beyond the repetition of
abunde. Both the Fury and Atreus start by striving to find a revenge of
unprecedented cruelty, and the Furys wish that the Thracian nefas be repeated on a larger scale (567: Thracium fiat nefas | maiore numero, let the
Thracian crime be done, but multiplied) is mirrored at the same relative
point in Atreus own speech: non satis magno meum | ardet furore pectus,
impleri iuvat | maiore monstro (the frenzy burning in my breast is not
great enough; some greater horror must fill me, 2524). Moreover, lines
1034 are a careful, explicit reworking of two lines in Ovids Philomela
7
8
At Aen. 7.552 terrorum et fraudis abunde est is uttered by Juno in direct response to Allecto, just before
the latters dismissal (559: cede locis).
See Braden (1985) 45 for the suggestion that bene habet, peractum est at Oed. 998 (cf. Her. F. 1035;
Ag. 901 with Tarrant (1976) 343 and references to comedy; Med. 1019; Her. O. 1457, 1472) might
be connected with the language of gladiatorial games. It would indeed be tempting to see the Fury
and Atreus as game-directors extraordinaires, a point which could well be supported by Calders
suggestion noted above (n. 6). It is worth noting that abunde est appears in the Aeneid only at 7.552.
Seneca, who has Atreus use the expression a third and final time at line 889 (bene est, abunde est, iam
sat est etiam mihi), multiplies its model, thus effectively depriving it of much of its value. On the
Virgilian construction see Fordyce (1977) 160.
180
episode:9 ut sensit tetigisse domum Philomela nefandam, | horruit infelix

totoque expalluit ore,10 and in the corresponding position in Atreus speech,
before he in turn exclaims bene est, abunde est, we find a direct reference to
the same story: animum Daulis inspira parens | sororque (inspire my soul, o
Daulian mother, and sister, too, 2756). These verbal and structural parallels, therefore, highlight the fact that Atreus is closely following in his own
speech the same sequence of thoughts and actions as was displayed by the
Fury in the prologue.
As the Furys description of the extraordinary outcome of Tantalus intervention acquires a cosmic dimension, and as nature forgets her habits,
she crowns her crescendo by pointing out that even the sun is uncertain
whether it should force the day to continue in its appointed course: Look!
Titan himself is in doubt whether to order the day to follow, and with
his reins to force it towards its destruction (1201: en ipse Titan dubitat
an iubeat sequi | cogatque habenis ire periturum diem).11 This image clearly
prefigures the withdrawal of the sun from its regular course12 and becomes,
later in the tragedy, the icon of the nefas which has been accomplished.13 It
is possible to argue that this elaborate parallelism denotes the Furys ability
to predict in great detail the events she has herself caused. Indeed, this
appears to be the most economical solution, since it implies no particular
idiosyncrasy on Senecas part. An alternative solution to the correlations
and discrepancies mentioned above would be to interpret the Furys overarching awareness of the events which will later unfold on the stage as a
manifestation of prophetic foreknowledge.14 This is certainly the case in
the prologue to Agamemnon, where it is Thyestes turn to anticipate the unfolding of the events. A comparison between the two prologues shows the
prophetic nature of Thyestes words, who welcomes the realization of an
uncertain prophecy (Ag. 38: sortis incertae fides) as the king returns home
9
10
11
12
13
Met. 6.6012.
Tarrant ((1985) 103) provides a detailed analysis of the analogies and differences between Senecas
and Ovids lines.
The use of the participle in se periturum indicates the inevitability of the action (Tarrant (1976)
178; see Tarrant (1985) 106).
The premature, unnatural setting of the sun as a mark of horror at Atreus deeds is mentioned by the
messenger (776), the chorus (789), Atreus (892) and Thyestes (990). We must assume that the sun
sets while Atreus carries out the infanticide, and the whole of act 4 must be taking place in at least
partial darkness (the messengers words at 6235 would thus have a somewhat paradoxical flavour).
When the chorus opens the following ode in a tone of surprised anguish (78991), they are not
witnessing the disappearance of the sun afresh, but rather commenting on it. The sun, supplanted
in its role by a star-like Atreus (8856), is still hidden when he meets Thyestes, as both brothers
remark (8912 and 9901). See Hine (1981) for the opinion that similar repetitions are standard
narrative devices.
14 Picone (1984) 32, n. 57.
I return to this point at the end of the chapter.
181
to be murdered: now he is near at hand to give his throat into his wifes
power. Now, now shall this house swim in blood other than mine (434:
adest daturus coniugi iugulum suae. | iam iam natabit sanguine alterno
domus). Both the careful arrangement of verbal tenses, and the final notation that Thyestes sojourn on earth is delaying dawn (536: sed cur repente
noctis aestivae vices | hiberna longa spatia producunt mora, | aut quid cadentes
detinet stellas polo? | Phoebum moramur: redde iam mundo diem, but why
suddenly is the summer night prolonged to winters span? or what holds
the setting stars still in the sky? We are delaying Phoebus: give back the day
now to the universe) exclude the possibility that the prologue is temporally
coextensive with the rest of the play. Thyestes video (46) should therefore
be considered a variation on Atreus boasting that he can picture in his
mind the whole imago caedis.
At the opposite end of the spectrum of possibilities, Shelton has suggested
that the Fury actually observes in a compressed period of time all the
events which the tragedy will gradually present to the audience.15 The whole
action of the tragedy Atreus plotting his revenge, the double murder, the
banquet would be encompassed in the prologue. This, I hasten to add,
would be perfectly in keeping with Junos metadramatic aspects: as befits
the author of the story, she watches all the phases of the plot as they rapidly
unfold in front of her eyes before they are shared with the audience. Atreus
will acknowledge that he can anticipate the whole sequence of his revenge
in his mind: tota iam ante oculos meos | imago caedis errat (already before
my eyes flits the whole picture of the slaughter, Thy. 2812).
Both solutions involve substantial difficulties. The prophecy theory
falters on the discrepancy between the two orders given to Tantalus by the
Fury, first that he stay to watch the banquet, then that he return to his
usual abodes. On the other hand, the fact that the prologue is taking place
just before dawn, while several references in the rest of the play make it
clear that we are in the middle of the day, make it difficult to assume that
the clock actually turns back at the beginning of act 2. One can see the
appeal of the claim that Seneca, as usual, just messed up his note cards
while attempting to juggle one too many models at a time.16
The fact remains that repetition and parallelism play a key role in the
structure of the play. Moreover the Furys orders notwithstanding the
plays later attempts to establish an almost regular internal chronology
do jar the audiences perception of temporal flow. The solution to this
puzzle probably lies in accepting that the Furys orders are intrinsically
15
Shelton (1975), (1978) 1720.
16
Lesky (192223).
182
contradictory. Act 2 does in fact repeat the structure of the prologue, as

the play ostensibly observes the regular conventions of the passing of time.
There is enough evidence in other Senecan plays, as we will see shortly, to
argue that Seneca is deliberately questioning a strictly mimetic notion of
linear dramatic time. The inconsistency between the Furys two orders is a
telling indication of the fact that we should expect an idiosyncratic treatment of time. In other words, if our (human) perception of events cannot
dispense with the usual notions of succession and regularity, there is a level
at which gods (and the playwright) devise the unfolding of actions with
a synoptic, all-encompassing knowledge which we can only dimly fathom
from our limited perspective. Actions, in a sense, are always already determined by a divine masterplot, a consideration which does not eliminate
human responsibility (far from it),17 but inscribes it in a much larger and
uncontrolled context of divine foreknowledge and planning. By disturbing
the audiences perception of time, Seneca marks repetition as an essential
component of tragic actions and of our theatrical understanding of them.
The confidence that actions unfold over an unwavering temporal continuum (and, therefore, according to a well-defined causal chain) is shattered
even by this limited disruption to the expectation of a regular, forwardlooking temporal flow. At the beginning of the second act of Thyestes the
audience is presented with events which seem to follow on directly from
the action of the prologue, but which in a very significant (if not strictly
literal) way should be traced back to a point in time in parallel with the
prologue itself. The movement from full daylight to the uncertain shades
of dawn marks, for the humans, a clear step forward in a temporal continuum. Enough clues emerge, however, to alert the careful spectator to
the fact that in an important sense we are moving backwards, returning to
a point in time that we have already witnessed once: the framing structure analysed in chapter two18 acquires a new, pregnant dimension: the
second layer of the plot is not only structurally embedded in the first, represented by the prologue, but is to a certain degree temporally coextensive
with it.
One of the most significant outcomes of this dramatic technique consists in a further blurring between the responsibilities of the characters on
the stage and the audience, and a deeper elision of the possibility of neatly
defining innocence and guilt. The audience is involved from the very beginning in the Furys scheming, and subsequently treated, either willingly
17
18
It is difficult to confine the significance of this treatment of dramatic time to the texts desire to
emphasize human responsibility, a point well made by Shelton (1975) 2637.
183
or malgre soi, to a detailed account of events already foreseen and foretold.

The audience could (should) leave. By continuing to watch, it can no longer
claim innocence. It is, in effect, an accomplice to the nefas on the stage.
Such a disruption of the conventions of tragic time is unparalleled in ancient drama19 and represents a form of self-conscious expression which
finds appropriate parallels in only a limited number of modern texts. Before twentieth-century epic theatre started to exploit discrepancies in the
treatment of dramatic time in order to draw an audiences attention to
the fictional nature of their experience,20 only a few authors had explored
the possibilities of a non-linear arrangement of time.21
The case of Thyestes, however, is not isolated in the Senecan corpus.
The prologue to Hercules furens is dominated by Juno, the only goddess
(or god, for that matter) ever to enter the realm of Senecas tragedy, and a
character rich with metadramatic resonances. These are particularly clear as
Juno, after lamenting the fact that no ordeal, no matter how extraordinary,
can worry Hercules or stop him (306), explicitly addresses the force of
her inspiration in order to discover, finally, an apt instrument of revenge
(757):
perge, ira, perge et magna meditantem opprime,
congredere, manibus ipsa dilacera tuis:
quid tanta mandas odia?
Onward, my anger, onward! Crush this overreacher! Grapple with him, tear him
apart with your own hands. Why delegate such hatred?
There follows a long invocation to the Eumenides (868) which leads a

few lines later to a direct description of their inceptive function (1006):
incipite, famulae Ditis, ardentem citae
concutite pinum et agmen horrendum anguibus
Megaera ducat atque luctifica manu
vastam rogo flagrante corripiat trabem.
hoc agite, poenas petite violatae Stygis;
concutite pectus, acrior mentem excoquat
quam qui caminis ignis Aetnaeis furit.22
19
20
21
22
See Taplin (1977) 2904, 3779 for important remarks on fifth-century tragedy. Little is known
about later developments, but see Taplin (1977) 49 with n. 2.
I will return to this particular aspect in the next chapter.
On the issue in general see Pfisters perceptive remarks (1977) 246ff., esp. 276ff.
The symmetrical quality of the language is an important indicator of the plotting strategy that Juno
has in mind, because Hercules will have to inflict on himself the revenge desired by the goddess.
Junos furor will derange Hercules, and his own furor will destroy him. The revengeful inspiration
that the goddess seeks for herself she will pass on to her unaware victim.
184
Begin, handmaids of Dis, brandish the blazing pine torch violently. Let Megaera
lead your troop, fearsome with snakes, and snatch a huge beam from a blazing
pyre in her baleful hand. To your work: avenge the desecration of the underworld!
Rouse your hearts; scorch your minds with fiercer fire than that raging in Etnas
furnaces.
Junos role in this prologue is in many respects similar to the Furys and
Atreus exhortation in the first two scenes of Thyestes. But another structural
element forcefully connects the two plays and more directly interests us here.
The prologue to Hercules furens describes various actions in the present and
past tenses, which the rest of the play will represent anew. Juno sees,
while she speaks, all the events that the play will slowly unfold before the
audience. A very provocative instance of this occurs early in Junos speech.
At lines 66ff. the goddess is worried that Hercules will bring his attack
against the sky, driven by a desire to dominate the entire universe: nec in
astra lenta veniet ut Bacchus via: | iter ruina quaeret et vacuo volet | regnare
mundo (668: and he will not reach the stars by a gradual approach, like
Bacchus: he will forge a path by destruction, and he will want to rule in an
empty sky). In the course of a few lines, as Juno indulges in the description
of previous Herculean deeds, the future challenge has already taken place:
at line 74 the future quaeret turns unexpectedly into the present quaerit:
quaerit ad superos viam.23 Moreover, at line 64 Juno states that Hercules
deserves to be punished because she is afraid that he will attack Olympus.
Yet Hercules does so only as a consequence of the madness the goddess has
inflicted on him. Juno seems able to conflate, in her divine epistemological
omnipotence, cause and effect, crime and punishment.24
The most provocative instance of the phenomenon concerns the return of
Hercules from the underworld. At 4752 Juno testifies to some of Hercules
achievements:
effregit ecce limen inferni Iovis
et opima victi regis ad superos refert.
parum est reverti, foedus umbrarum perit:
23
24
The rapid succession of the two moments in which Juno foreshadows Hercules attack (67: iter ruina
quaeret) and actually sees it happen (74: quaerit ad superos viam) provides a pertinent instance of what
has been dubbed temporal compression (Zwierlein (1966) 29). This and similar passages (see Tro.
3512) are offered by Zwierlein as evidence that the tragedies were not performed, since the author
broke away from the conventions of a realistic treatment of time. In Greek tragedy, he notices, the
arrival of characters on the scene is always carefully arranged in such a way that a reasonable amount
of time elapses between the decision to summon them and their appearance, while a text written
for recitation rather than performance, he argues, would be less interested in preserving a strictly
realistic arrangement of time, and such compressions could be easily glossed over in the audiences
imagination. I agree with Calder (1970) that recitation increases, rather than reduces, the need for
clarification in such passages, while pauses on the stage could easily be filled by different means.
See Braden (1985) 48.
185
vidi ipsa, vidi nocte discussa inferum

et Dite domito spolia iactantem patri
fraterna.
See, he has broken through the gates of nether Jove, and brings spoils of triumph
over that conquered king back to the upper world. It is not enough to return:
the terms governing the shades have been breached. With my own eyes I watched
him, after he had shattered the gloom of the underworld and subdued Dis, as he
showed off to his father spoils won from that fathers brother.
The hero has already left Erebus dragging Cerberus with him, and has
flaunted to Jupiter the victory he obtained against the gods own brother
(512). Hercules, however, is still described as inhabiting the underworld at
the opening of act 2, and he will not proceed upwards until line 520, when
Amphytrion exclaims (5203):
cur subito labant
agitata motu templa? cur mugit solum?
infernus imo sonuit e fundo fragor.
audimur! est est sonitus Herculei gradus.25
Why is the shrine rocking and shaking with sudden movement? Why is the earth
rumbling? A thunderous noise comes from the depths, from the underworld. We
are heard! It is the sound of Hercules step.
The choral ode that begins immediately after this remark seems to occupy
the time needed by Hercules to complete his ascent: at the beginning of
act 3 his initial words are consistent with the notion that he has just reemerged from the underworld. In this instance it is difficult to assume that
at 4751 Juno is simply foreshadowing future events, since she takes care to
insist, by the emphatic repetition of vidi, that she has already seen Hercules
return.26 Here, again, there is a strong suggestion that the prologue covers
all the events up to Hercules bout of madness (which in the play will begin
at 926), and that at the beginning of act 2 we turn back to a moment in
time when he is still in the underworld.
The unfolding of Hercules return is further complicated by Theseus
remarks at 81321, where he harks back to the span of time covered
by the second choral ode27 and informs Amphitryon that Hercules had
25
26
27
Shelton ((1978) 201) suggests that the same treatment of dramatic time applies to yet another
instance. In her opening speech Juno provides a detailed analysis of the psychological processes
which will lead Hercules to his final, self-defeating ordeal. This is a case, however, where it is
quite appropriate to suspect that Juno is simply foreshadowing future events thanks to a more
comprehensive form of knowledge.
The anaphora of vidi is usually referred to events (often of a cruel nature) actually witnessed by the
speaker. In Senecas tragedies see Her. F. 2547; Tro. 170 (vidi ipse, vidi); Ag. 656 with Tarrant (1976)
294; Her. O. 207. See Hansen (1934) 40. An influential model must have been Virg. Aen. 2.499501.
Wagenvoort (1933) 177.
186
returned with a reluctant Cerberus to the mouth of the river Tenarus, not
to Thebes. The latter detail, though consistent with the internal chronology of events once we accept the proleptic nature of Junos prologue, seems
to contrast with Amphitryons words at 5203, provided that we assign
them a specific topographical function.28 In any event, Theseus narrative
also disrupts the expected overlap between the level of the story and the
level of the plot, albeit one which is made acceptable by its insertion in a
rhesis.
Regardless of whether Juno is a vatic character, or whether the prologue
really covers a temporal expanse which successive phases of the tragedy
will repeat, as the curtain rises on the first act the audience will inevitably
experience the anxious feeling of dejà vu. Junos words do not simply anticipate the course of events described in the tragedy, but offer a number
of specific details which will make those events, as they actually happen
for the spectators, repeat a masterplot they already know, hence disrupting
the linear arrangement of events which is a cardinal feature of dramatic
texts.
Junos words do not describe reality as much as they in fact create it:
she envisages Hercules challenge, and behold the challenge really takes
place shortly thereafter. In this respect she embodies the creative power of
the author, and the plot that she conceives is the tragedy which happens.
This is one more metadramatic feature of the prologue, which functions as
a preliminary, superordinate phase of the play, encompassing the rest of the
drama from a superior, proleptic viewpoint which replicates the authors
perspective.
The limited examples I have presented show a treatment of dramatic time
which emphatically breaks away from the linear structure that is consistently
observed as the norm in ancient drama. By creating a framework in which
certain sections of the play appear to revolve back to a point in time that
has already been treated, and by substituting iteration for linearity, these
tragedies make repetition an essential modality of tragic representation.
By emphasizing repetition and serialization, the tragedy produces a strong
alienating effect.29
This alienation is largely determined by the conspicuous short-circuiting
of a literary canon whose rules had so far been scrupulously observed
28
29
I agree with Caviglia ((1979) 238), who rightly argues against the suggestion that line 523 or even the
whole section 5203 should be deleted because they contradict 81317; see Leo (187879) 375 and
Lindskog (1897) ii. 41. Fitch (1987) 252 argues that [a]s elsewhere, Sen[eca] is ignoring consistency
in favor of immediate dramatic impact.
I will discuss in the next chapter how these alienating structures impact on the audience.
187
(at least, I should add, this is how it appears).30 The treatment of internal time is one of the most radical ways in which tragedy and epic can
be differentiated, because creative manipulation of the temporal links between events is widespread in epic. The importing of typically epic narrative
techniques into a dramatic text inevitably produces a jarring effect. Epic,
with its complex, multiple plot-lines, had explored early in its development the possibility of narrating simultaneous events in a linear scheme.
Homer, here as elsewhere, proposed a solution and imposed a rule. Simultaneous events would have to be separated and juxtaposed, arranged serially
one after the other.31 Virgil, too, exploits the potential of linear arrangement, creating in his poem a complex web of temporal intersections which
are essential to an understanding of the narrative.32 One should resist the
temptation to read the development of this aspect of epic technique as the
progressive complication of a supposedly simple archetype (Homer), and
to detect in Virgil the first symptoms of narrative disorders magnified by
his Silver successors; in this particular instance, however, it is in Ovid that
we find the first extended, systematic deviation from the recognized decorum of temporal linearity.33 Ovid extensively explores the signifying value
of a fragmented and often confusing representation of time. His use of
and prolepsis
transcends the limited function that they have in the
analepsis
Aeneid, where they are rather rigidly controlled by the internal narrator(s),
and become a pervasive feature of the text, poised to exploit the interplay
of different narrative times to the full. If the Heroides stand as a particularly
effective example of this technique,34 the Metamorphoses complicate and
30
31
32
33
34
We do not know enough, in this respect as in others, about post-classical drama, and it is reasonable
to assume that significant variations in the treatment of time did occur. It is unlikely, however, that
Seneca could have found a direct model for his treatment of dramatic time.
See Stanley (1993) 6: Like the Homeric sentence, Homeric narrative presents an ongoing series in
which each successive action seems at first glance to receive equal status just as time, in direct
narrative, generally moves forward in an uninterrupted flow in which even simultaneous events
seem to be treated as a linear sequence. The phenomenon was first described by Zielinski (1901),
hence the definition Zielinskis law (on which see the reservations of Rengakos (1995)). The most
recent and exhaustive treatment is offered by Stanley (1993), especially 69, 1535 and, for further
bibliography, 306, nn. 1819. See also Whitman and Scodel (1981) and Janko (1992) 1501.
On the use of interea in the Aeneid see Reinmuth (1933). As Heinze ((1928) 306, n. 31) remarks, this
is a case of loose use of interea, which approximately equals now. But the ambiguities, as Heinze
points out, remain. Three books of the Aeneid (5, 10 and 11) have interea in the first line, but only
in 5.1 does the meaning meanwhile appear to be fully active. See Harrison (1991) 58 and Kinsey
(1979) 2634.
On Ovids treatment of time see now Feeney (1999) and Zissos and Gildenhard (1999). Virgil can
also be seen to experiment with chronology: see Heinze (1928) 305 for a discussion of how difficult
[Virgil] found it to deal with [ . . . ] simultaneous actions. It need hardly be remarked that the
simplicity attributed to Homer is little more than convenient shorthand.
See the fine discussion in A. Barchiesi (1992) 1619.
188
muddle the temporal linearity established at the beginning35 to such an

extent that they end up offering what we might call a cubist representation of time.36 It is tempting to assume that Ovids sustained engagement
with temporal (dis)continuity is a metaliterary register of his posteriority.
Going back in literary time as Ovid does in his intertextual dialogues
becomes one of the ways in which the text performs its ideological negotiation of the past. (In the next chapter I will attempt to situate intertextuality
within the painful, oppressive ideology of the past that Senecan tragedies
seem to privilege.37 )
In epic, the narrators boundless power facilitates the organization of
complex temporal structures into a clear, intelligible arrangement. Theatre,
on the other hand, was bound to the rule of the hic et nunc, to Aristotles
dictum that it should be made up of actions, not of stories,38 and that
narrative interventions should be limited to specific, self-contained loci
such as the messengers speeches. The temporal discontinuities that we
have observed are evident enough to elicit an active interpretative gesture
from the audience, yet subtle enough to dawn on us only gradually. For
instance, the Furys speech in the prologue to Thyestes contains internal
discrepancies that should alert the audience to the fact that there are
at the very least different modes of perceiving time, and that to rely
exclusively on the notion of linear, unidirectional movement is perhaps
inappropriate. But it is only as Atreus own speech unfolds in the following
act that the audience is made to recognize a number of striking parallels
between the two characters, and to note the causal relationship between
their utterances and actions. At an even later stage when, for instance,
Atreus voices for a third time the refrain bene est, abunde est we perceive
both the internal correlations between causes and events and the imperfectly
linear structure of time that subtends them. If this deconstruction of linear
narrative structures can never compare to its much more intense modern
counterparts, a comparison with the latter retains nonetheless a considerable
heuristic value, as I will show in the next chapter. What we can certainly
say about the structure of Thyestes is that, once we perceive the repetition of
a key phrase in a specific moment of the dramatic action, we are inevitably
gripped by a sensation of dejà vu, coupled with the realization that again
we have been moving backwards. The Fury herself, it should be noted again,
35
36
37
See Ludwig (1965) 56.

Fasti, too, are obviously engaged in the treatment of time, juxtaposing discordant scenes which twist
and turn below the surface of temporal linearity. The most stimulating recent readings of the poem
are A. Barchiesi (1994) and Newlands (1995).
38 Poetics 1449b26.
189
reversed through a well-trodden literary background that is prominently

inhabited by Euripides and Virgil.
In Thyestes, the upsetting effect of such a sensation stems primarily from
the denial of meaningful closure encoded in the structure of the prologue
and in its relationship with the rest of the play. Not only does the prologue tear open wounds which its most prominent model, the Aeneid , had
struggled hard to heal (or, at least, to give the consoling illusion of doing
so), but it fashions, in its compact brevity, the illusion that the new ordeal
can be concluded swiftly, if not altogether painlessly; in other words, the
illusion that the audience could be treated this time only to the superhuman level of deliberations without having to face in excruciating detail the
actual unfolding of events: imagine the second half of the Aeneid with only
divine meetings, and no battles. But the closural sign which brings the
Furys speech to an end is also rapidly exposed as illusional. The tragedy
starts all over again in Atreus chambers, and this time we are going to see
it all, in its horrific, unabridged version.
To deny closure means that everything will happen again and again, that
regression will know no end. The denial of closure encoded in the contrast
between the apparently final ending of the inner level of the plot and the
emphatically open finale of the play as a whole turns out to be anticipated
at its beginning, too.39 In Thyestes, the non-linear organization of dramatic
time and the complex framing structure of the play as a whole allegorize
the force of regressive repetition which can be seen as the tragedys driving
dynamic. Indeed, regression, at different levels and in different guises, is
arguably the single most relevant operating principle of Thyestes and of
Senecan tragedy. If the prologue encompasses in its short frame a much
larger portion of the tragic action, then we are bound to perceive the
successive stages of the play, in which the actions unfold one by one, as
a repetitive, regressive exploration of an already accomplished misdeed.
There is no progression in the play: from the end of the prologue onwards,
the tragedy is trapped in the repetitive exploration of the consequences
of Tantalus pollution. The plot of Thyestes, we realize, is bent backwards,
not forwards: it does not aspire to the consoling sense of progress crowned
by final resolution, to reach a meaningful point of closure. The lack of
closure inscribed in the end of Thyestes, with its call to further revenge,
and therefore to further engagement with the past, mirrors the illusory
closure of the prologue and highlights the overall regressive movement of the
plot.
39
190
Atreus, the protagonist of the play, embodies the enduring pleasures

and pains of memory. His overarching goal throughout the tragedy is to
avenge past wounds, and his conduct allegorizes the virtues of memory,
even obsessive memory. His superiority vis-à-vis Thyestes is based on the
fact that, unlike his brother, he firmly believes that the past cannot and
will not be erased. Thyestes is tricked into thinking that Atreus has finally
forgotten the slights he has endured and is ready to turn a new page in
their relationship. His dogged determination proves that the past cannot
be undone, and its memory lives on to determine future actions. But Atreus
knows no forgetting. While his revenge in the play could be construed as
a means to actively reshape the past (this time, after all, Atreus will win),
he is the first to recognize, both during the play and, most emphatically, at
its end, that the spiral of revenge and counter-revenge cannot find a final
resting place, and that even his current victory must be seen in the light of
a cyclical arrangement of history.
Prevented from moving forward, Thyestes is condemned to oscillate between returning to and returning from. The future, if we can now call
it that, promises only the repetition of a well-known pattern: as Thyestes
consummates his revenge, he will merely repeat once more the fixed script
which holds his whole family hostage.
iii
Repetition and regression sustain the dramatic tension in Troades. Pained
reflections on the hopes and despairs of repetition loom large throughout
the tragedy and especially in two crucially important scenes, the altercation
between Agamemnon and Pyrrhus in act 2 and the dramatic confrontation
between Andromache and Ulixes in act 4.40 But the tragedy also engages
its audience in a complex evaluation of the effect and nature of repetition
from the point of view of dramatic structure. Troades has long been considered to mark Senecas resignation to a loose structure with very limited
attempts at a unified plot. The result, not unpredictably, has been explained
mainly on the basis of the unsuccessful mixture of disparate Greek sources,
compounded by the bold, but confusing, decision to unite in one play the
fates of Polyxena and Astyanax.41 I propose to set aside for the moment a
40
41
In my reading of Troades I have found Fanthams (1982) and Boyles (1994) commentaries very
valuable, and the articles by Schetter (1965) and Owen (1970) particularly useful.
On sources see Fantham (1982) 5075 and Calder (1970). Although I do not agree with Calders
strictures on the play as a whole, his analysis of the Greek models is excellent. On the lack of unity
cf. Zwierlein (1966) 91.
191
discussion of the plays structure, and to concentrate instead on the role

that repetition plays in its thematic texture.
Even before Pyrrhus appears on the stage, the Greek messenger Talthybius has introduced into the play, with a sinister Ringkomposition, the tragic
spectre of a previous mora, one infamously brought to an end by the sacrifice of Iphigenia. The Greeks, stranded at Troy by bad weather, are poised
to repeat a well-known pattern of delay (1645):
o longa Danais semper in portu mora,
seu petere bellum, petere seu patriam volunt.
O that there is always this long delay in harbour for the Greeks, whether they will
set out for war, or set out for their homeland.
Pyrrhus urges the leader of the Greek army to honour his dead father,
Achilles, by sacrificing Polyxena, who had already been promised to him.
The sacrifice would represent a decisive victory of the past over the present in
more ways than one. It is, literally, a request coming from the underworld.42
Pyrrhus does not seem to be aware of that, but Achilles ghost had appeared
to Talthybius, as we hear in the first scene of act 2, and had demanded that
the Greeks respect their promise: otherwise he will prevent their departure, and their return home (Achilles could count on his mothers help).
Moreover, it is a request made in the name of past practices whose value
Pyrrhus readily accepts despite Agamemnons tormented doubts: it is, in
many senses, a re-enactement of the dispute that dominates the Iliad from
its very beginning. The last words of Pyrrhus impassioned speech foreground the traditional nature of his request, and the tragic precedent that
Agamemnon should keep in mind (2489):
at tuam gnatam parens
Helenae immolasti: solita iam et facta expeto.
But as father you slaughtered your own daughter for Helen: I ask for things now
customary and with precedent.
Agamemnons rebuttal shows that he has undergone a significant transformation during the war, and especially at its end. To him, a simple repetition
of past patterns of behaviour does not seem to make any sense. His position is based on a series of compassionate, rationalistic assumptions which
question the appeal to tradition that is powerfully voiced by Pyrrhus. We
find Agamemnon, here, impersonating the voice of resistance to the evils
42
The underworld as home of the dead who successfully impose their rule on the upper world is a
symbol for the past. See the crucial evocation at the heart of Oedipus, above, pp. 8ff.
192
of tragedy to its nefas precisely like Tantalus shadow or the satelles in

Thyestes. In fact Agamemnon goes as far as stating that he would have liked
to prevent the destruction of Troy (279), and, in any event, that enough,
more than enough punishment has already been dealt out: (exactum satis |
poenarum et ultra est, 2867). Unlike most of his counterparts in Thyestes
or other plays (the satelles, Phaedras nurse . . .), however, Agamemnon is a
prominent character with a well-defined past from which he now tries to
free himself. Unlike Tantalus ghost, who is loath to engage in new crimes
(Thy. 867), but whose psychology is only defined in terms of rather simple
oppositions, Agamemnon articulates in detail his psychological evolution.43
Irresistibly, though, Agamemnon is ensnared in the discourse of the past
that is advocated by his opponent. The insults that Pyrrhus and Agamemnon start exchanging at line 336 are focused on their past: both look to it
for compelling explanations of what they are doing or what they should
be doing.
Agamemnon, of course, must yield, and he does so abruptly by promising
to heed Calchas orders (3512: potius interpres deum | Calchas vocetur: fata si
poscent, dabo, rather let Calchas, the spokesman of the gods, be called: if the
Fates demand it I will grant the sacrifice). The seer establishes at the very
beginning of his response the connection between the present predicament
of the Greeks and their bloody past (36070):
dant fata Danais quo solent pretio viam:
mactanda virgo est Thessali busto ducis;
sed quo iugari Thessalae cultu solent
Ionidesve vel Mycenaeae nurus,
Pyrrhus parenti coniugem tradat suo:
sic rite dabitur. non tamen nostras tenet
haec una puppes causa: nobilior tuo,
Polyxene, cruore debetur cruor.
quem fata quaerunt, turre de summa cadat
Priami nepos Hectoreus et letum oppetat.
tum mille velis impleat classis freta.
The fates grant a way to the Greeks at their customary price: the virgin must be
sacrificed on the tomb of the Thessalian leader; but in the costume that is worn
for marriage by brides of Thessaly or Ionia or Mycenae. Let Pyrrhus present her as
wife to his father: thus she will be properly given in marriage. But this is not the
only cause that detains our ships: a blood more noble than your blood is owed,
Polyxena. Let him whom the fates demand, the child of Hector, grandson of Priam,
43
On the character of Agamemnon see Anliker (1960) 65 and Schetter (1965) 401, with further
bibliography (and diagnoses ranging from weakness to noble humanity).
193
fall from the highest tower and so meet his death. Then let the fleet cover the sea
with a thousand sails.
These lines exploit verbal repetition as the iconic correlate of the repetition
that Calchas advocates: solent occurs twice (360, 362),44 and so does fata
(360, 368). The polyptoton cruore . . . cruor (367) further reinforces the effect. Repetition, moreover, stresses the ritual nature of Calchas order. In
fact, his order that the sacrifice of Iphigenia be re-enacted specifies also that
it proceed with all due respect for religious ritual: Polyxenas death, masked
as a wedding, will thus be a perfect repetition of Iphigenias murder and
will testify further to the disturbing connection between ritual and murder
which can be identified in other parts of the Senecan corpus.45
It is also worth stressing the dramatic effectiveness of Calchas concise
intervention, which is similar to the dialogue between Creon and Tiresias
in Sophocles Antigone,46 but which by its very brevity elides any space for
discussion: it is, in all senses, a final and irrevocable decree.47
Atreus obsession with a maius nefas draws attention, as I remarked
earlier,48 to the fact that any repetition of nefas is necessarily worse than its
precedent. This obsession finds a novel incarnation in Calchas unexpected
order that the Greeks sacrifice not just Polyxena, but Astyanax, too. This
particular combination of horrors is especially striking because nothing in
the plot so far has led the audience to expect Calchas request: indeed,
there seems to be no precedent in the tragic tradition for combining in this
manner the fates of the two Trojan youths. It is a development, however,
which stands out as a direct, emphatic rebuttal to Agamemnons statement
that more than enough punishment has been exacted already (2867):
Calchas shows that more can be asked for, and more can be obtained.49
It is important to notice that Calchas is acting here as the structural
counterpart of the Fury in Thyestes: his appeal to fata (360, 368) perversely
parallels her reliance on furor. Both forces are responsible for the continuation of nefas in the face of moderation and restraint, and their connection
disrupts any rigid opposition we might have counted on so far. The voice
of fata appears to be steeped in the cruel repetition of the past and is as
44
45
47
48
49
Although in a different perspective (that of tracing elements of rhetorical colour), Leo (187879)
14955 has very interesting remarks on Senecas use of soleo. See Tarrant (1976) 208, and above,
p. 27. On iconic repetition in general, Wills (1996) 67 and passim.
46 Calder (1970) 76.
On Atreus ordo sacri see above, pp. 85ff.
As Braden ((1985) 37) aptly remarks: Senecan dialogue is not an exchange of news or feeling, but a
test of [the contestants] self-possession.
See above, p. 95.
Schetter ((1965) 408) rightly emphasizes the central importance of Calchas words in the structure
of the play.
194
complicit with the forces of the underworld as the furor it would supposedly counteract: the voice of fata, we should remember, coincides with the
dark menaces uttered by Achilles ghost from Acheron.
The contrast between Calchas brief speech (his only presence on the
stage) and the following choral ode, which contains a remarkable number
of Epicurean reflections on the nature of death, highlights the deceiving
rationality of fata. The choruss rationalistic doubts about afterlife seem to
present both an apt continuation of Agamemnons speech, with which they
share a sense of scepticism and uncertainty, and a tentative reassessment of
the relationship between life and fate. Thus the chorus picks up both on
Talthybius appearance, which stressed the need for Polyxenas sacrifice, and
on Agamemnons own doubts about this solution. The choruss Epicurean
overtones stand in complete contrast to earlier choral lyrics, a structural
problem which would deserve separate, detailed consideration. Here it will
suffice to say that the contrast between Calchas and the chorus inevitably
portrays the seers words as belonging to the vatum terriloqua dicta criticized
by Lucretius immediately after his depiction of Iphigenias death:50 both in
De rerum natura and in Troades the sacrifice of Iphigenia is invested with
enormous paradigmatic importance.
The moral balance of the story is at this point painfully clear: Agamemnons resistance to the invariable pattern of repetition is brutally cut short by
Calchas and, quite literally, silenced forever at the end of act 2. Repetition
will dominate in its most literal, obsessive form, as the apparently endless
replica of past nefas. Agamemnons suggestion that a radical modification
of the past could represent a valid alternative to this pattern and to the
plot of the tragedy is rejected outright. In revolving around Achilles
wrath over Polyxena, Troades as a whole represents a re-enactment of the
archetypal menis-epic, the Iliad. In this, as in other Senecan plays, revenge
retribution for deeds and obligations long past becomes the most reliable guarantee of tragedy, since it imposes a cycle of actions which can
never stray from its archetypal model and can never alter significantly the
psychological profile of the characters or the actual course of events they
inevitably choose. Thus revenge encodes both the supreme power of the
past over present and future, and the very essence of this form of literary
production, in that it provides the masterplot which the author must follow
while also clearing space for his creative innovation.
Andromaches actions, no less than those of her Greek foes, are motivated
by the spell of the past. As she enters the play in act 3, she retells the
50
De rerum natura 1.1023.
195
appearance of Hectors shadow in exact parallel to Talthybius report about

Achilless ghost in act 2. Although Hectors message is helpful,51 it shows
nonetheless the cognitive subordination of the living to the dead: Achilles
and Hector know more, and they can enforce other peoples behaviour. In
this respect, the appearance of Hector stands as an immediate denial of the
choruss latest reflections on the non-existence of the afterlife. But Hectors
appearance is intrinsically fraught with ambiguity: while it is ostensibly
geared to prevent a repetition, in the case of Astyanax, of his own tragic fate,
it visibly embodies repetition as it alludes to Hectors archetypal appearance
in the second book of the Aeneid (27097). Thus the stage is set for the
inevitable, and pointed, contrast between the results brought about by the
two apparitions: while Virgils Hector will succeed in saving Aeneas and
will thus ensure the Trojans a future that is significantly different from their
past, he will be denied any such success in the Troades, where, it seems, the
only permitted form of repetition is ad litteram.
Andromache repeatedly voices her hope of a Virgilian future for her
son. As a new Hector, futurus Hector in Ulixes words (551), he will be able
to build a new Troy and take revenge against the Greeks (46974):
o nate sero Phrygibus, o matri cito,
eritne tempus illud ac felix dies
quo Troici defensor et vindex soli
recidiva ponas Pergama et sparsos fuga
cives reducas, nomen et patriae suum
Phrygibusque reddas?52
Dear son, born too late for the Trojans, too soon for your mother, will that time
come and that blessed day when you, as defender and avenger of the Trojan land,
will found a Troy renewed, and bring back your people scattered in exile, giving
back its name to your country and your Phrygians?
Recidiva Pergama directly alludes to an idea which, with different overtones,

is central to the Trojans negotiation of their past destruction and future
hopes in the Aeneid.53 For Ulixes, of course, the possibility that Astyanax
will in fact avenge his country represents a compelling reason to kill him
now: thus repetition inevitably displays, once more, its dark, negative
potential.
The solution that Andromache devises in order to save her son grimly
foreshadows the eventual outcome of her efforts: she hides Astyanax in
51
52
53
A contrast especially emphasized by Schetter (1965) 409.

The theme is further elaborated in 65961, 77185.
The iunctura occurs at Aen. 4.344, 7.322, 10.58.
196
Hectors tomb, and her envoy expresses the distressing ambiguity of the
solution (51921):54
dehisce tellus, tuque, coniunx, ultimo
specu revulsam scinde tellurem et Stygis
sinu profundo conde depositum meum.
Earth, open up, and you, dear husband, rend the earth away from its farthest
cavern and bury my dear treasure in the deep gulf of Styx.
Hope of future salvation is sought in the ultimate embodiment of the

past a grave which literally swallows Andromaches hope for the future:55
in a similarly poignant scene at the end of the tragedy, Polyxenas blood
is sucked up by the thirsty soil covering Achilles grave; hence Achilles
establishes with an undisputed sense of finality his right over the slain
maiden (11624):
non stetit fusus cruor
humove summa fluxit: obduxit statim
saevusque totum sanguinem tumulus bibit.
The spilled blood did not stay or float on the surface of the ground: but the mound
instantly swallowed and savagely drained dry all the blood.
Thyestes, too, realizes at the end of his ordeal that he has become a monstrous coffin for his children: genitor en natos premo | premorque natis
(I, the father, overwhelm my sons, and by my sons am overwhelmed,
Thy. 10501).56 Andromaches final words find a disturbing, almost verbatim parallel in two other Senecan passages. Oedipus employs the very same
words when he is finally made aware of the fact that he is the sinner (Oed.
8648):
o e. si ferus videor tibi
et impotens, parata vindicta in manu est:
dic vera: quisnam? quove generatus patre?
qua matre genitus? ph. coniuge est genitus tua.
o e. dehisce, tellus . . .
o e. If I seem harsh to you, and out of control, revenge is in your hands; speak the
truth: who is he? of what father begotten? of what mother born?
ph . Born of your wife.
o e. Open up, earth!
54
55
56
The image realizes what in Thyestes is a perverse metaphor: Thyestes is the tomb of his children,
and they can be buried, paradoxically, only if he is cremated and buried himself (10902).
Almost certainly a Senecan innovation; see Schetter (1965) 418, n. 8. On the association of tombs
with kleos, and the literary implications of this connection, see Goldhill (1991) 1202.
Lines 10501 fulfil Atreus vow at 8901: pergam et implebo patrem | funere suorum. Accius had been
even more explicit: natis sepulchro ipse est parens (226 Ribbeck2 = 57 Dangel).
197
The revelation of incest is met by the desire to hide in the depths of the
earth, to return to it. As we will see presently, Andromaches decision to
have Astyanax hide in Hectors tomb is fraught with upsetting resonances
connected with incest and adultery: the Oedipal connection certainly intensifies them.57 And Phaedra, too, at the end of her tragedy, wants to
disappear into the depths of the earth (Phaed. 12389):
dehisce tellus, recipe me dirum chaos,
recipe . . .
Open up, earth, receive me, dread Chaos, take me back . . .
Even more striking, however, is the hallucinated internal debate about

the fate of Hectors tomb which torments Andromache at Troades 64262.
As she probes her feelings, Ulixes informs her that since Astyanax is reportedly dead,58 the only way in which the Greeks can perform the requested
ritual purification that is essential for their departure is by tearing down
Hectors tomb. In a harrowing aside Andromache weighs the alternative
whether she ought to preserve her husbands tomb or save her sons life.
The alternative, of course, simply does not exist: if the Greeks tore down
the tomb, both Hectors remains and Astyanax would be destroyed. Yet
Andromache desperately clings to the distinction, as she tries to preserve the contrast between past and future which has been an essential component of her thinking all along. As she inclines towards saving
the tomb, she desperately wants to spare Hector a repetition of his fate:
better to see Astyanax thrown from a tower than Hector killed once again
(6535):
potero, perpetiar, feram,
dum non meus post fata victoris manu
iactetur Hector.
I will be able, I will endure and bear it, so long as my dear Hector is not abused
after his death by the victors violence.
Later she realizes that saving her son means saving a chance of a different
future: serva e duobus, anime, quem Danai timent (my heart, of those
two, save the one the Greeks fear, 662). But there is no real possibility
of choice, and Andromache finally must admit to the harsh truth which
57
58
On the connection between womb and tomb in Seneca see Robin (1993) 11011 and, in general,
duBois (1988) 54. Racines Andromaque explores more fully the intricate set of erotic and sexual
implications that are tentatively suggested in Seneca: the plot itself, of course, hinges on romance.
A particularly poignant confession is registered at line 279, where Andromaque declares, referring
to Astyanax, that il maurait tenu lieu dun père et dun epoux alluding to Il. 6.42930.
So Andromache had claimed at 5947.
198
for the audience has ironically been clear all along. Ulixes trick has in fact
obliterated any distinction between past and future for her: utrimque est
Hector (659), there is no choice between preserving the memory of the past
and rescuing the seed of future revenge. Andromaches ineffectual debate
poignantly underlines a truth that the play has already upheld not once, but
several times, namely that the only movement allowed by fata is a repetition
bound to keep as close as possible to its model. If anything, Andromaches
inconsequential emotional struggle shows that she herself is ultimately more
inclined to preserve the past than to give the future a chance. After all, by
asking Astyanax to come out of his shelter, and by leaving him at the mercy
of Ulixes (Andromache and the senex had deliberated at length on the
danger of the situation, and Hectors ghost had been extremely clear in this
respect) she chooses the only option that could guarantee the preservation
of her husbands tomb.
A reading of these conflicting allegiances can naturally enough be expanded into an analysis of the emotional intricacies of Andromaches character, especially in the light of what might appropriately be dubbed a significative lapsus at line 501. Before committing a reluctant Astyanax to his
frightful refuge, Andromache invokes Hectors protection: Hector, keep
safe the stolen treasure of your loving wife, and with trusty ashes welcome
him so that he may live (Hector, tuere: coniugis furtum piae | serva et fideli
cinere victurum excipe, 5012). Coniugis furtum is a surprising definition
which the oxymoronic addition of piae (and fidelis in the following line)
does nothing to tame: it still refers, in no uncertain terms, to adultery.59
Thus, even if we set aside the tense Oedipal implications to which Andromache effectively draws our attention with these unguarded words, it is
plausible to infer that she is thinking here that her ploy to save Astyanax
victurus constitutes a betrayal of sorts of her dead husband. By committing her son to her husbands grave, Andromache vicariously and perversely
fulfils a wish which is well known to lovers, classical and otherwise: that
they might be joined together in death.60 The spell of the past, it seems,
holds hostage even the one character in the play who seems to be generally
sincere, if not without a degree of ambiguity, in her desire to guarantee
of
her son and her country a different, better future. The tomb, the sema
epic honour and the symbol of epic grandeur (here charged with a gesture
towards elegiac wish fulfilment), violently encodes in the drama of Andromache the demands both tragic and appealing of the past. Determined
to save Hectors tomb, Andromache echoes her Virgilian counterpart. In
59
Cf. Fantham (1982) 288.
60
Ov. Met. 4.157, with Bomer ad loc.
199
the third book of the Aeneid she is recognized by Aeneas while she offers a
libation to Hectors empty grave, an empty tomb (tumulus inanis, 3.304)
which seals her unwavering determination to live in the past, in a miniature Troy built around a false Simoenta (3.302: falsi Simoentis ad undam).61
Against Andromache stands the unwavering request of immutability and
repetition, voiced in different guises by Pyrrhus, Calchas and Ulixes, even
as the latter proclaims that killing Astyanax is necessary in order to avoid
a new Trojan war:62 as the protagonists who catalyze the string of events
which constitute the tragedy, they also embody the narrative progress of
the play and guarantee its successful arrival at an end, of a sort. But they
advocate an apparent progression which ultimately results in the denial of
meaningful change, and strives in fact to assure that no substantial change
will ever occur.
Even Astyanaxs physical appearance is moulded by the spell of the past:
he greatly resembles his father; indeed, too much to give him any hope
of a happy future, or to keep at bay the incestuous undertones of Andromaches feelings (Sen. Tro. 6468).63 This similarity, which Euripides had
briefly remarked upon (Eur. Tro. 117881), is especially magnified in Troades
(4618):
o nate, magni certa progenies patris,
spes una Phrygibus, unica afflictae domus,
veterisque suboles sanguinis nimium inclita
nimiumque patri similis. hos vultus meus
habebat Hector, talis incessu fuit
habituque talis, sic tulit fortes manus,
sic celsus umeris, fronte sic torva minax
cervice fusam dissipans iacta comam.
Dear child, true offspring of a mighty father, sole hope of Troy and of your shattered
family, scion of an old race, too glorious and too like your father; these were my
dear Hectors features; he was like this in his walk and in his bearing; he held his
gallant hands just so; just so he carried his shoulders high and seemed to threaten
with frowning brow, shaking his streaming hair with the toss of his neck.
Only later do we discover why Astyanax is nimium . . . similis to his father.

Seneca departs from the tradition according to which the boy is buried on
61
62
63
See also Andromaches reaction to the appearance of Aeneas, which seems to be echoed in 3.31012:
verane te facies, verus mihi nuntius adfers, | nate dea? vivisne? aut, si lux alma recessit, | Hector ubi est?
Owen ((1970) 130) argues that to the present victor belong the future and the power, which is
certainly true at the immediate level of action witnessed in the play. But it is important to remark
that the future that Ulixes has in sight is essentially a repetition of the past.
Compare Phaedras comments about Hippolytus close resemblance to his father as a youth at Phaed .
6468.
200
Hectors shield,64 and he has the messenger describe in graphic detail the
complete destruction of his body after the fall from the tower (111117):65
ossa disiecta et gravi
elisa casu; signa clari corporis,
et ora et illas nobiles patris notas,
confundit imam pondus ad terram datum;
soluta cervix silicis impulsu, caput
ruptum cerebro penitus expresso iacet
deforme corpus.
His bones are fragmented and crushed by the violent fall, his weight cast down to
the earth below blurs the features of his noble body, his face and those lineaments
of his glorious father; his neck is broken by the impact of the flint, his head split
open and the brain squeezed out from inside the body lies there a shapeless
horror.
It is at this point that Andromache interrupts the messenger and sees that
her hopes are to be cruelly realized: Astyanax will follow in his fathers
footsteps only in the most gruesome of senses in this, too, he is like his
father (sic quoque est similis patri, 1117).66
One more scene embodies the power of repetition in a perversely effective way. In the fourth act of the play Helen reflects alone on her involvement with marriages that are destined to be unhappy and sorrowful (861:
funestus, inlaetabilis). She is referring, of course, not only to the devastating
long-term consequences of her wedding, but more specifically to the nefas
which opens the Trojan expedition, the sacrifice of Iphigenia. Her aside
(86171a) declares her completely self-conscious nature as a character who
is aware both of the moral implications of her actions and of the literary
background to which she is inevitably connected.
Dismemberment is, of course, a disturbingly common feature of Senecan
death, as, for that matter, of other first-century authors.67 As the basic
model of natural order, the body, dissected and scattered, is warped into
the supreme emblem of disruption, and symbolizes a breaking down of
spatial order which is germane to the dissolution of temporal linearity that
I have been discussing so far. Sometimes a disjointed body, as in the case of
64
65
66
67
Eur. Tro. 1133ff.; see Enn. scen. 73 Ribbeck2 = 1067 Jocelyn.

Homer does not describe Hectors dismemberment, but places enough emphasis on the damage
that the corpse suffers because of Achilles cruelty: see Il. 22.396404, and especially 4013, which
climax, as Seneca does, at the head. See also Virg. Aen. 1.483: ter circum Iliacos raptaverat Hectora
muros.
A poignant reversal of Catullus wish in 61.214: sit suo similis patri (cf. Procnes words as she sees
Itys, Ov. Met. 6.6212: a! quam | es similis patri); Leigh (1995) 471.
See especially Most (1992).
201
Hippolytus, who is disfigured beyond recognition, aptly portrays the cracks

in a reading of reality as a coherent, organic whole. The conflict between
points of view and competing narratives, together with the weak causal links
between acts and scenes, testifies to a pervasive epistemic quandary. How
can we comprehend and represent a reality which has long ago lost firm
internal points of view? How can we reconstruct a narrative which overcomes the puzzles and limitations of human understanding? In Theseus
anguished question at the sight of his dismembered son Hippolytus hic
est? (Sen. Phaed. 1249) we can infer a much larger question on the nature
of representation and the understanding behind it. Reality, diffracted in a
range of conflicting perspectives, escapes a unified approach, as if the very
abundance of details (of moments) denies the possibility of a complete,
immediate vision.
Thus the threat that circularity and repression pose to the norms of continuity and linear progress is also played out in the spatial dimension of the
tragedy. In the metonymic, often less than organic, segues that regulate the
development of the plays actions lies the intimation that the fundamental
categories of perception must face a new reality, where the natural order has
been ripped up and replaced by an obsessive regression and return, where
the present is menaced by the spell of the past (as represented by a literary
heritage). So much so, in fact, that only individual moments survive, arranging themselves in erratic, shifting structures. Spatially, the fundamental
inversion of above and below fatally connects upper world and underworld, and inverts their traditional hierarchy. This is not, however, the only
instance of spatial disruption. Troades, for instance, by alternating scenes in
the Trojan and Greek camps in direct succession, and using two different
choruses (one of Trojan women, the other of Greek sailors),68 represents a
reality torn between two competing, opposed points of view, and resists a
uniform approach. Troades, it seems, can understand and express emotions
and events only by partial, metonymic approximation. Gone is the illusion
that a linear Aristotelian plot can ensure a united, coherent vision of events,
and can signify a logical chain of causal connections. Events succeed each
other in a less than orderly fashion, forcing the sometimes puzzled audience to reconstruct the relationship that glues them together. Considerable
effort is required simply to understand the underlying structure of the plot.
68
In Hellenistic tragedy (Sifakis (1967) 11326) and Senecan tragedy the chorus is no longer present
uninterruptedly from beginning to end and might well have followed the Hellenistic practice of
leaving the stage after each ode, thus making it possible for the scene to change and for time to elapse
(Calder (1975), (197677) 6; Tarrant (1978) 2218; Davis (1993) 1138). On secondary choruses in
classical Greek tragedy see Wilamowitz (1909) 116, n. 13; Lemmers (1931) esp. 13142; and Carrière
(1977); on Eur. Hipp. 5871, see Barrett (1964) 1678 and Taplin (1977) 2308).
202
Astyanaxs death is thus highly symbolic, as are the deaths of so many

other children in Senecan tragedy. In Hercules furens, Troades, Medea, Phaedra and Thyestes, children are sacrificed to a regressive logic of revenge,
punishment and furor. Their deaths testify to the overwhelming power
of the past over the future. In this appalling elimination of potentiality,
Senecan tragedy also overturns one of the main principles of epic narrative. Epic strives to construct a bridge between the past and a future that
should normally be different from the past. In the Aeneid, for instance, it
is Anchises who dies, leaving Aeneas (and Ascanius with him) free to construct his own new identity and that of his fellow citizens, once Trojans,
soon to be Romans. It is remarkable that while Senecan tragedy adopts, as
we have seen, distinct traces of epic discourse, it deprives it of any forwardlooking implication and constantly imposes its conflictual, gnomic, static
perspective.69
iv
Endowed with prophetic powers, Cassandra foresees a distortion of the
linear continuity of time at the conclusion of her prophecy in Agamemnon;
the king will die, Troy will be avenged (7528):
haec hodie ratis
Phlegethontis atri regias animas vehet,
victamque victricemque. vos, umbrae, precor,
iurata superis unda, te pariter precor:
reserate paulum terga nigrantis poli,
levis ut Mycenas turba prospiciat Phrygum.
spectate, miseri: fata se vertunt retro.
Today this boat of dark Phlegethon will carry the royal souls, vanquished and
vanquisher. I pray you, o shades, and equally I pray you, waters upon which the
gods take their oath: for a little open up the cover of the dark world, that the
insubstantial crowd of Phrygians may look at Mycenae. Look, poor souls: the fates
turn backward on themselves.
Here, the opening of the gates of Acheron portends Cassandras desire that
her fellow Trojans (although it is difficult to overlook the larger impact of
spectate) see the breaking down of historical progression which had determined their demise: the decrees of fate seem to be turning backwards, and
now it is time for the Greeks to suffer. Paradoxically, the prophetess is able
69
It is useful to compare this opposition with the different connotations of the main characters of
Lucans Bellum Civile. While Caesar embodies epic discourse (winning, progressive, geared towards the future), Pompey (as Cato) gives voice to a tragic instance which hopelessly reverts to an
irretrievable, vanquished past.
203
to conjure up an image of the future course of events precisely because her

eyes are turned backwards,70 in more senses than one (71215):
stetere vittae, mollis horrescit coma,
anhela corda murmure incluso fremunt,
incerta nutant lumina et versi retro
torquentur oculi, rursus immoti rigent.
Her fillets hang still, her soft hair rises in horror, her gasping heart rumbles with
pent-up murmuring, her glance roams uncertainly, and her eyes, turned backwards,
circle, then again stare unmoving.
Cassandra looks back to the past history of Troy, but by denying her eyes
their normal forward-looking perspective, she also privileges a point of view
which is the antithesis of the norm. She looks back,71 and she looks down,
just as she hopes that dead Trojans will be allowed to look up from the
underworld, another form of vision which is exceptional and unnatural.
By looking back and down, Cassandra signals the arrival on stage of a
reversal of fortune which transforms winners into losers, even if it does
not manage to accomplish the opposite feat. Inspired by her underworldly
furor, Cassandra sees more and better than anybody else, but the price
to pay for her epistemological prowess is inscribed in her powers dark,
chthonic origin. The movement backwards, accordingly, has none of the
empowering overtones that connote the archetypal reversal of the Trojans
defeat elaborated in the Aeneid. Cassandras own death, pointedly linked
to Agamemnons in the iunctura victamque victricemque (754), denies the
possibility of escaping from the web of the past. Dying together, as the
inhabitants of the underworld express their impotent, purely negative joy
at Agamemnons demise, signals the regressive nature of their desire to
repeat the past, albeit as winners. As Cassandra herself had desperately
acknowledged, Troy is forever destroyed, and her prophetic abilities seem
utterly pointless, since they have not been heeded when they should have
been: Now Troy has fallen what have I, false prophetess, to do? (725:
iam Troia cecidit falsa quid vates agor?).
At the very beginning of Agamemnon another character had testified
to the regressive quality of backwards movements. As he dominates the
prologue, Thyestes reflects that his incest has subverted the law of nature
(346):
versa natura est retro:
avo parentem, pro nefas, patri virum,
gnatis nepotes miscui nocti diem.
70
71
See Tarrant (1976) 304 for parallel descriptions of frenzied ecstasy.

On torqueo as a sign of frenzied anger see Hershkowitz (1998) 923.
204
Nature has been turned backwards: I mixed father with grandfather monstrous!
husband with father, grandsons with sons and day with night.
Incest forces a repulsive mixing of different generations and perturbs their

natural motion forward. As unnatural as streams rushing back towards their
sources,72 incestuous offspring move in the wrong direction: they look at
the past, not at the future.
A similarly upsetting image of incest is established in the tragedy of
Oedipus. Creon scathingly attacks the king who has returned to his
mothers womb (Oed. 2368):
nec tibi longa manent sceleratae gaudia caedis:
tecum bella geres, natis quoque bella relinques,
turpis maternos iterum revolutus in ortus.
You will not enjoy the pleasure of your slaughter for long: you will wage war against
yourself, and will bequeath war to your children too, you who have foully returned
to your mothers womb.
Later in the play, Oedipus will mark his belated awareness of his nefas by
invoking a similar image of reversal (86870):
dehisce, tellus, tuque tenebrarum potens,
in Tartara ima, rector umbrarum, rape
retro reversas generis ac stirpis vices.
Open up, earth! And you, lord of darkness, and king of the shades, drag to the
deepest Tartarus this succession of ancestry and progeny which has turned back
on itself.
But the most extensive engagement with images of unnatural and ominous
reversal comes in one of the most powerful scenes of the play, the sacrifice
performed by Manto at 303402. Unable to see, Tiresias asks his daughter
to describe to him the manifesta . . . signa (302) marked in the entrails of a
slaughtered cow. The findings are portentous (36671):
mutatus ordo est, sede nil propria iacet,
sed acta retro cuncta: non animae capax
in parte dextra pulmo sanguineus iacet,
non laeva cordi regio, non molli ambitu
omenta pingues visceri obtendunt sinus:
natura versa est; nulla lex utero manet.
72
For retro in such contexts see Thy. 115 (iam Lerna retro cessit) and, somewhat differently, Thy. 459
(the unnaturalness of pushing back the sea by building in it). Similarly, of blood flowing backwards,
in the sacrifice at Oed. 349.
205
The position has changed; nothing lies anymore in its place, but all has been
reversed: on the right side lies the lung, filled with blood and unable to breathe;
the heart is not on the left; no caul with soft covering stretches with rich folds over
the entrails. Nature is subverted: no rule is left for the womb.
The retroflection of the internal organs of the cow is the iconic correlate of
the incest which Oedipus has committed,73 an equally devastating upheaval
of natural laws.74 No less stomach-churning is the presence of a foetus in
the womb of an unmarried cow (3735):
quod hoc nefas? conceptus innuptae75 bovis,
nec more solito positus alieno in loco,
implet parentem
What monstrosity is this? A foetus conceived by a virgin heifer, unusually placed
in a strange location, fills its mother.
The oxymoron fits Oedipus own situation, since the real monstrosity of his
marriage to Jocasta is, in effect, a return to the status quo ante of his prenatal
existence.76 In his nefarious regression, Oedipus cancels the passing of time,
denies the normal flow of events which should preclude his renewed union
to Jocasta, and questions the necessary correspondence between causes and
results: as an unmarried cow can conceive (a hysteron proteron of sorts),
so can he overstep societal and natural boundaries and return to Jocastas
womb. The tragic suffering never seems to deter Oedipus obsession with
returning. Even as he fully acknowledges his monstrosity, the punishment
he initially proposes for himself privileges a repetitive modality which would
actually entail the endless rehearsal of his crimes, an endless, explicit return
to Jocastas womb (9427):
illa quae leges ratas
Natura in uno vertit Oedipoda, novos
commenta partus, supplicis eadem meis
novetur. iterum vivere atque iterum mori
liceat, renasci semper ut totiens nova
supplicia pendas . . .
73
74
75
76
Bettini (1984) offers a persuasive analysis of this scene and its anthropological implications.
In the prologue to Thyestes the Fury points out that as soon as Tantalus ghost has polluted the house,
water starts to flow backwards, another appropriate indication of retroflection as sign of perversion:
cernis ut fontis liquor | introrsus actus linquat . . . ? (1078).
The unnatural association of conceptus and innuptae has perturbed the manuscript tradition as well,
part of which (A), prefers the facilior infaustae. See Bettini (1984) 14950 for a more extensive
discussion of this point.
Bettini (1984) 1512, who also compares Soph. OT 121315: Time condemns the
(1214) because, Bettini argues, lunione incestuosa si configura simultaneamente come generante
e come generata .
206
Let Nature, who in Oedipus alone reverses her fixed laws, devising strange births,
be changed anew for my punishment. Let me live again and again die, be born
again forever to pay new penalties as many times . . .
Later, the first scene of Senecas Phoenissae finds him, old and blind, moving
slowly with Antigone in the aftermath of the disaster that has wrecked his
life. He resists his daughters help, and voices his desire to turn back, to seek
once again the slopes of Mount Cithaeron, the place where he was found,
a thinly veiled disguise for an unavowable desire to return.
Oedipus embodies in Senecas tragedies the temptation and danger of
returning to an impossible past, a retrogression which, taken to its extremes,
threatens nature and history alike. No less than Thyestes and Oedipus
incests, Cassandras desire to even the score with her Greek foes is a nefas,
predicated on a violent reversal of history which is geared to annihilate
the past even more than simply to reverse fortune. Cassandra is ready to
die, provided that she can see the death of Agamemnon, and that the dead
Trojans can ascend briefly from the underworld in order to see what could
have happened, but did not, and can no longer happen.
The intertextual thread connecting Cassandra to Dido encourages comparison of the negotiation of Trojan past and future which is at the core of
Virgils Aeneid. Resolved to die, the queen of Carthage furens (4.465)
is assailed in her sleep by tormenting images (4.46573):
agit ipse furentem
in somnis ferus Aeneas, semperque relinqui
sola sibi, semper longam incomitata videtur
ire viam et Tyrios deserta quaerere terra,
Eumenidum veluti demens videt agmina Pentheus
et solem geminum et duplices se ostendere Thebas,
aut Agamemnonius scaenis agitatus Orestes,
armatam facibus matrem et serpentibus atris
cum fugit ultricesque sedent in limine Dirae.
As she slept Aeneas himself would drive her relentlessly in her madness, and she
was always alone and desolate, always going on a long road without companions,
looking for her Tyrians in an empty land. She would be like Pentheus in his frenzy
when he was seeing columns of Furies and a double sun and two cities of Thebes;
or like Orestes, son of Agamemnon, driven in flight across the stage by his own
mother armed with her torches and black snakes, while the avenging Furies sat at
the door.
It is worth noting the elaborate allusive game that Agamemnon plays with
its model, as the striking detail of the geminus sol (Ag. 728), inserted in a text
207
in which Cassandra describes Agamemnon (if not Orestes) agitatus scaenis,

thus fulfils the theatrical reference suggested by the Aeneid (Ag. 72831):
sed ecce gemino sole praefulget dies
geminumque duplices Argos attollit domus.
Idaea cerno nemora: fatalis sedet
inter potentes arbiter pastor deas.
Look, the day shines with a double sun, and double Argos lifts up twin palaces. I
can see the groves of Ida: the fateful shepherd sits in judgement of the powerful
goddesses.
At the very core of the Aeneid stands the complex, often obsessive elaboration of the relationship between past and present, and the Aeneid , too,
highlights the temptations and dangers inherent in the desire simply to
return to an unattainable status quo ante. But there are conspicuous differences between the way in which the poem negotiates these opposite trends
and Senecas own approach to the same critical theme. The battle between
past and future which dominates the first part of the Aeneid finds a resolution, albeit a painful and uncertain one, in the ultimate predominance
of a teleological solution of the plot which breaks with the repetitive, ineffectual compulsions displayed by the Trojans in the earlier phases of their
wanderings.77
In Agamemnon, but also in Oedipus and Thyestes, any such teleological
drive is conspicuously absent. Regression and return impose seriality as
the dominant organizing principle of the plays, and there seems to be no
counterbalancing force which might eventually displace them. There is, in
effect, no clinamen which may lead to a future that is significantly different
from the masterplot of the past. Looking back, and obsessively insisting
on the repetition of a past nefas, prevents the successful repression of nefas
which many characters in these tragedies advocate, often with less than
compelling force. While there is no guarantee that looking forward, to the
future, will bring no new crimes, it is certain that further engagement with
age-old ones will only perpetuate the spiral of revenge and counter-revenge.
Rather than looking at epic as a possible (if far from entirely successful) solution, Senecan tragedy makes it an integral part of the problem.
Epic models of representation import into the tragedy a disruption which
is ethical as much as it is narrative. The radically different status of epic
vis-à-vis dramatic representation once freed from its prescribed boundaries, such as the messengers speech contaminates the texture of tragedy
77
Quint (1993) 5096.
208
and precipitates the incumbent threat of nefas. Metastatizing as an alien entity in the play, epic narrative and temporal structures thematize the breaking down of narrative conventions and the ethical boundaries they imply.
The opposite process is well documented in the Aeneid, where tragedy
powerfully deconstructs the forward-looking, self-assertive conventions of
epic narration.
Seneca shows that the epics linearity is illusory, perhaps true only, if at all,
in strictly relative terms; in a different context, he shows that epic, too, is
the embodiment of nefas. In the prologue to Thyestes it was precisely an epic
intertext which vigorously introduced nefas; we see now that the association
extends to other important aspects of the play. But this complementary
demonstration is hardly neutral. To show that epic can be to tragedy what
tragedy was to epic proves that relative hierarchies and privileges are hardly
tenable, that there is no haven safe from the menace of nefas, and that
the illusory strength of the epic masterplot is precisely that illusory. In
retrospect, Thyestes and other Senecan tragedies question the discursive
assumptions on which the Aeneid was built, and shatter any optimistic
ideals that it may have nurtured.
v
An impossible dream of return also torments Medea, the arch-heroine of
Senecan drama. While it is undoubtedly difficult, and perhaps dangerous,
to privilege specific thematic links in the compact poetic texture of Medea,
it is more difficult, still, to forgo such a partial investigation in the name
of a hypothetical organic approach (which, incidentally, might be faulted
even more in the case of Senecas tragedies than in others). Thus it is legitimate, I believe, to privilege in a reading of Medea a particular obsession, a
recurrent thread which lends the protagonists actions and emotions their
common denominator: her desire to push her life backwards, to deny the
future any real possibility of unfolding and deviating from the past. Medea,
although we might want to see her portrayed as an unruly, furious and uncontrollable maenad, in fact consistently evaluates her predicament and
displays a dogged determination to achieve her goals. She is far from irrational: irrationality is a weapon she wields with poise and sophistication,
clear intent and strategy. Far more irrational, to her, is the supposedly
rational explanation of events which Jason half-heartedly tries to uphold.
Like Atreus, Medea could easily be dismissed as a dreadful embodiment of
boundless revenge: yet, even more than Atreus, she forcefully communicates
her clear-headed reasonings to the audience.
209
Medea strives to arrest the implacable sequence of events set in motion

by the announcement of Jasons wedding. When the tragedy opens she
already knows what is going to happen, and the epithalamion sung by the
chorus after the prologue only confirms the truth in a suitably contrasting
shallow tone. Deeply wounded in her pride, Medea longs for a return to
a past in which she was Jasons partner, a role from which she has now
rather hastily been displaced by Creusa. Images of return, accordingly, play
a central role in the tragedy.78 Creon, in act 2, orders her to return home,
but the return she has in mind is less literal and less circumscribed. She
insists that Jason should be given back to her, that their story (history) turn
back to the point before the tragedy started, before it all happened. She
might be guilty, but if she is, she still deserves what was once hers (Med.
2456):
si placet, damna ream;
sed redde crimen.
If you so decide, condemn the accused woman; but give me back my crime.
Towards the end of the confrontation her pleas become insistent (2723):
profugere cogis? redde fugienti ratem
vel redde comitem fugere cur solam iubes?
non sola veni.
You force me to flee? As I flee, give me back my ship, or rather give me back my
companion why order me to flee alone? I did not arrive alone.
A similar iteration returns in her dialogue with Jason: redde supplici felix
vicem (give me, a suppliant, my reward, 482), and again redde fugienti sua
(give your wifes property back as she flees, 489), are the high points of
her appeal.79 It is important to stress the fact that this particular aspect
of Medeas psychology is wholly Senecan. Euripides heroine makes no
attempt to win back the object of her passion. She is resigned to her fate,
determined only to take on Jason as agonizing a revenge as possible.
In her triumph, Senecas Medea finally achieves the nefarious return she
has so much longed for. Her actions describe a stripping away of features
that she considers external, to reveal only the inner core, the true Medea
(or Medea). She sheds first her status as a wife, returning to be only a
mother: materque tota coniuge expulsa redit (the wife in me is driven out,
the mother is completely reinstated, 928). But even that is too much, and
78
79
A comparably impossible dream is voiced (perhaps deceivingly) by Clytemnestra in Ag. 241: sed
nunc casta repetatur fides.
Compare a similar exchange between Electra and Clytemnestra in Ag. 9678: CL. redde nunc gnatum
mihi. | EL. Et tu parentem redde.
210
the path backwards can proceed further: rediere regna, rapta virginitas redit
(my kingdom has been restored, my raped virginity is restored, 984). She
is willing to eradicate any trace of motherhood from her very womb: in
matre si quod pignus etiamnunc latet, | scrutabor ense viscera et ferro extraham
(if any pledge even now lurks unseen within its mother, I shall probe my
womb with the sword and tear it out with the steel, 101213). Mater, which
a few lines earlier meant mother (928), is now (1012) a strictly physiological
marker, pure anatomy, simply a womb. At the conclusion of her revenge,
before flying away, driven in a blaze of serpents, she has succeeded in
bringing back a past which is paradoxically both before crime, and yet
full of it. Her virginity is back, but at the price of multiple murder, and
infanticide. She has held true to the vow she expressed in her magic rite:
indeed, she has bent the courses of the seasons (75964):
temporum flexi vices:
aestiva tellus horruit cantu meo,
coacta messem vidit hibernam Ceres;
violenta Phasis vertit in fontem vada
et Hister, in tot ora divisus, truces
compressit undas omnibus ripis piger;
I have bent the courses of the seasons: the summer earth has shivered at my chant,
Ceres has been compelled to watch a winter harvest. The Phasis has turned its
violent waters back to its source, and the Hister, which branches out into so many
mouths, has held back its sullen waters, reluctant to move in any of its channels.
Again, it is worth while to contrast this turn of events with its Euripidean
model. The Greek Medea displays a realistic preoccupation with her future,
and ensures for herself a safe refuge at Aegeus palace in an episode which
is completely omitted by Seneca. In sharp contrast to Euripides Medea,
who plans her departure towards Erechtheus land (1384), at the end of
this play Medea does not have a new city to move to, and simply disappears
into the sky, returning to the ancestral abodes of her family.
In her triumph over the laws of time and nature Oedipus own triumph Medea denies the constrictions that reality imposes on emotional
drives. She erases the notion of temporality with the same determination
with which the unconscious refuses to acknowledge times existence80 and
80
On Freuds central tenet that the unconscious ignores temporality see especially The Interpretation
of Dreams (Freud (1900) = SE iv.328 and v.5778; the latter passage, in which Freud discusses the
relationship between lack of temporal awareness and anger, is especially interesting in connection
with Medea); On Beginning the Treatment (Further Recommendations on the Technique of PsychoAnalysis vol. i) (Freud (1913) = SE xii.130); New Introductory Lectures on Psycho-Analysis (Freud
(1932) = SE xxii.74). In general, see De Ritis (1991) 233.
211
the unidirectionality of causal connections81 another crucial aspect of its

symmetrical, generalizing logic which emerges in contrast to the prevailing
adult logic that is based on Aristotelian principles.82 Medea is, at a certain
level of abstraction, a wish fulfilled, a compulsion satisfied against the
requirements of logic and reality. We are reminded of Atreus similar victory
over the basic principle that what is done cannot be undone; by punishing
Thyestes with abandonment, he is able to restore what has been lost, to
assure himself that his children are really his own and that his wife has
never been seduced by a wanton brother: liberos nasci mihi | nunc credo,
castis nunc fidem reddi toris (now I am convinced that my children are my
own; now I believe that I can trust again the purity of my marriage-bed, Thy.
10989). This is, quite literally, a dream come true, or, in psychoanalytic
terms, an instance of Ungeschehenmachen, the retroactive annulment of an
event,83 which is perfectly possible, of course, in the unconstrained realm
of symmetrical logic and negative magic.
Medea, like Calchas or Ulixes in Troades, embodies the power of the past
over the present and the future. She does not accept Jasons change of mind,
and she alters the regular succession of times, both natural and historical,
in order to achieve her goals. The moral balance, as is usual with Senecan
characters, is less easy to determine than one might be tempted to think.
Jason is a lame, ineffectual character, excused without much enthusiasm by
the chorus, on the basis of his intrinsic weakness. His desire for change for
a new bride is essentially rooted in selfishness, his desire to save himself
and, perhaps, his children. Functionally, his role is similar to Andromaches,
but nothing remains of her emotional appeal, of her complex, engaging
inner turmoil. Medeas reasons compared with Ulixes are also basically
private in nature. She cannot claim that the future of a whole nation is at
stake, that Astyanaxs survival will provoke endless agony for the mothers of
Greece. Yet, paradoxically, her actions acquire a grandiosity that completely
overshadows Ulixes Realpolitik. Her greatness lies precisely in her adamant
conviction, deeply personal and undebatable nonetheless, that she will
halt the course of events. The tone is set from the beginning, when she
defiantly asks the sun, her ancestor, whether it can still bear to proceed in its
81
82
83
On the parallelism between the elision of chronology and the unconscious undermining or abolition
of causal links see Fragments of an Analysis of a Case of Hysteria (Dora) (Freud (1905) = SE vii.17),
with The Interpretation of Dreams (Freud (1900) = SE iv.247). It is interesting to read in conjunction
with Freuds remark the important arguments put forth by Goldschmidt ((1969) 479, 16886) on
the prevalence of the present in the Stoics ideology of time.
Bodei (2000) 319.
Notes on a Case of Obsessional Neurosis (the Rat Man) (Freud (1909) = SE x .2356); negative
magic, Inhibitions, Symptoms and Anxiety (Freud (1926) = SE xx.11920 at 119).
212
path should it not turn back right now (Sen. Med. 2831)? Creon had
dimly foreseen the truth: giving Medea time, even a little time, means
giving her the very weapon she needs (292).84 She wants to control time
in order to bend it backwards. The certainty of her furor is rooted in the
certainty of natural events (4017):
dum terra caelum media libratum feret
nitidusque certas mundus evolvet vices
numerusque harenis derit et solem dies,
noctem sequentur atra, dum siccas polus
versabit Arctos, flumina in pontum cadent,
numquam meus cessabit in poenas furor
crescetque semper . . .
while the earth stays at the centre and keeps the heavens balanced, while the
bright universe maintains its constant revolutions, while the grains of sand are
innumerable, while day attends the sun and stars the night, while the pole keeps
the Bears dry as they revolve, while rivers flow down into the sea, never shall my
madness falter in its search for vengeance, and it will increase constantly.
But she also carries out her proposal to subvert and destroy everything:
sternam et evertam omnia (I shall ruin and destroy everything, 414). At the
end of the play her furor and her dolor do acquiesce: she has altered the
regularity of time with her rites, and she has found a limit to her revenge
(101820). The apparent adynaton is fulfilled precisely as she kills her second
son:
misereri iubes
bene est, peractum est. plura non habui, dolor,
quae tibi litarem.
Youre telling me to have pity. [She kills the second son] Good, it is finished. I had
nothing else, anguish, to sacrifice to you.
Her cry of joy, which turns on its head the choruss moralizing sententia
at the beginning of the play (175: tempori aptari decet, it is right to adapt
to circumstances), is revealing: meus dies est. tempore accepto utimur (The
day is mine; I am enjoying the time I have been granted, 1017).
By taking control of time, Medea ensures that there will be, in a sense, no
future: Jason has lost Creusa and his children, and Medea disappears into
the sky, back to her ancestors home. All Jason can do, as Medea poignantly
84
In Euripides, Creon remarks explicitly on the fact that the time he finally grants Medea is not
enough for her to commit any of the deeds he fears: see 3556 (these lines are deleted by Nauck and
Diggle, but retained by Murray and Page).
213
remarks, is to take back his children, the emblem of their past life together
that he had rejected: recipe iam gnatos, parens (1024).85
It should be observed that Medea pointedly refrains from any idealization
of the past, even as she longs for its return. Her intent is largely negative:
things should not change, Jason should not marry again. Her fixation on a
surpassed state of events finds a significant echo in the second chorus of the
tragedy, where the women of Corinth elaborate on the nefas of seafaring.
It is easy to file this ode in the bloated category of laudationes temporis acti,
and to dissect it in search of topoi: the dangers and intrinsic impiety of
navigation, the longing for a long-gone golden age with no ambitions and
no sorrows, the deprecation of the geographic and moral disorder which
marks a degenerate present. Yet the chorus itself resists such a simplification.
The contrast that it institutes between past and present is less facile and
less reassuring. In the midst of expected judgements, it is remarkable, for
instance, that the idealized life of the farmer who has reached old age in his
ancestral fields (patrioque senex factus in arvo, 332) should be characterized
as lazy (piger, 331), as if it lacked, together with the nefarious practice of
navigation, a vital elan which is not intrinsically immoral, and which we
would in fact expect to be praised in the context of Roman ideology.86
Medea draws attention more than once to the weight and implications
of her name. In her dialogue with the nurse, she is fully aware of the
potential embedded in it: Medea superest ( Medea is left, 166), Medea
as a recognizable entity, as a persona somehow distinguishable from the
person who carries it. Again: Medea fiam (Ill become Medea , 171). At
the end of the play, as her revenge is being carried out, she feels that she has
lived up to the expectations: Medea nunc sum (now I am Medea , 910).
It is time for the others, particularly for Jason, to acknowledge this fact:
coniugem agnoscis tuam? (do you recognize your wife?, 1021). Recognition
is a belated act of cognition which reveals something previously hidden,
or unfocused, which, in this sense, stands as the ultimate victory of the
past. To be able to recognize Medea as Medea, or Atreus as Atreus, is
predicated on the immutability of fundamental characteristics which define
them as what they are. These characters can deceive and disguise, but their
inner nature sooner or later shines through and is revealed in a flash of
horror. Both Medea and Atreus embody the superiority of the past over the
present and the future. They both guarantee that past patterns will prevail;
85
86
The order does not imply any action on either Medeas or Jasons part: there is no need to suppose
that Medea actually throws the childrens corpses from the roof, and to construe this event as
evidence of the fact that the play could not have been staged.
See Biondi (1984) 87141 and Nussbaum (1994) 464ff. for a fine analysis of this ode.
214
they rise from the certainty of a model which their antagonists need time to
learn. Once they do, once they recognize, they admit the fallibility of their
desire, or hope, for change. Thyestes, for instance, had indeed suspected
that Atreus could not possibly have changed, and that Tantalus exhortation
was therefore dangerous. He was right: his only mistake was not acting on
such a good hunch. Even at a later stage he tries to dispel his depressing
forebodings, to discard his past worries: the old Thyestes should make
room for a new, impossibly happy one (Thy. 937: veterem ex animo mitte
Thyesten, dismiss the old Thyestes from your thoughts). Similarly, Jason
entreats Medea to change, to accept something new and different in lieu
of what she can no longer have. On a more positive note, Andromache,
too, recognizes her sons character, which is strikingly similar to her dead
husbands: I know your nature: you are ashamed to show fear (agnosco
indolem: | pudet timere, Tro. 5045).
vi
As I remarked in the opening section of this chapter, regressive repetition
is complicit with the action of repressed forces and impulses of destruction and upheaval. At the most basic level, regression denies the movement
forward inscribed in the natural passing of time. This turning back implies, sometimes literally, a return to darkness, the darkness of unspeakable
crimes, of nefas and furor. As her eyes turn back, Cassandra sides with a
compelling form of knowledge which is chthonic and subversive, and she
reveals nothing less than the violent overturning of the fates decrees.
In their struggle upstream against the linear determinism of time,
Senecan characters stage a rebellion against the notions of law and order represented by times unerring flow. If time is inreparabile, it is a sort
of counterfactual, ultimately impossible reparation that they stubbornly
try to achieve. The desire to turn back the clock on history, personal and
otherwise, finds its most poignant expression in the emphasis placed on
the past, which slowly bulges out of proportion as it invades the present
and conditions the future. We should ask ourselves, at this juncture, what
may be the overall implications of this obsessive regression that seems to
characterize many of the tragedies on the various levels I have discussed.
We might perhaps take our cue from an incisive passage in Senecas
Letters to Lucilius 122 that raises an intriguing set of associations. In his attack
against those who live backwards retro vivunt (122.18) Seneca expounds
the immoral connotation of fighting times natural forward movement.
Inverting the order of day and night is the most powerful icon, and to a
215
certain extent the cause, of a denial of nature which immediately affects

morality. It is a sign of dubious distinction to alter the normal arrangements
of time, the temporum dispositio (122.18), even if the sinners who choose
to live in this manner reveal their obsession in a number of seemingly
innocuous activities: they crave spring flowers in the middle of winter
and will do anything in order to obtain them (122.8), just as they strive
to prevent, even to erase, natural ageing (122.7). In the latter case Seneca
makes explicit the connection between the artificiality of tampering with
time and the connotations of sexual immorality which configure the whole
process as nefas: do they not live against nature, those who strive to retain
the glow of adolescence at the wrong age? Can there be something more
cruel or miserable? Hell never be a man so that he can continue to lie under
a man. His gender should have spared him this iniquity, but now not even
age will87 (non vivunt contra naturam qui spectant ut pueritia splendeat
tempore alieno? quid fieri crudelius vel miserius potest? Numquam vir erit, ut
diu virum pati possit? et cum illum contumeliae sexus eripuisse debuerat, non
ne aetas quidem eripiet?, 122.8).
The connection that is raised in the letter between sexual deviation
and the subversion of natures laws of ordered time finds a remarkable
counterpart in a tragic passage which I have already had an opportunity
to discuss. In his speech at the beginning of Agamemnon, the shade of
Thyestes complains that he has mixed together things which the norms of
nature keep separate: children with their fathers, grandparents and their
descendants, day and night (Ag. 346).88 Merging day and night is revealed
as the emblem of a much more upsetting tampering with natural laws. (We
will see in a moment how this image can be connected with the repeated
instances of the sun turning its course that dot several tragedies.)
Less upsetting, but equally revealing, is the connection between morally
inappropriate behaviour and interference with the natural flow of time that
is inscribed in a well-known mythical episode, which is repeatedly mentioned in the tragedies and voiced with particular emphasis by an enraged
Juno in the prologue to Hercules furens. From the very moment of his conception when Jupiter, eager to prolong the night he was spending with
Alcmena, prevented the dawning of a new day the hero symbolizes the
disruption of natural order which will be especially evident in his trampling
of the thresholds of the underworld (246):89
87
89
88 See above, p. 203.

Authors translation.
Appropriately, the structure and names of the constellations are now a permanent cosmic reminder
of Jupiters unfaithful behaviour. See especially line 5: tellus colenda est: paelices caelum tenent, and
618.
216

in cuius ortus mundus impendit diem
tardusque Eoo Phoebus effulsit mari
retinere mersum iussus Oceano iubar.
[This son] for whose begetting the whole world lost a day, when Phoebus, with
orders to keep the sunlight immersed in Ocean, shone forth late from the Eastern
seas.
This very detail of the mythical narrative, which elsewhere in the tragedies
is explicitly referred to as a breaking down of natural laws,90 is mentioned
by Seneca at De brevitate vitae 16.5 in the context of his attack against the
poetarum furor: Jupiters crime is magnified and perpetuated in the poetic
descriptions of his lascivious deeds.
The immorality, even perversion, of holding back personal or public
history must arguably be read in conjunction with the notion of time that
structures Stoic thinking about the physical universe. Although repetition is
prominently inscribed in the Stoic concept of palingenesis, I have observed
a number of instances in which repetition acquires distinctly negative overtones and is portrayed as a dangerous obsession. The contradiction is only
apparent. The kind of repetition that, for instance, Medea advocates, unduly forestalls the natural and expected evolution of the cosmic cycle which
91 This form of repetition
will eventually culminate in a purifying ekpyrosis.
slows down the movement of the cycle and constantly threatens its potential for renewal. It is a sinful form of restitutio in integrum, which we should
contrast with the righteous attitude described by Seneca at De providentia
5.8: our fate has been determined since the moment of our birth, and wisdom resides in a complete acceptance of its decrees; to go with the flow of
the universe is indeed a relief: it is a great consolation that it is together
with the universe we are swept along (grande solacium est cum universo
rapi). Acceptance of nature, a basic tenet of Stoic thought, necessarily implies acceptance of time, and of the movement forward which will lead to
destruction and purification.
A rebellion against Stoic orthodoxy alone, however, can hardly account
for the far-reaching prominence of this theme in the tragedies, and it might
90
91
Ag. 81415: cui lege mundi Iuppiter rupta | roscidae noctis geminavit horas.
as final catastrophe,
It has been argued that Seneca presents a distinctively pessimistic view of ekpyrosis
which is at odds with the traditional Stoic notion of a natural, purifying event (Rosenmeyer (1989)
1519). It is rather clear that Senecan heroes try hard to precipitate what they would like to see not
just as a phase in the unstoppable process of creation and destruction, but as a final catastrophe,
which might come in the form of a flood (Q Nat 3.2930). Also, contrary to the surviving (Greek)
Stoic authorities, Seneca posits a direct correlation between their wickedness and the catastrophe,
which, again, is portrayed in terms of punishment. On this issue, see Barnes (1978); Lapidge (1978);
Mansfeld (1979); Long (1985).
217
be worth while to expand our points of reference to include other firstcentury authors. I will start, though, from Virgil. The teleological ambition of an epic such as the Aeneid is inextricably linked with a treatment
of narrative time that privileges linearity and control. The story has to
proceed forward, to approach its ultimate goal. Regressions and digressions, tempting as they are, must be restrained lest they obstruct the chosen
path forward. Virgil gives this principle emphatic expression not in the
Aeneid but in his Georgics, at a point when the insisted description of the
furor equarum92 threatens the ordered unfolding of his didactic project
(G. 3.2847):
sed fugit interea, fugit inreparabile tempus,
singula dum capti circumvectamur amore.
hoc satis armentis: superat pars altera curae,
lanigeros agitare greges hirtasque capellas.
But in the meanwhile time flies, flies irretrievably, while, seized by love, we linger
around each topic. Enough now about herds; the second part of our labour remains,
to lead the woolly flocks and shaggy goats.
There is something to be gained by insisting on the contrast between the

furor-led temptation to digress and the rationality of the move that Virgil
opposes to it. The flow of time is undeniably determined by Fate.93 There
is, literally, no going around that fact. Or is there? The narrators fiddling
with the linearity of time is an act complicit with furor, with the subversive
forces down below. Arguably, there is no act of poetic creation which could
be deemed completely innocent in this respect. By retelling events, poetry is
bound to repeat, and thus to disclaim the uniqueness and linearity of time.
Moreover, any act of poetic evocation inevitably disrupts the temporal
framework of the events that it narrates, whether they are historical or
not: there can be no perfect coincidence of histoire and recit. Yet, as I
remarked earlier, certain works display much more obviously than others
their willingness to challenge the ordered unfolding of time.
In the Bellum Civile, too, the comforting logic of chronology94 is abandoned in favour of fractured, competing narratives which stubbornly refuse
any call to order, not to mention unity. Lucans poem, as Ralph Johnson
elegantly puts it, has no unity unless it is the absence of unity, and, having no unity, it needs no heroes to enact unity.95 Time becomes one of
92
93
G. 3.266, with Schiesaro (1993a) 140. It is worth remembering in this connection the importance of
Junos furor as a principle of delay (explicitly acknowledged by the goddess herself at 7.315) which
structures in a very basic sense the Aeneid as a whole.
94 Johnson (1987) 110.
95 Johnson (1987) 110.
Cf. Sen. Ben. 4.7.2; Helv. 8.3.
218
the most charged signifiers in the poem, endowed with competing ideological values: delay is pro-Pompeian as much as proceeding forward is
pro-Caesarian. As he engages in the re-enactement of a nefas, and thus inevitably displays his Caesarian allegiances, Lucan multiplies morae which
can at least postpone Caesars inevitable victory.96 The momentary nature
of Lucans heroes (many are able to hold the stage convincingly for a while,
but none can give the poem in its entirety a sense of unity) can provide
useful insights into Senecan tragedy as well, especially if we consider it
alongside the cubist diffraction of time in Ovids Metamorphoses which I
recalled earlier.97
Taken together, the experiments with time undertaken by Ovid, Lucan
and Seneca testify to a shared sense of uneasiness and anxiety. Never before
(pace Lycophron, an exception that confirms the rule) had classical texts
built up such a comprehensive onslaught against the linearity of time. By
the beginning of the twentieth century similar challenges, however, had
indelibly marked the experience of Western culture. In this more recent
scenario it is possible to connect experimental attitudes towards time with
a flurry of scientific discoveries and philosophical reflections which substantially modified our perception of time and imposed uniformity and order
on that which was previously defined only by ignorance and superstition.98
But Roman culture, too, had experienced its own Copernican revolution.
Bringing order to the confusion that had reigned for several centuries,
Caesar introduced, just one year before his death, a new, reliable calendar.
It was, by all accounts, no mean feat, as Lucans Caesar remembers in his
meeting with the Egyptian priest Acoreus (10.1847):
fama quidem generi Pharias me duxit ad urbes,
sed tamen et vestri; media inter proelia semper
stellarum caelique plagis superisque vacavi,
nec meus Eudoxi vincetur fastibus annus.
For sure, I was brought to Pharos cities by report about my son-in-law, but still
report about you too; always in the midst of battles I found time for higher things,
for regions of the stars and sky, nor will my own year be worsted by Eudoxus
calendar.
It is undoubtedly striking that Caesar seems here almost to equate his

pursuit of Pompey to his insatiable drive to know. In his desire to reveal
the inner secrets of nature, Caesar momentarily turns into a cultural hero
of sorts, redolent in many respects of Lucretius portrayal of Epicurus,
96
97
One of Masters most important acquisitions, see Masters (1992) passim.

98 Kern (1983).
See above, p. 188.
219
who pushed himself to the edge of the world in order to comprehend

the regulatory mechanisms of all things, and even more of Alexander the
Great, the ruler who embodied a profound link between the thirst for
knowledge and the thirst to conquer. As the inventor of the new calendar,
Caesar marks the ordered passing of time as a personal accomplishment
intertwined with the vicissitudes of Roman history. There will be, as we
know, a well-ordered time for Augustus,99 but it was his predecessor who
had explicitly paved the way. Caesars ordered annus proceeds smoothly and
without uncertainties, not unlike his plan to catch Pompey and progress to
a new form of political organization.
It is not simply a matter of observing that political power is intrinsically
involved in the regulation of the calendar and especially of its festive days,
as Ovids Fasti amply attest. Rather, we should try to recapture at least part
of the fundamental sense of (new) order and predictability that lay at the
core of the Julian reform, and to imagine that such a powerful revolution
could come to be seen as the perfect target of equally forceful counterreactions. It is precisely because there can now be, literally, such a thing
as time for Augustus, that the active and unpredictable manipulation of
time in poetry can acquire a significant disruptive force. Reacting to the
myth of progress and renewal that is so central to Augustan rhetoric, the
disruption of time portrayed by Ovid, Lucan and Seneca problematizes in
different ways the most basic category of human understanding, implicitly
questioning its very foundations. Causality and chronology waver under the
repeated attacks of analogy, association, regression, repetition and delay.100
The unstoppable vector of history that was promoted by Augustus can now
be shown to be simply one of many possible movements of history and
hence deprived of any teleological impact. History can bend back on itself
and explore darkness and regression, can eliminate the future and proclaim
the triumph of the past. Not even time is safe from nefas: it can actually
be one of the ways in which nefas achieves its victory. Neither is poetry
safe: poetry, too, bends backwards, thanks to the multiple opportunities
afforded by a novel, creative poetics.
99
100
Wallace-Hadrill (1987).
I do not believe, however, that the foregrounding of irrational elements and the development of
avant-garde poetics especially in first-century a d literature should be read only as a reaction to
Augustanism and its undoubted impact on cultural protocols. The drive towards rationalization
is an important feature of Roman culture at least from the second century b c onwards (Moatti
(1997)), and any reaction to it should be read not simply as antagonistic to the political powers
who may at certain times appear to be fostering it (too reductionist a move), but primarily in the
context of an ongoing cultural debate (see Schiesaro (1997); and Schiesaro (1993b) 263).
220
In the reversal of upper and lower worlds, as in the constant return of

the past, Senecan drama gives voice to the dead who emerge from Acheron
to impose their dreadful rule on the living. Time turns back not so much
to redress the misfortunes of the past as to guarantee that no change and
no variation can occur. It is difficult not to read in this reversal a pained
scepticism about the regularity imposed by Augustus on time and space
a regularity and faith in the future that Augustan poets had tried to
make their own. In, for instance, Virgils promise of imperium sine fine101
(a promise that the Aeneid powerfully questions at several critical junctures),
or in Horaces Epode 16, we can read the hope that the new Augustan beginning has interrupted the law of cosmic cycles and, thanks to the kings
vital force, has inspired a progression which will not have to be tempered
by catastrophe.102 Two or three generations after these hopes had been tentatively expressed, they have lost even their qualified problematic appeal
and are ripe for deconstruction.
Seneca, as I mentioned, presents us with a dramatic visual icon for the
fundamental disruption of natural laws which recurs with remarkable frequency in tragedy. His tragedies often display the suns uncertainty in
following its course or even its extraordinary retrocession.103 Shocked at
the sight of the nefas perpetrated by Atreus, the sun withdraws, throwing
mankind into utter darkness as the text repeatedly emphasizes. It is indeed
tempting to recall that Caesars calendar, for the first time based entirely
and reliably on the suns regular movements, seems to have been one of the
immediate motivations behind a surge in the popularity of solar cults.104
The sun, elevated on the one hand to the position of supreme guarantor
of the regularity of time, displaying in its perturbed and unpredictable
movements the irrational criminality of human actions, can now become, in Senecas obsessed and distraught universe, the ultimate symbol of
disorder.
101
103
104
102 Eliade (1954) 13340, esp. 136.

Aen. 1.25796.
Rosenmeyer (1989) 160 and Schmitz (1993) 90.
Bickerman (1980) 51, with Nilsson (1932) 166 and Weinstock (1948) 37.
chapter 6
i n t e rt e x t ua l i t y a n d i ts d i s co n t e n ts
i
The thematic insistence on the role of the past in the tragedies is displayed
to great effect in the extensive and for some even hypertrophic allusive
dimension of Senecas writing.1 I now want to explore this dimension of
intertextuality further. By problematizing the relationship between intertextuality, poetry and the past, I also want to look again at the connection
between poetry and passions, and at what implications such connections
might suggest for the interpretation of Thyestes and other plays. In the
second section of this chapter I will therefore focus again on the Stoics
theoretical discussion of the role of the audience confronted by theatrical
outbursts of passion, and in the third section I will investigate how spectatorship is dealt with in the tragedies themselves. The last portion of this
chapter will be devoted to an issue which is crucial both within the tragedies
and in the possible modalities of their reception, that is, the relationship
between this form of drama and epic. I will also look further at possible
analogies between Senecas plays and modern epic drama, which may be
characterized by the particular mode of reception it demands and purports
to foster. Can Senecas theatre be considered epic in any Brechtian sense?
Was his notional audience predicated on a similar set of presuppositions?
Intertextuality is indeed one way of looking back, of allowing a past constituted by texts, words and narratives to shape the present and possibly the
future. All texts are by definition intertextual, even if and when they happen
1
In his important discussion of Senecas intertextuality Segal ((1986) 20214) focuses especially on the
implications of the sword at Phaed. 896 (hic dicet ensis, this sword will tell you). The sword, which
replaces the writing tablet with which Theseus accuses Hippolytus in Eur. Hipp. 877, is, according
to Segal, the visible mark of Senecas own anxiety of influence , the trace of the earlier writers
absence (208).
221
222
to be only marginally engaged in overt allusive strategies.2 Yet it is generally and rightly admitted that Seneca and, for that matter, several other
first-century authors revel in the expressive possibilities afforded by a sustained dialogue with their models. The tragedies not only insert themselves
explicitly in a history of literary production which revolves around wellknown and repeatedly staged myths, but their language, too, signals at every
turn their genetic connection with previous points of reference, tragic and
otherwise. Considered in this perspective, intertextuality acquires a crucial
role both as the marker of an ideological obsession with the past, and as a
self-conscious form of critical reflection on a texts literary affiliations.
In deciphering the connections between intertextuality and the ideology
of the past, it is interesting to start by comparing the relationship established
by the Aeneid with its Homeric models and the modalities that configure
Senecas connection with his tragic antecedents.
Senecan tragedy differs drastically from the Aeneid in its exploitation
of and reflection on literary models. The Aeneid rewrites Homer in more
senses than one. Its two halves follow the lead of the Iliad and the Odyssey,
but the inverted order in which they are recalled establishes an active gesture
of modification. The Aeneid follows in Homers steps, but this time the
Trojans will win, proving that the curse of repetiton can indeed be broken.
Similar considerations arguably extend to Latin epic as a genre. Steeped
as it is in a dialogue with its Greek models, Latin epic shows from its
very inception the ability to Romanize a past that is connected with, but
ideologically independent of, its Greek counterpart. Thus, from Ennius
Annales to Naevius Bellum Poenicum, Roman epic emerges as the preferred
medium for a negotiation of tradition and innovation in the literary (and
historical) realms.3
Tragedy does not enjoy the same kind of freedom. Of course the playwright is able to choose from different versions of the same myth, and, by
plotting it creatively, can easily shape his own distinctive voice. But tragedy
as a whole, rooted as it is in a mythical continuum which is relatively flexible but ultimately fixed, is impervious to the insertion of completely new
material, and to the same kind of active modification of the plot which is
encouraged by epic. Even a play such as Thyestes, which testifies to the degree of originality and freedom that the poet enjoys in the treatment of his
2
The divide between allusion and intertextuality is questionable, and, while it is certainly useful to
retain a terminological distinction, these two concepts must often be considered more as different
points on a graded continuum than as alternative options. On this point see now Fowler (1997a) =
(2000) 11537, and Hinds (1998).
On this topic see now Goldberg (1995).
223
chosen theme, cannot escape a largely predetermined series of events. The

contrast is especially glaring if we compare the plays, again, with the epic
Aeneid, which had enjoyed unprecedented freedom in the manipulation of
the mythistorical histoire that it stages. There is no mythic masterplot to
rule what should happen to Turnus, while Thyestes children still cannot
escape their fate. This lesser degree of freedom paradoxically entails a higher
degree of responsibility. If the Aeneid can attempt to impose a new narrative order which carefully balances nefas and ratio, if it tries to continue
and modify the Homeric plot by staging a reversal of fortunes which looks
forward to general reconciliation, as opposed to endless revenge, Senecas
tragedies are constitutively denied any such illusion. They will only be able
to reiterate the nefas, to return to it, incessantly and regressively.
Senecas tragedies stem from a continuous, even obsessive confrontation
with their models. As they give new life to the tragic experience first brought
onto the stage by the Greeks, thus questioning with an elaborate scheme
the apparently seamless experience of watching a tragedy, they inevitably
problematize the relationship with their Greek counterparts. The metadramatic activation of the plot and the consequent denial of any immediacy
to the tragic experience intensify features that Greek tragedy had only intermittently displayed (especially in such plays as Euripides Ion or Helen,
or, in different forms, in the case of Sophocles), and come to represent the
most individual hallmark of Senecas tragic writing.
The pervasive characteristic of Senecas tragedies is their belatedness: they
represent an anachronistic return to the past, a frustrated desire for lost
forms mediated by an overwhelming and oppressive intertextual memory.
Senecan tragedy validates its existence (and its novelty) by displaying total
awareness of its epigonic nature and by laying bare its internal mechanisms.
Oedipus embodies perhaps more vividly than any other play the torment
of Senecas post-Virgilian quandary. And the thematization of posteriority
is what makes the opening scene of Phoenissae, for instance, so harrowing. In general, all the characters in these tragedies are intensely aware of
their previous existence in the domain of literature, and they reflect, as
do the tragedies themselves, on the intermediate stages of writing which
divide these belated mythical narratives from their supposed originals. It is
precisely the metadramatic structure and the manipulation of time which
colour the existence of these plays, and our enjoyment of them as acts of
defiance aesthetic as much as ethic. The author-on-stage be it Medea,
Thyestes or Juno advertises in no uncertain terms the constructedness
and artificiality of the text as well as the fact that a fresh representation of
nefas could (should) have been avoided.
224
If all post-Ovidian literature is programmatically self-conscious to a very

high degree, Senecas own narcissism takes the form of a sustained critique
of authorial responsibility as it is showcased in the authors staged counterparts a group of obsessed, determined criminals. It is not just that poiesis
is (literally) staged as a constructive process: it is portrayed as a pollution
which inevitably involves author and audience alike. Whether they were
staged or not, Senecas tragedies imply a form of communication which is
not primarily or exclusively written; yet the tension they dramatize between
the visible (staged, spoken) and the invisible (implied, hidden, written) is
their defining characteristic.
ii
The analysis of intertextuality, especially this particular brand of Senecan
self-conscious, metadramatic intertextuality, can fruitfully move towards
an evaluation of rhetorical features as they interact with the psychological
processes of the reader. All texts, especially those involved in the narration
of mythical events, are necessarily repetitive: they repeat a mythic story,
they repeat each other. But some texts more than others display a high
degree of perceptible, thematized repetition, which they force the reader to
acknowledge. The intertextual gesture that impels the tragedy of Thyestes
from the shadows of non-existence makes explicit, as I showed in chapter
one, the ethical problems implicit in the creation and fruition of this play.
This intertextual impulse vehiculates the crucial metadramatic theme of
the moral responsibility of a certain type of poetry and, at a more general
level, makes an issue of the decorum of poetry by calling into question its
limits.
Intertextuality, at any rate, can never be a neutral operation in either
its contents or in the dynamics of its perception. It will offer the well-read
reader the pleasure of recognition, a chance to share with the author control
over the text and its signification; or it may puzzle and disempower those
who perceive it in an unfocused manner. It will be a way to activate memories of pleasurable events (the very act of remembering can be pleasurable),
or to recreate the painful experience of nefas; to encourage identification
with the emotions provoked by the past, or to look at them with relief,
anguish, hope or terror.4 It is arguably in this intrinsic tension that we can
locate the peculiar force of Senecas poetic project (and of those authors
4
It would be interesting to extend the interpretation of intertextuality that I sketch here through an
engagement with modern philosophical and psychological theories of memory and its ambivalent
nature as either a source of pleasure or as a catalyst for the onset of psychiatric conditions. See, for
instance, Hall (1996).
225
who are closest to him in this respect). Gone is the illusion that the world is
ordained in a logical sequence of discrete events, of clearly defined ethical
and aesthetical alternatives. Andromaches desperate monologue in Troades
offers a sequence of thoughts that can be extrapolated as a more general
epistemic protocol: the categories she carefully defines (living and dead,
husband and son, honour and safety) turn out to be so intertwined as to
be useless. There exists no real choice between giving up Hectors tomb
or Astyanaxs safety. Both have already been lost. The text provokes both
pleasure and pain; its poetics invite us to rejoice in agnition and to recoil in
horror as intertextual memories exhibit their violent, confusing potential.
Infandum, regina, iubes renovare dolorem, in Virgils words. Or then again:
forsan et haec olim meminisse iuvabit.5
From this perspective the vast mass of circumscribed intertextual points
of contact with previous poets, especially Virgil and Ovid, becomes in
Thyestes, and in Senecan tragedy at large, a source of horror and at the same
time a reiterated if imperfect apology for its legitimation. Intertextuality becomes an internal, ambiguous mode of defence for inconceivable
monstrosities: it partially displaces responsibility while almost compulsively
deepening the original wound, in a solution that is apparently preferable
to the repressive force of silence. This is possible only in all conceivable
senses after: after Virgil, after Ovid, after almost one hundred years of
engagement with the memory of the systemic disruption that was witnessed
in the first century.
As the poet Atreus has made abundantly clear, imitation of and competition with the past are also essential components of his poetics of
the sublime.6 The urge to transcend the limits of humanity seems to
emerge from the same terrifying depths of horror from which his unquenchable thirst for revenge has originated. Atreus offers a compelling
image of the mutually reinforcing connection between furor, poetic inspiration and an obsession with the past. Determined to take revenge for
the wrongs he has suffered, he is equally determined to surpass the poetic
models he explicitly chooses as points of reference. In the end, we are left
to contemplate with surprising satisfaction the pleasure that his revenge,
and his poetry, have been able to afford us.
It is precisely with these intimations in mind that I now return to the
scene from Oedipus with which I opened this book.7 There Laius evocation
of the ghosts of the underworld suggested a compelling vision of poetry
slowly and painfully emerging from its chthonic dens. Tiresias, the seer
5
6
The sorrow you bid me bring to life again is past all words (Aen. 2.3); The day will come, perhaps,
when it will give you pleasure to remember even this (Aen. 1.203).
7 Above, pp. 8ff.
See pp. 117ff.
226
whom Seneca forces to confess his impotence against the plague which ravages Thebes, had ordered this rite once his own attempts at understanding
had failed (3907):
nec alta caeli quae levi pinna secant
nec fibra vivis rapta pectoribus potest
ciere nomen; alia temptanda est via:
ipse evocandus noctis aeternae plagis,
emissus Erebo ut caedis auctorem indicet.
reseranda tellus, Ditis implacabile
numen precandum, populus infernae Stygis
huc extrahendus . . .
Neither the birds which on their light wings cut the depths of heaven, nor entrails
grabbed from still living breasts can summon up the name; another route must be
attempted: the king himself must be evoked from the regions of eternal night, so
that, released from Erebus, he may point out the author of the murder. The earth
must be opened up, the implacable power of Dis must be implored, the people of
infernal Styx must be drawn forth here . . .
It is from the dead, not from the living, that knowledge will eventually
arise, and for that purpose the earth has to be prised open and forced
to yield its secrets. (This is, incidentally, a scene that would have pleased
Ps.-Longinus, the author of On the Sublime, who had not hesitated to
include among his examples of sublimity a similar image that he found
in Homer.8 ) As he prescribes a procedure that in Senecas poetry acquires
strong metadramatic overtones, Tiresias comes close to a number of other
first-century characters whose competence in the domain of divination
and prophecy was closely intertwined with the nature and effect of poetic
creation.9 Tiresias own words are implicated in the language of poetics. Alia
temptanda est via (392) echoes the programmatic passage from the beginning
of the third book of the Georgics in which Virgil declares his dissatisfaction
with a number of overexploited poetic subjects, and vows to embark on
a project of startling novelty.10 The Virgilian allusion not only underlines
Senecas analogous intention to innovate (in fact the whole scene with
Laius is conspicuously absent from his Sophoclean model), but strongly
reinforces the supposition that the scene as a whole can be read, at one
level, as a reflection on poetics. What should not be overlooked, however,
is the sharp antiphrastic undertone of the allusion. Virgils aspiration to
fame is couched, in metaphoric terms, as the desire to lift himself up
8
9
10
Subl. 9.6, which quotes together Il. 21.388; 5.750; 20.615.

Cf. above, ch. 1, pp. 8ff.
The parallel is pointed out by Palmieri (1989).
227
from the ground and to be able to fly from mouth to mouth (G. 3.89):
temptanda via est, qua me quoque possim | tollere humo victorque virum
volitare per ora (I must attempt a path whereby I, too, may rise from earth
and fly victorious on the lips of men). Tiresias project revolves around the
diametrically opposed movement, an exploration of the depths of Acheron
which will be made possible by the ghosts ascent. The new way of poetry
outlined by Tiresias in his own programmatic statement will have to emerge
from a deep involvement with chthonic forces, with the shocking world
of Acheron. Poetry is to be the medium which enables underworld and
upper world to communicate, and in so doing transgresses the semiotic and
ethical boundary which should guarantee the separation of those worlds
and protect the living from pollution. Poetry blurs the thresholds between
dead and living and between past and present, thus preventing the former
from being lost for ever, and subjecting the latter to the constant anxiety
of unwelcome returns.
The choral ode that separates Tiresias intimation from its actual fulfilment celebrates in a peoples hymn (populare carmen) the achievements of
the god Bacchus.11 Its seemingly digressive intonation heightens the level
of dramatic suspense after the rites have been announced, and the audience knows that they are being performed elsewhere in the same time span
(4012: dum nos profundae claustra laxamus Stygis, | populare Bacchi laudibus
carmen sonet, while we loosen up the gates of Styx in its abyss, let the peoples hymn resound with Bacchus praise); the carmen is thus coextensive
with those arcane procedures, is part of the evocation through carmina
that had been mandated by Tiresias.12 In fact we will hear from Creon that
Tiresias pours wine Bacchus on the earth as he begins his rites (5667).
Bacchus is a uniquely appropriate character to be called into the picture at
this point. (It is, as we know, with Bacchic figures, Agave, Pentheus, the
Bacchae themselves, that Laius tragic catalogue ends.13 ) In Tiresias negotiation between the living and the dead, Bacchus appears as a powerful
intermediary who embodies both the joyful enthusiasm of inspired love
and the dreadful dangers of orgiastic rites. He is not only the divine patron
of Thebes, but also a powerful reminder of the tragic confusion of roles
and natural norms which Oedipus has brought on his city: he is, in fact,
the very symbol of nature acta retro.14 The ode focuses at an early stage
on the confusion of sexual identity that characterizes the god, and on his
power to metamorphose objects and human beings alike (41822; 4868).
11
12
13
See esp. Mastronarde (1970); Davis (1993) 2027; Tochterle (1994) 3627; Caviglia (1996).
See carmen magicum (561); canit (567); rata verba fudi (572).
14 See Paratore (1956) 125.
Lines 61618. See p. 9 above.
228
Bacchus continuously shifts between the dark violence of blood and the
resplendent light of his smiling appearance; he is the principle of life as
much as of deadly violence.
The numerous points of contact between the ode and the following
necromancy scene15 reinforce the notion that the invocation to Bacchus is
an essential component of those rites and shares to a certain extent their
metapoetic implications. Bacchus, the god of life and death, of light and
darkness, of masculinity and femininity, of blood and milk, is also the
perfect symbol of a poetry which constructs a bridge between the dead and
the living, the joyful frenzy of the Bacchic orgy and its obscure, threatening
undertones.
The form of poetry that emerges from Oedipus metadramatic reflection is centred on a pained yet inevitable relationship with the past, one
which passion, furor and Bacchic enthusiasm can access and elaborate for
the living. The underworld is not only the repository of all things dead,
but also of all the dark forces that project their grim shadow on powerless
mortals. It is in the context of this ideology of the past that intertextuality
should be set and allowed to acquire some of the eerie connotations that
must inevitably accompany such an extended interaction with the world
of the dead. It can hardly be meaningless that it falls to Oedipus to outline
a poetics of regression, but it is Thyestes that focuses extensively on the extraordinary power which accrues to the character who has fully understood
those principles and knows only too well how to exploit them. And Atreus,
the Dionysiac poet, is the ultimate embodiment of this form of poetics, at
once sublime and regressive.
pa s s i o n s a n d h e r m e n e u t i c s : t h e au d i e n c e
There is no such thing as a good influence, Mr. Gray. All influence is immoral
immoral from the scientific point of view. (O. Wilde, Portrait of Dorian Gray)
ac nescio quomodo ingenia in immani et invisa materia secundiore ore expresserunt
sensus vehementes et concitatos; nullam adhuc vocem audii ex bono lenique animosam. (Seneca, De clementia 2.2.3)
The choral ode in Oedipus prompts us to confront once again the overwhelming force of poetry, its ability to dredge up from memory and nefas
dark secrets and terrible truths, a cocktail of pleasurable and painful passions. It is a combination of emotions which brings us back to the same
issues that I faced when I began this investigation: what are the function
15
See Mastronarde (1970) 31011.
229
and effect of poetry? How are we, the audience, supposed to confront the
emotional reality of Senecas plays, sprung from their underworldly roots?
What role do these emotions play as we try to assess the ideological balance
of each play? Or, to put it in wider terms: can there be a truly coherent
Stoic theatre?
In search of answers to these questions it is interesting to turn once more
to Stoic thinkers in order to clarify, first of all, the theoretical framework
of the issues at stake. Afterwards, however, it will be poetry, again, which
must offer its own less systematic but hardly less compelling answers, in
the form of metapoetic images culled from the plays themselves.
Although they insist on the potential educational value of poetry,16 the
Stoics are also keenly aware of the possible dangers it presents to the audience. Plutarchs How the Young Man Should Study Poetry (Quomodo adulescens poetas audire debeat), which is very close, as far as we can ascertain,
to traditional Stoic thinking,17 expresses all these concerns very clearly.
Poetry can be dangerous for the very same reasons as it can be useful. It
is a form of expression which is more effective than prose, both because
it enables the poet to express himself more concentratedly, and because it
arguably captures the audiences attention better than prose. Cleanthes, as
we learn from a reference in Seneca (Ep. 108.10) formulated this thought
with particular clarity: Cleanthes used to say, As our breath produces a
louder sound when it passes through the long and narrow opening of the
trumpet and escapes by a hole which widens at the end, even so the fettering
rules of poetry clarify our meaning. 18
There are two main ways in which poetry can be dangerous. First of all,
listening to poetry produces pleasure in the listener, a passion that has to be
accounted for, justified and contained. It is an irrational movement of the
soul, and the Stoics must negotiate its existence by taking into account, on
the one hand, the necessity to avoid passions altogether and, on the other,
the potential benefit of the excitement of poetry for the reader. Secondly,
poetry can deceive the reader into endorsing morally objectionable ideas,
and, by representing passions in the characters, it can induce passions in
the audience.
These different forms of influence rely on the assumption that poetry
itself can be analysed into two separate entities, matter and form, or
in Stoic terms logos and lexis, a rhythmical pattern of sounds which is
16
17
18
De Lacy (1948) and Tieleman (1992) 21948.

De Lacy (1948) 250, n. 47. Nussbaum (1993) 122, too, uses Plutarch, with caution.
For a similar evaluation of Cleanthes see Philodemus, De musica (28.114 Neubecker = SVF
1.486).
230
peculiar to poetic expression. Much like music,19 to which Stoic authors

often compare it, poetry affects the hearer with harmonious sounds and the
appropriately composed relationship between different parts,20 and a wellcrafted form (lexis) alone can produce pleasure in the hearer, irrespective
of the content it expresses. As an irrational movement of the soul, such
pleasure should normally be avoided, but, if controlled and moderated, it
can be justified because of its educational benefit, provided that the content
of the poetry is morally acceptable. However, only those still struggling in
the way to wisdom, the proficientes, will need such inducements, which
the wise (sapiens) will normally eschew.21 It is necessary to remember, in
any case, that the ability of poetry to produce pleasure irrespective of its
moral contents is a constant danger which educators should carefully guard
against.
At the level of logos, or content, as well as of education, poetry appears to
be an equally double-edged form of expression. For one thing, poetry can
produce sympathetic passions, and, for this reason, it is essential that the
audience should restrain itself. The reader, says Plutarch (Mor. 16e):
will check himself when he is feeling wroth at Apollo in behalf of the foremost of the
Achaeans . . . he will cease to shed tears over the dead Achilles and over Agamemnon
in the nether world . . . and if, perchance, he is beginning to be disturbed by their
suffering and overcome by the enchantment, he will not hesitate to say to himself,
Hasten eager to the light, and all you saw here lay to heart that you may tell your
wife hereafter [Od . 11.2234].
Poetry can also represent immoral ideas and forms of behaviour. I quote
Plutarch again on this point (16de):
Whenever, therefore, in the poems of a man of note and repute some strange and
disconcerting () statement either about gods or lesser deities or about
virtue is made by the author, he who accepts the statement as true is carried off
his feet, and has his opinions perverted (
); whereas he who always
remembers and keeps clearly in mind the sorcery of the poetic art in dealing with
falsehood . . . will not suffer any dire effects or even acquire any base beliefs.
A little later (17d) Plutarch quotes two short passages from the Iliad and
Euripides Iphigenia at Aulis, and comments:
19
21
20 De Lacy (1948) 246 and n. 20, 248, n. 33.

De Lacy (1948) 248, n. 33.
But the Stoics also posit that even the philosopher can derive pleasure from poetry. His is not
but , a state of rational elevation which derives from a correct judgement, that is from a
correct evaluation of the contents of the poem in question: De Lacy (1948) 250, with n. 48.
231
These are the voices of persons affected by emotion and prepossessed by opinions
and delusions. For this reason such sentiments take a more powerful hold on us and
disturb us the more, inasmuch as we become infected by their emotions ()
and by the weakness from whence they proceed. Against these influences, then,
once more let us equip the young from the very outset to keep ever sounding
in their ears the maxim that the art of poetry is not greatly concerned with the
truth . . .
These two possibilities that poetry will stir passions, and that it will
liberate wrong ideas come closest to the core of the issue here at stake, for
they are inescapable in the case of tragedy. Plutarch gives an interestingly
simple answer to the issue (18ab). In the case of artists, such as painters,
who depict unnatural acts ( ), for instance Medea slaying
her children, it is especially necessary that the young man should be trained
by being taught that what we commend is not the action which is the subject
of the imitation, but the art, in case the subject in hand has been properly
imitated. To imitate beautifully means fittingly and properly and ugly
things are fitting and proper for the ugly (18d).
While Plutarch insists on the fact that education and judgement must
avoid the audiences endorsement of wrong ideas represented on stage,
Strabo seems more confident that the text can safely orient the reactions
of the audience. A pleasant mythos (he says at 1.2.8) produces in the audience an impulse ( ) towards that particular behaviour, while
a frightening mythos exercises a deterring effect called . This
clear-cut distinction, however, is undermined by the observation, in the
same paragraph, that fear, too, can produce pleasure,22 here called
or . De Lacy notes that the term is used in On
the Sublime precisely to describe the pleasurable excitement provoked in the
audience by great literature,23 a state closely akin to the enthousiasmos of
the poet producing such literature.24 Strabos apparent trust in the idea that
the intrinsic shape of the text can produce the desired effect of
or matches the rhetorical writers confidence that the reaction
of the audience can be safely oriented. This confidence, in turn, presupposes a set of shared values and cultural norms which shape the audiences
22
23
And what is new is pleasing, and so is what one did not know before; and it is just this that
makes men eager to learn. But if you add thereto the marvellous ( ) and the portentous
( ), you thereby increase the pleasure (), and pleasure acts as a charm to incite to
learning . . . Now since the portentous ( ) is not only pleasing (), but fear-inspiring
(
) as well, we can employ both kinds of myth [pleasing and fear-inspiring] for children,
and for grown-up people too.
24 Russell (1964) 62.
Subl. 1.4. De Lacy (1948) 270.
232
reactions and allow the rhetor to foresee them.25 As Ps.-Longinus points

out, however, phantasia operates differently in poetry and in rhetoric, since
the latter is bound by verisimilitude and must eschew exaggeration. The
impact of poetic enthousiasmos and phantasia are altogether less easy to
gauge in advance even within a cultural system which does not admit the
existence of a free, creative, modern imagination.26
So far, I have tried to show that Stoic theories on poetry fail to provide
an adequate explanation of how the moral and educational value of poetry
can be safeguarded in forms of poetry which represent negative exempla.
The next step is to investigate the dynamics of aesthetic reception: how the
audience perceives poetry and is influenced by it.
The Stoic theory of passion posits a fundamental distinction between
instinctive reactions and rational assent. When presented with a frightening appearance (species), any human being, wise or not, will receive an
involuntary impulse (ictus) which will make him or her jump, or blush, or
involuntarily perform any such acts. As Seneca points out in De ira 2.2.1,
these reactions are not controllable, and therefore reason cannot prevent
their unfolding (quorum quia nihil in nostra potestate est, nulla quominus
fiant ratio persuadet). Among the various examples of involuntary reactions,
De ira includes episodes related to the effect of poetry and other forms of
artistic expression (2.2.36):
This [impulse of the mind] (ictus animi) steals upon us even from the sight of plays
upon the stage and from reading of happenings of long ago. How often we seem to
grow angry with Clodius for banishing Cicero, with Antony for killing him! Who
is not aroused against the arms which Marius took up, against the proscriptions
which Sulla used? . . . [4] Singing sometimes stirs us, and quickened rhythm, and
the well-known blare of the War-gods trumpets; our minds are perturbed by a
shocking picture and by the melancholy sight of punishment even when it is entirely
just . . . It is said that Alexander, when Xenophantus played the flute, reached for
his weapons.
This impulse (ictus), however, is not a passion, because, again in Senecas

words (2.3.1), none of these things which move the mind through the
agency of chance should be called passions; the mind suffers them, so to
speak, rather than causes them (nihil ex his quae animum fortuito inpellunt
adfectus vocari debet: ista, ut ita dicam, patitur magis animus quam facit).
Essential to the existence of a passion (adfectus) is the assent (assensio) which
the receiver of such an ictus will or will not grant to the ictus. Passions,
says Seneca (2.3.1), consist not in being moved as a result of impressions of
25
Webb (1997) 1235 at 125.
26
Webb (1997) passim, esp. 1234; Pedrick-Rabinowitz (1986) 107.
233
things, but in surrendering oneself to them and following up this fortuitous

movement (ergo adfectus est non ad oblatas rerum species moveri, sed permittere se illis et hunc fortuitum motum prosequi). The involuntary reactions
to poetry or painting, in so far as they are the movement of minds not
wishing to be moved (motus . . . animorum moveri nolentium), are not proper
adfectus, but only principia proludentia adfectibus (2.2.5):
Such sensations, however, are no more anger than that is sorrow which furrows
the brow at the sight of a mimic shipwreck, no more anger than that is fear which
thrills our minds when we read how Hannibal after Cannae beset the walls of
Rome, but they are all emotions of a mind that would prefer not to be so affected;
they are not passions, but the beginnings that are preliminary to passions.
If we apply this doctrine to the case of poetry, or, more specifically,

to dramatic poetry, we can say that the text produces an image (species)
which provokes an impulse (ictus) of, for instance, fear, pleasure or hate,
but that only when the mind has granted its assent to such an impulse will
poetry have produced a real passion (adfectus). This structure appears to
be consistent with Plutarchs almost exclusive emphasis on the fact that the
audience should be educated to resist the impulse of poetry, and it entails
interesting consequences for the critical interpretation of Senecan tragedy.
Before, however, turning to these consequences by way of conclusion, I
would like to elaborate further on the dynamics of passions and on other
Senecan passages which shed light on the relationship between poetry and
adfectus.
In Letters to Lucilius 115, Seneca criticizes poetic endorsements of wealth
as a value (115.12): Verses of poets also are added to the account verses
which lend fuel to our passions (quae adfectibus nostris facem subdant),
verses in which wealth is praised as if it were the only credit and glory of
mortal man.
The torch is a remarkable metaphor for the ictus, since it implies not
only that the tragic endorsement of negative thoughts can produce an involuntary impulse, but also that such thoughts constitute intrinsically
dangerous temptations.27 As Seneca states in Letter 7.2, nothing is so damaging to good character as the habit of lounging (desidere) at the games;
for then it is that vice steals subtly upon one through the avenue of pleasure. The impulse, we know, can be resisted, but it is still a temptation,
indeed a dangerous one since it is associated with the pleasure of poetry, the
pleasure of hearing certain sounds and forms of expression, and, according
27
The torches are also traditionally associated with the Furies, whose connection with poetry I explored
above.
234
to Plutarch, the pleasure essentially linked with the very act of mimetic
representation.
In the following paragraphs of Letter 115, Seneca relates an anecdote about
Euripides career. Hearing some lines praising wealth,28 the audience rushed
forward, pushed away the actors and tried to end the performance. Euripides came to the stage and asked them to wait and see what end (exitum)
that admirer of gold would eventually get (115.15). This episode seems to
imply that, in order to preserve its educational value, tragedy should stage
evil actions only if it also shows the retribution they deserve. Again, we are
confronted not so much with the fact that Senecan tragedy represents immoral conduct, but that it conspicuously fails to offer a convincing image
of punishment.29
The relationship between passions and poetry that is established in the
passages above implies a remarkable shift of responsibilities from the author to the audience. Surely, the author is responsible for his intentions,
and should be judged accordingly (if we so desire). But, whatever these
intentions, the real burden of interpretation falls on the audience, and ultimately lies outside the authors sphere of influence. It is left to the audiences
interpretation, if they so wish, to fulfil the educational intentions of the author. It is perfectly possible to assume that Senecas intention in portraying
Medea was to move his audience to a stern criticism of the passions which
dominate her. But it is equally possible that, contrary to his supposed aims,
a reader will end up feeling great sympathy for Medea and her crimes,
and thus forgo any chance of being morally improved. In the hermeneutic
process that links the author to his text and the text to its audience, there
is a gap which the authors intention cannot bridge and which effectively
renders the question of the educational value of poetry aporetic.
Another section of Letter 7, the famous letter on circus games, provides
an interesting confirmation that this particular brand of the Stoic theory
of interpretation is heavily focused on the reactions of the audience. At
7.5 Seneca exclaims: Come now, do you not understand even this truth,
that a bad example reacts on the agent? Thank the immortal gods that you
are teaching cruelty to a person who cannot learn to be cruel. Although
some interpreters choose to read in these lines a disguised praise of Neros
28
29
Danae, fr. 324 Nauck2 ; Seneca mistakenly attributes the lines to the Bellerophon.
Letter 108 also acknowledges the possibility that adfectus be stirred by poetry (108.1112): but our
minds are struck (feriuntur) more effectively when a verse like this is repeated: he needs but little
who desires but little or He has his wish, who wishes only for what is sufficient. When we hear
such words as these, we are led towards a confession of the truth. Even men in whose opinion nothing
is enough, wonder and applaud when they hear such words, and swear eternal hatred against money.
When you see them so disposed, strike home, keep at them . . .
235
clemency, I am inclined to think that the person who cannot learn to be

cruel is Seneca himself, caught watching the slaughter of the arena despite
himself. Unlike the crowd, Seneca is able to resist the ictus that comes from
the performance because his moral principles lead him to deny his assent to
such a monstrosity. But, again, the point is precisely that he is uninfluenced
by the evil example in front of his eyes not because of its intrinsic evilness,
but simply because his principles are already against it. If his principles had
been different, the same text could have provoked very different results.
This applies to the spectators of tragedy as well. A negative spectacle will
probably not affect the sapiens, but it might well invite a wrong assent from
the morally weaker proficiens (let alone the insipiens). And, regrettably, there
are very few sapientes in this world.
a l l e g o r i e s o f s pe c tato r s h i p
gaudet magnus aerumnas dolor
tractare totas. ede et enarra omnia.
(Seneca, Troades 10667)
Senecan tragedy can often be seen to dramatize the emotional quandaries

of spectatorship, referring to watching and being watched as essential components of the actions performed on stage. It is easy enough to connect
this emphasis with more or less obvious features of Silver Latin aesthetics: an obsession with form over content; the substitution of voyeuristic
detachment for a sense of active purpose that is now regrettably lost; the
baroque (or Mannerist) penchant for overwrought, graphic descriptions.
Ovid started it, it is often assumed, with his flamboyant fantasy and his
passion for ekphrasis. The history of later reception seems to have crystallized this figurative quality of Ovidian and post-Ovidian representations
into consistent patterns: comparing the two is much like distinguishing
between the restrained Virgilian landscapes of Poussin and the colorful reworking of Ovidian themes offered by Titian or Rubens. To say that this
contrast is, by now, commonplace, is not to deny it a measure of truth,
even less a prima facie appeal. Yet it pays to set aside for a moment the
temptation to outline too neat a history of post-Virgilian literature, and
to concentrate instead on how seeing and watching function in a specific
Senecan context. I thus choose to discuss a scene from Troades which, I believe, offers the most engaging and poignant of paradigms. First, however,
I will concentrate for one last time on Thyestes itself, in order not to lose
sight of the relevance of the topic as far as this particular play is concerned.
236
At the very centre of the dramatic action in Thyestes stands the elaborate
relaying of the slaughter scene to a hardly composed chorus by an equally
distressed messenger. Seneca retains in this scene the structural pattern that
characterizes the whole play: the messenger, an unseen spectator, guarantees
that Atreus deeds are not lost for the chorus and the audience in spite of
the messengers own wish that the image of that nefas might abandon him:
what whirlwind will headlong bear me through the air and in murky cloud
enfold me, that it may snatch this awful horror from my sight? (6235: quis
me per auras turbo praecipitem vehet | atraque nube involvet, ut tantum nefas |
eripiat oculis?). In fact the messengers words haeret in vultu trucis | imago
facti (the picture of that ghastly deed still lingers before my face, 6356)
echo an expression previously employed by Atreus himself: already before
my eyes flits the whole picture of the slaughter (2812: tota iam ante oculos
meos | imago caedis errat), and thereby guarantee that the spectacle Atreus
had envisaged has now been carried out in all its enormity. We should also
remember that all those watching are in turn being watched: the Fury had
said early on that Tantalus ghost would be forced to watch, unseen, the
macabre developments of the plot: let blood mixed with wine be drunk
before your eyes (656: mixtus in Bacchum cruor | spectante te potetur).
The detailed description of the site where the slaughter takes place,
Atreus royal palace, is a privileged locus for analysing the mixed emotional
responses that are elicited by this spectacle, since the messengers reaction to
the horrific scene exemplifies the reactions that his own rhesis is expected to
provoke in the reader. (The chorus, who might otherwise fulfil this function,
is, as usual, rather restrained.30 ) The messenger makes only a lame attempt
to stifle the overpowering force of the human sacrifice that he has watched.
He starts by hoping that the dire image might be erased from his memory
(6245), and a few lines later he begs to be dragged away because that image
is still painfully stuck in his mind (6358). But his resistance is weak and
fleeting: the chorus does not have to prod him much in order to obtain
a full and emphatic report on the most minute aspects of the sacrificial
rite. The words of horror carry an irresistible force: they flow unchecked
past the superficial attempt at removal advocated by the messenger in selfprotection. Once his initial opposition is overruled, the horrendous scene is
recalled and described in an impressive display of eloquence. In his capacity
to recreate for our eyes an otherwise irretrievable scene, the messenger is
yet another authorial persona in the text and is subject in turn to the
disruptive dialectic of the repressed and the repressive that shapes the play
30
On the choruss reaction see above, p. 170.
237
at different levels. Here again we have a graphic account of a negative

force that the messenger wishes he could resist, but which demands to
be represented, and to which he can only succumb: there is no hiding
what he has seen, just as Tantalus shadow cannot help stirring up the
dramatic actions demanded by the Fury. The rhetorical elaboration of the
speech thus acquires a significance which goes beyond Senecas Baroque
predilection for florid expression. The self-conscious richness of language
in the rhesis becomes the clearest textual signifier of the unopposed literary
triumph of Atreus deeds; the pleasure of the messengers words, both the
pleasure he provokes in the well-read audience and the pleasure to which
he yields as he relishes his own description, stand in direct contrast to the
moral judgement of Atreus deeds that is suggested by the messengers and
the choruss incidental remarks. The Ovidian subtext which powerfully
structures Atreus thoughts about his revenge is also important here. We
remember that the figurative story sent to Procne by Philomela provokes
an explosive response (Met. 6.5816):
evolvit vestes saevi matrona tyranni
fortunaeque suae carmen miserabile legit
et (mirum potuisse) silet: dolor ora repressit,
verbaque quaerenti satis indignantia linguae
defuerunt, nec flere vacat, sed fasque nefasque
confusura ruit poenaeque in imagine tota est.
the savage monarchs wife unrolled the cloth and read the tragic tale of her calamity
and said no word (it seemed a miracle, but anguish locked her lips). Her tongue
could find no speech to match her outraged anger; no room here for tears; she
stormed ahead, ready to confuse right and wrong, her whole soul filled with visions
of revenge.
Fasque nefasque confusura ruit expresses, even more than the qualified
moral detachment of the narrator, the notion that such a reading (or vision)
will provoke emotional consequences of a morally ambivalent nature, a
collapse of clear-cut distinctions between fas and nefas.
The rhesis of Thyestes, as we have seen, problematizes the emotional
response of the readers and questions their possible identification with
the chorus: once again the poetic word appears to be working against the
repressive force of rational criticism. Nowhere in the Senecan corpus is
this issue thematized more effectively than in the final scene of Troades, a
tense three-way dialogue between the messenger, Hecuba and Andromache.
As the messenger appears and announces to both women that they have
suffered tremendous losses, Andromache invites him to relate the events in
detail (10657):
238

expone seriem caedis, et duplex nefas
persequere: gaudet magnus aerumnas dolor
tractare totas. ede et enarra omnia.
Recount the order of the slaughter, and unfold the story of the double crime: a
great grief delights to consider its sorrows entire: speak out and recount it all.
The messenger obliges by describing first the death of Astyanax. His language makes it clear that the underlying association is with a theatrical
performance, a tragically real one. The centre of the stage is occupied by
the tower from which the Greeks plan to throw the boy. Around the tower
stands a crowd of Trojans and Greeks alike (107587):
haec nota quondam turris et muri decus,
nunc saeva cautes, undique adfusa ducum
plebisque turba cingitur; totum coit
ratibus relictis vulgus. his collis procul
aciem patenti liberam praebet loco,
his alta rupes, cuius in fastigio
erecta summos turba libravit pedes.
hunc pinus, illum laurus, hunc fagus gerit
et tota populo silva suspenso tremit.
extrema montis ille praerupti petit,
semusta at ille tecta vel saxum imminens
muri cadentis pressit, atque aliquis (nefas)
tumulo ferus spectator Hectoreo sedet.
This tower, once famous and pride of the walls, is now a cruel outcrop, surrounded
on all sides by the spreading crowds of princes and common folk. The whole throng
has assembled abandoning the fleet. For some a distant hill offers a free view from
open ground, for others a high cliff, upon whose summit the eager crowd poised
the tips of its feet. A pine tree supports one man, a laurel another, another a
beech tree, and the whole wood quivers with its load of people. Another man
makes for the edge of a steep hill; yet another treads on a half-burned dwelling,
or the projecting masonry of the falling wall, and there is even (o abomination!) a
barbarous spectator who sits on Hectors mound.
Unlike the boy, the crowd weeps (10991100: non flet e turba omnium | qui
fletur); as soon as the messenger relates Astyanaxs brave resolve in jumping
voluntarily from the tower, Andromache interjects with a histrionic lament
(11049). Finally, we come to Polyxenas death. This time, the messenger
emphasizes even more the theatrical aspects of the scene (111828):
praeceps ut altis cecidit e muris puer
flevitque Achivum turba quod fecit nefas,
idem ille populus aliud ad facinus redit
239
tumulumque Achillis. cuius extremum latus

Rhoetea leni verberant fluctu vada;
adversa cingit campus et clivo levi
erecta medium vallis includens locum
crescit theatri more.31 concursus frequens
implevit omne litus: hi classis moram
hac morte solvi rentur, hi stirpem hostium
gaudent recidi.
When the boy fell headlong from the lofty walls and the Greek crowd had wept
for the wickedness it had committed, the same people turned back to another
wicked act and the tomb of Achilles. The Rhoetean waters beat on its far side with
gentle breakers; a plain fringes the near side, and a valley grows, rising with an easy
slope and enclosing a central space, like a theatre. The numerous throng filled the
whole shore: some believe the delay of the fleet will be ended by this death, others
are glad that the young shoot of the enemy has been pruned back.
The crowds reactions are especially singled out, as the messenger insists on
them three times (112831, 11367, 11478):
magna pars vulgi levis
odit scelus spectatque; nec Troes minus
suum frequentant funus et pavidi metu
partem ruentis ultimam Troiae vident.
A great part of the shallow crowd both hates the crime and watches it; the Trojans
as eagerly attend their own burial and panicked with fear look on the last fragment
of falling Troy.
terror attonitos tenet
utrosque populos.
Awe holds both peoples in shock.
omnium mentes tremunt,
mirantur ac miserantur.
The minds of all are trembling, they marvel and they pity her.
The final remark on the reaction of the spectators once again conjoins
the two different groups (11601):
uterque flevit coetus; at timidum Phryges
misere gemitum, clarius victor gemit.
Each crowd wept: but the Trojans sent up a hesitant groan, while the victor groaned
more loudly.
31
On the textual issues raised by these lines, and their bearing on the interpretation of theatri more,
see Fantham (1982) 377 and Zwierlein (1986) 11112.
240
There are two intertwined elements of great interest in this scene. The
first is the explicit use of the theatrical analogy which the messenger exploits
throughout as a structuring device for his narration. The second is the
insisted focus on the emotional reactions of the people watching the deaths
of Polyxena and Astyanax, which, in the context of that analogy, invites
reflection on the relationship between that form of involved spectatorship
and the reactions which the audience would be expected to experience
while watching the play.
The latter issue should perhaps be dealt with first. By watching watchers
watch an allegory of spectatorship32 the audience is naturally invited to
acquire a critical distance from the very act of watching. This is especially
true since the messenger carefully distinguishes the reactions of Greeks and
Trojans, and points out that the latter are able to refrain from crying out
although they would be expected to be hit harder by the tragic events unfolding before them. If the Trojans are able to avoid uncontrolled despair,
this should prove that a form of restrained spectatorship is indeed possible. One could consider this a form of critical spectatorship,33 since the
Trojans moderate reactions must doubtless be rooted in their rational determination to offset the effects of the painful scene they are watching with
the desire to maintain a dignified appearance in front of their oppressors.34
Furthermore, both Trojans and Greeks are watching real events, not a mimesis of those events. Thus, a fortiori, spectators should infer that they can
avoid being completely overwhelmed by emotions mediated by a mimetic
representation.
The analogy between the Greeks and Trojans watching the deaths of
Polyxena and Astyanax and the audience watching the play, however, is
not immediate. The tragedy, in fact, presents an intermediate level which
frames this allegory of spectatorship and complicates its extrapolation, since
the messenger relates the events not directly to the external audience (us),
but to an internal audience made up chiefly of Hecuba and Andromache.
They, too, react to the speech, and, although their emotional involvement
is obviously of a more directly compelling nature, their reactions represent
another point of reference and comparison which is offered to the ultimate audience of the play. After the first part of the messengers exposition
(1068103), Andromache reacts with an outburst of pained indignation
32
33
34
A term I borrow from Stam (1992) 29 and passim.

I anticipate here a reference to Martha Nussbaums notion of critical spectatorship which I discuss
more fully in the next section of this chapter, p. 244.
Presumably they must also moderate their cries for fear of retribution tyrants may even order
relatives to display joy after the execution of their family members: Jal (1963) 2867.
241
(110410). Shortly after, upon hearing of her sons brave behaviour, her reaction is one of pride, as she observes the similarity between son and father:
sic quoque est similis patri (1117). Finally, Hecuba expresses her feelings by
voicing her despair and regretting that she has to survive the demise of her
family (116577).
Between the events narrated and the audience watching the play there
are, therefore, a number of layers. The event is (i) watched by the Greeks,
the Trojans and the messenger, whose reactions are (ii) described to
Andromache and Hecuba, whose own reactions are, in turn, (iii) displayed
on the stage. At each of the first two levels, however, the tragedy foregrounds different modalities of reaction. The Greeks weep uncontrollably,
and watch despite hating what they have to see; the Trojans who cannot
but detest the events in front of them are able to superimpose a level
of rational consideration onto their reactions; Hecuba, Andromache and
the messenger react with violent emotions. The differing attitudes of the
Greeks and the Trojans prove that emotional reactions are not a direct, univocal consequence of emotional involvement: hence we lose, I suspect, the
exemplary value of metadramatic alienation. The mise en scène of spectatorship does invite the audiences critical reflection on its own acts, and thus
fosters the possibility of critical viewing. But the further complication of
this model finally turns it on its head, as it shows that no definite pattern of
behaviour is really predominant. The audience is left with the tantalizing
impression that a form of critical distancing is indeed possible, but that
no coherent prescription for it can be given. What the scene ultimately
provides is the illusion of critical spectatorship, a form of controlled reaction which is theoretically possible but actually elusive, since it depends
too much on individual attitudes and reactions. By multiplying the internal points of reference, and thus (apparently) offering substantial stimuli
for a critical analysis of the implications of spectatorship, the play finally
leaves the audience alone with, and probably puzzled by, its own critical
burden.
A critical juncture in the mirroring of spectatorship occurs in the final
scene of Troades, and in the poignant observation that the text reserves for
the behaviour of most Greeks in the shallow mass (magna pars vulgi levis),
which is described as hating the crime and watching it (odit scelus spectatque, 1129). The paratactic arrangement of these verbs cannot obliterate
the concessive force of odit: the epigrammatic tension of the expression
represents intrinsically contradictory emotions: the authors of the scelus
abhor it and yet are compelled to watch it. There are several disturbing
implications of this remark which need to be untangled. On the one hand,
242
we notice that the scelus resists control on the part of its creators, since it
elicits widely differing reactions from different groups of people: some of
the Greeks rejoice at the Trojans demise (11268: hi classis moram | hac morte
solvi rentur, hi stirpem hostium | gaudent recidi), others hate the crime. The
latter also show by their reaction that the staged scelus they have planned
cannot be controlled to its end, and in fact turns against them emotionally.
Moreover, it is clear from the messengers observations that the scelus commands attention in spite of its loathsome nature.35 It should be noticed,
however, that the concessive force of odit is far from determined by the syntax of the phrase; the modern reader, alerted by Freuds well-known dictum,
might well be inclined to suspect in such a vague syntactical arrangement
traces of a causal connection between odit and spectat. The scelus keeps the
vulgus riveted precisely because it is cruel and hateful.36
This suggestion finds circumstantial corroboration elsewhere in the
speech. Shortly after line 1129, the messenger describes Polyxenas arrival on
the scene of her death, exploiting the pathetic quality of the wedding-asfuneral motif.37 The extended simile at 11402 foregrounds the perverse
reaction of the crowd to Polyxenas beauty, even as the reference to natural events tries to downplay the disruptive potential of the pleasurable
association between imminent death and moving beauty. The three-line
portrait of Polyxena is fraught with erotic overtones. Her beauty is especially resplendent, the messenger confesses, in spite of her demure and
shy behaviour (1138: tamen), a detail which would clearly befit a bride,
and which in this context focalizes her as an object of sexual desire in
the eyes of the male narrator and the predominantly (we assume) male
crowd. Such an implication, moreover, is clearly brought out after the simile, as the messenger confirms that Polyxenas beauty stirs strong emotions
in the beholders and for the second time reinforces the association
between aesthetic pleasure and the awareness of a cruel, imminent death
(11438):
stupet omne vulgus, et fere cuncti magis
peritura laudant: hos movet formae decus,
hos mollis aetas, hos vagae rerum vices;
movet animus omnes fortis et leto obvius,
Pyrrhum antecedit; omnium mentes tremunt,
mirantur ac miserantur.
35
36
37
A paradox already discussed by Plato, Resp. 439e440a.

See Boyle (1994) 229 on the theatrical force of spectare.
The suggestion is made explicit at line 1132: thalami more praecedunt faces. For further references to
this motif see Schiesaro (1985) and Seaford (1987).
243
The whole crowd is dazed, and generally all men praise more what is about to die:
some the grace of her beauty moves, others her delicate youth, others the straying
course of fate: but her brave spirit moves all as it goes to confront death. She goes
before Pyrrhus, and the minds of all are trembling, they marvel and they pity her.
The generalizing remark at 11434 (et fere cuncti magis | peritura laudant) has
been suspected as an interpolation,38 but it actually reinforces the poignant
tone of the simile, and suitably glosses the contrasting emotions that are
experienced by the vulgus watching Polyxenas death.
The analysis of the final scene of Troades is consistent with my previous reading of Stoic theoretical statements on the emotional impact of
performance. Critical distance is severely jeopardized, if not annulled, by
the voyeuristic involvement of the audience in the spectacle that they are
watching. It is not difficult to see how this involvement plays an important
role in Thyestes as well. There Atreus himself is the delighted spectator of
his own creation, who relishes all the details of Thyestes distress: miserum
videre nolo, sed dum fit miser (907), he exclaims, as the curtain rises on the
crowning glory of his masterpiece, a drunken, ignorant Thyestes acting out
a grotesque combination of inner pain and outward intoxication. While
the reading I offer is always open to the objection that, in the final analysis,
Atreus could actually be set up to provide a negative model of spectatorship
(he is pleasurably affected by tragic events), this strategy would be less easy
to uphold with respect to the Trojan audience, which is legitimately upset
by the horrors it observes. In both cases, however, what we, as the audience,
are offered, is the example of an internal audience deeply affected by the
events in front of them, and unable to resist the ictus they receive.
the challenge of epos
If Stoic theories fail to offer a coherent and fully convincing account of how
the emotional impact of the tragedies could be contained, and prevented
from affecting the audiences internal balance, the allegories of spectatorship that I have examined appear to confirm that it is indeed difficult, if
not impossible, effectively to insulate viewers from the pathos that they
experience. Stoic theories emphasize the critical burden of the spectators,
who are asked to evaluate the moral implication of the text without a firm
and unequivocal internal point of reference.
38
Lines 1143b1144a as well as 1147 are deleted by Zwierlein (1976) 190 and 188 respectively, as interpolated comments. Fantham (1982) 380 deletes 1143b1144a but not 1147. Boyle (1994) rightly retains
both. Further see Boyle (1994) 231 for a discussion of the role of spectacle in tragedy according to
Aristotle, with reference to these lines.
244
My reading of the Stoic sources, and their implications for an understanding of Senecan drama, is, however, only one possible reading. For
instance, Martha Nussbaum has eloquently argued that the Stoics intend
to promote a critical spectatorship, and that such a form of spectatorship
may escape the seemingly unsolvable contrast outlined by Stoic sources between the potential benefits and dangers of poetry.39 The critical spectator
will observe the tragedy with a concerned but critical detachment and will
analyse every aspect of the play with a reasoned coolness: the Stoics hope
to construct a spectator who is vigilant rather than impressionable, actively
judging rather than immersed, critical rather than trustful.40 I sympathize
with Nussbaums assertion that we should look at the spectator as the locus
for a resolution of these tensions. I am less optimistic, however, about the
expectation that Senecas tragedies can be seen to bear out such a hypothesis. True, the insistence of many plays on passions and their inner workings
does highlight one of the elements that are crucial to the formation of a
critical, detached spectator, who is reminded of the existence of such mental processes. But I am not sure that the repellent nature of many central
characters discourages the audiences emotional identification, or that we
could consider the choruss moralizing orthodoxy as a guide for the spectators response.41 But is the spectators identification with such powerful
characters as Medea or Atreus really discouraged, especially when their apparent lack of reason (of ordinary reason) is set against the commonplace
superficiality or moralizing dullness of the characters who surround them,
primarily the chorus? And especially when the central negative character of
the play is invested with the responsibility and prestige of creating his or her
own play: can we really loathe Atreus if we enjoy Thyestes? This is not to say
that the notion of critical spectatorship could not be precisely the answer
that Seneca himself would have given if asked how he would justify his
poetic project on a theoretical level. What he could not have guaranteed,
however, is that this solution would actually work in his tragedies, that the
philosophical infrastructure of his plays effectively avoids the possibility of a
39
40
Nussbaum (1993). Nussbaum distinguishes between two different Stoic views, the non-cognitive
and the cognitive, represented respectively by Posidonius (and, to a large extent, Diogenes of
Babylon), on the one hand, and Chrysippus, Zeno, Seneca and Epictetus, on the other. The noncognitive position argues that emotions are non-rational movements which poetry can order by
equally non-rational forces such as rhythm, harmony and melody. The authors of the cognitive
line insist that emotions are evaluative judgements, and that poetry has an educational function
in as far as it tries to modify those judgements. Pratt (1948) argues, on the contrary, that Senecas
tragedies reject Chrysippus theory of passions while embracing Posidonius notion that irrational
emotions have no cognitive value. Tragedies would thus be, like music, means to affect the irrational
(!") and emotional ( ) part of the soul through the irrational ( # $ ").
41 Nussbaum (1993) 148.
Nussbaum (1993) 137.
245
misinterpretation which would transform his supposedly didactic project

into a dangerous source of passion and turmoil.
The notion of critical spectatorship is rooted in Brechts theories on
drama. To bring Brecht42 into the picture is doubly interesting, first because
it enables us to explore further the issue of the audiences reaction, and
secondly because it may be helpful in explaining the generic interaction
between drama and epic which we have already had an opportunity to
confront.
The idea that tragedies should inspire strong emotional reactions finds
its most influential expression in Aristotles canonic definition of the genre
(Poetics 1449b248):
tragedy is, then, a representation of an action that is heroic and complete and of
a certain magnitude by means of language enriched with all kinds of ornament,
each used separately in the different parts of the play: it represents men in action
and does not use narrative, and through pity and fear it effects relief to these and
similar emotions.
The opposition that Aristotle draws here between drama and narration
(""%) is predominantly concerned with the tragedys different impact on the audience and is echoed in the sections of the Poetics which discuss
the relative features and merits of tragedy and epic.43 Thus, towards the end
of the surviving part of the essay, Aristotle notices that the tragedys reliance
on music (something which epic does not have) increases the pleasurable
effects of tragic poetry by making them more evident (&");44 tragedy
is more compact and concentrated than epic, and greater concentration
is more pleasurable than dilution over a long period: suppose someone
were to arrange Sophocles Oedipus in as many hexameters as the Iliad .45
Drama affects the audience more deeply than epic does, since its strategies
of communication, its plotting of actions, and its forms of expression hold
readers and spectators more deeply enthralled in the mimesis in front of
them. Drama relies on a swift, tightly connected succession of events, on
a forward-looking momentum which inevitably builds up suspense and
preludes to climax. As Friedrich Schiller remarked several centuries later in
his insightful commentary on Aristotles fundamental opposition,46 dramatic action moves in front of me, while I move around epic action, and
it does not, as it were, appear to move.47 Schiller emphasizes especially the
different effects on the readers:
42
43
46
There is only a cursory reference to Brecht in Nussbaum, who is mainly concerned with ancient
sources. For a cogent argument on the limitations of Brechts view of Greek theatre see Lada (1996).
44 Poetics 1462a.
45 Poetics 1462b.
Poetics 1449b; 1461bff.
47 Letter of 26 December.
In an epistolary exchange with Goethe in December 1797.
246
When the event moves in front of me [i.e. in the case of dramatic action], I am
firmly shackled to the present as I apprehend it through my senses, my phantasy
loses all freedom, a constant restlessness rises in me and stays in me, I must always
remain by the object, all looking back, all reflection is forbidden, because I am
following an external force.
Not so with epic, which allows ample freedom for readers to set their own
pace: I can proceed at uneven steps, I can, according to my subjective need,
linger for a longer or shorter time, I can move backward or forward . . . I
maintain a quiet freedom.48
Schillers masterful amplification of the contrast drawn by Aristotle already brings us back to some of the central interpretative concerns about
Senecan drama which I have highlighted above. Yet it is necessary to move
one step ahead and interrogate another theoretical application of Aristotles
opposition before I return to our author. A direct link connects Schillers
words with what arguably remains the most incisive attempt at a redefinition of theatrical communication in the twentieth century. Around the
names of Peter Szondi, Erwin Piscator and especially Bertolt Brecht, a fullyfledged theory of epic theatre has evolved, one which brings the contrast
between epic and drama to its most radical consequences and suggests that
drama should utterly renew itself by abandoning the essential characteristics
which set it apart from epic. Brechts theoretical reflection proves invaluable
for widening our notion of theatre beyond the norms that are powerfully
encoded in Aristotle. Comparing his notion of epic theatre with Senecas
experimentalism, anachronistic as it obviously is, can have a heuristic value.
Beginning in the 1920s, Brechts theoretical reflections and his dramaturgic activity are affected by his thorough re-evaluation of the basic premises
on which theatre, even modernist theatre, had so far been predicated. His
central point of dissatisfaction with bourgeois theatre is that it encourages
the audiences emotional identification with the characters and actions on
the stage, and thus prevents them from reflecting critically on the circumstances which govern their lives, on the power structures which silently
articulate their fate. Alienation (the A-factor) is the only effective means
of acquiring knowledge, and alienation should govern both the authors
construction of the play, the directors staging of it, the actors acting and,
finally and consequently, the audiences reactions. With epic theatre:
[t]he stage began to tell a story. The narrator was no longer missing, along with
the fourth wall. Not only did the background adopt an attitude to the events
on the stage by big screens recalling other simultaneous events elsewhere, by
48
Ibidem.
247
projecting documents which confirmed or contradicted what the characters said,

by concrete and intelligible figures to accompany abstract conversations, by figures
and sentences to support mimed transactions whose sense was unclear but the
actors too refrained from going over wholly into their role, remaining detached
from the character they were playing and clearly inviting criticism of him.
The production took the subject-matter and the incidents shown and
put them through a process of alienation: the alienation that is necessary to
all understanding. When something seems the most obvious thing in the
world it means that any attempt to understand the world has been given
up.49 It is thanks to these strategic choices that the spectator was no longer
in any way allowed to submit to an experience uncritically (and without
practical consequences) by means of simple empathy with the characters
in a play .50 This form of theatre which Brecht, drawing on Aristotles
opposition, calls epic theatre effectively becomes a form of instruction,
forcing the audience to acquire critical distance and to react rationally to the
staged scene. The contrast between dramatic theatre and epic theatre thus
becomes central to his theoretical approach. The opposition, he cautions,
is not made of absolute antitheses but of mere shifts of accents.51 It is
mostly a matter of different methods of construction which depend on
the different way of presenting the work to the public.52
The audiences reactions to the epic, as opposed to the dramatic theatre,
are described by Brecht in terms which echo Schillers intimations:
The dramatic theatres spectator says: Yes, I have felt like that too Just like
me Its only natural Itll never change The sufferings of this man appal me,
because they are inescapable Thats great art; it all seems the most obvious thing
in the world I weep when they weep, I laugh when they laugh.
The epic theatres spectator says: Id never have thought it Thats not the way
Thats extraordinary, hardly believable Its got to stop The sufferings of this
man appal me, because they are unnecessary Thats great art: nothing obvious in
it I laugh when they weep, I weep when they laugh.53
There are numerous elements of great interest in Brechts and Schillers

analysis of the spectators reaction to these two different forms of theatrical
communication. Indeed, it would be possible to adopt their description of
the effects of epic theatre on the public and apply it to Senecan tragedy,
where it would buttress the assumption that the plays are meant to furnish
elements of moral and philosophical instruction, and do so by exploiting a
form of critical spectatorship derived from alienation. However, I will first
explore further the contribution that Schillers and Brechts insights make
49
52
Brecht (1936) 71.

Brecht (1936) 70.
50
53
Brecht (1936) 71.

Brecht (1936) 71.
51
Brecht (1930) 37, n.1.
248
to an understanding of the different methods of construction presiding

over epic and dramatic theatre.
As Brecht rightly emphasizes, dramatic elements can be found in epic
works, and vice versa:
The bourgeois novel in the last century developed much that was dramatic, by
which was meant the strong centralization of the story, a momentum that drew
the separate parts into a common relationship. A particular passion of utterance,
a certain emphasis on the clash of forces, are hallmarks of the dramatic. The epic
writer Doblin provided an excellent criterion when he said that with an epic work,
as opposed to a dramatic, one can as it were take a pair of scissors and cut it into
individual pieces, which remain fully capable of life.54
Brecht is looking at a central feature which distinguishes epic from drama

from the point of view of plotting and pacing. Drama creates tension by
its strict adherence to unity of action, but also by organizing each scene in
a continuum which creates an ever-increasing tension. What Brecht calls
the linear development of drama as opposed to the epics development in
curves55 is responsible for the dramas premium on emotional solutions
which are precipitated by an unstoppable crescendo of tension. Walter
Benjamin provided an apt analogy for this structural model when he intimated that the peripeteia (and, we might add, anagnorisis with it) is the crest
of the wave which breaks and sweeps the audience with it and rolls forward
to the end.56 Epic relishes a plot which might ultimately be teleological, but
which relies heavily on delay rather than suspense, on structural parataxis,
on changes of viewpoints, on carefully orchestrated movements backwards
and forwards in the temporal frame of the narration. The narrators who
control the epic narrative are free from many of the constraints imposed
on the dramatic writer, since they can alter the linear arrangement of the
plot almost at leisure. They can pause to describe and refrain for a while
from portraying actions. They can move sideways to a different plot or subplot without destroying the texture of their creation. In dramatic theatre,
Brecht says, one scene makes another. In epic theatre, as evolutionary
determinism is replaced by jumps, each scene [is] for itself .57
I have insisted especially on the different notions of internal time and
scene-succession which characterize epic and drama because they recall
important (if problematic) aspects of Senecan theatre. Seneca, it is often
claimed, constructs his plays as a sequence of relatively unconnected scenes
which are not organically linked: such a lack of what Brecht would call
54
56
55 Brecht (1930) 37.

Brecht (1936) 70.
57 Brecht (1930) 37.
Quoted by Stam (1992) 41.
249
growth (as opposed to epic montage) is strongly reinforced by the choruss suspended, dream-like interventions, which do nothing to ease the
transition among different moments of the plot. The internal articulation
of time, too, is remarkable. As we have seen in Thyestes and in Troades,
for instance, Seneca tampers with the linear, unidirectional flow of time
which is essential to traditional drama, and he replaces it with a complex
intertwining of different temporal levels. In Thyestes, we also noticed, the
centrality of Atreus killing and his subsequent revelation to his brother
are further problematized by the framing structure, which makes it clear
that the central plot, so traditionally hinging on such climactic moments,
is but a part of a larger whole whose borders are nowhere to be grasped
in the play as we see it: all the play ultimately guarantees is that we can
glimpse snippets of an extended sequel of actions without being able to
know exactly how they may evolve.
What are we supposed to do with the substantial presence of epic elements in Senecan tragedy? One solution could be to extend Brechts observations about the fact that dramatic and epic elements inevitably coexist in
various artistic forms, and thus, in effect, expropriate the issue of much of
its interpretative potential. Or, on the other hand, we could fully embrace,
as I mentioned, all the corollaries which Brecht suggests concerning the
effects of epic theatre on the audience. Again, I would first like to explore
a middle ground which privileges literary history before turning to the
epistemological and cognitive implications of Senecas strategy.
We would do well to recognize, first of all, that by the time Seneca wrote
his plays epic had become an extremely flexible and far from unambiguous
medium in Latin literature. The epic thread in a text which is still predominantly dramatic is much less surprising after Virgil and Ovid, who
had shown that within an epic frame dramatic and dialogic scenes could be
combined with sustained narratives and the constant presence of one (or
indeed more than one) narrator. Virgil had shown, too, that extensive contacts with tragedy were essential to his poem, and he appropriated both
specifically and generically a great deal of the Greek and Roman tragic
tradition. Ovid, for his part, demonstrated that the interplay of frames and
contents, narrators and tales, were essential elements in the works fractured and polyphonic signification. The Metamorphoses are structured on
the combined claims to truth and reliability of different narrators, whose
narratives are often nested one inside the other in a confusing array of layers: some narrative frames, for instance, remain open for hundreds of lines.
And, of course, what Ovid ultimately accomplishes is a violent disruption
of the notion of closure.
250
The bewilderment which is part and parcel of the effect of the Metamorphoses ultimately deprives epic of some of its supposedly critical function
vis-à-vis drama. The Metamorphoses show how an energetic exploitation of
epics structural freedom its possibility to shift narrative, narrators and
time can lead not to the critical distancing and empowerment of the readers, but rather to a form of confusion which effectively denies the privilege
of insulation, rendering them victims of an emotional identification which
epic could theoretically discourage. A certain level of framing helps readers
understand their position in the flow of narratives, provides critical distance
and ultimately affords them a privileged point of view which they might
understandably mistake for that of the narrator. But Ovid multiplies this
effect to the point that readers, dazzled by the web of frames within frames
and narrations within narrations, can only forget the larger, critical picture, and are as likely as the audience of a play to focus on the tale in hand,
identifying with a character while forgetting what is implied or suggested
by part or all of the metadramatic framing. There is a noticeable analogy
between this technique and Senecas predilection for rather detached acts
within a play, and both authors have often been censored for what critics have seen as their inability to create a coherent continuum between
scenes. Such a continuum is exactly what works such as Metamorphoses or
Thyestes radically question, by adopting, as we have seen, an internal logic
which makes more space for patterns of thought akin to the working of the
unconscious.58
When Seneca conspicuously introduces epic elements into his drama he
accomplishes more than a mechanic Kreuzung der Gattungen, but less than
a total revolution. Epos no longer guarantees the interpretative effects that
would be consistent with a distinctly didactic view of Senecan drama, since,
at a basic level, its narrative norms are no longer based on clearly articulated
structural patterns. Long rheseis and extended similes violate the relative
stylistic homogeneity of the tragedy, without offering, on an ideological
level, any less troubled or disconnected impression of reality. Stories such
as those of Tereus and Polyxena hark back to epic, where in turn they had
been imported from tragedy. The safety of an external epic that is relatively
unscathed by the pervasive violence of passions is questioned when tragedy
and epic intersect not just once, but twice, and project onto each other a
relativizing, troubling shadow.
58
On episodic narratives in imperial literature see Williams (1978) 24653 and Johnson (1987) passim.
For the unconscious rejection of temporality and causality see above, p. 211.
251
In the context of Senecan drama, therefore, even the Brechtian notion

of epic theatre as a guarantor of critical distance can hardly ensure the
ethical and didactic viability of the plays. Rather, the comparison between
Brecht and Seneca illustrates how Seneca contaminates epic with tragedy
far more than he disinfects drama with it. Framing, we have seen, is the
main vehicle of metadramatic reflection in the play, as it enacts, before the
spectators eyes, the constructedness of the tragic experience and encourages
speculation on the specific literary features of the tragedy. Responsibility is
emphatically foregrounded, and an involuntary acquiescence on the part
of the audience made more and more difficult.
Epilogue
Verum . . . nulli . . . nisi audituro dicendum est

(Seneca, Letters to Lucilius 29)
The analysis of Stoic perception and evaluation of literary phenomena offers

an interesting way out of the dilemma of reading the tragedies as either
enactments or extended refutations of Stoic dogmata. The very explanation
of how passions work and are perceived which Seneca offers in accordance
with Stoic principles makes the effect of a literary utterance less safely
ascertainable than one would like to expect. At the level of assensio, which
is the critical juncture in the development or forestalling of a passion, readers
are left alone with their hermeneutic burden. They might have thoughtful
teachers to guide them in the process, much as Plutarch recommends, but
the author of the text, with his responsibilities and intentions, is inevitably
out of the picture.
Predictably, the situation is more muddled than this. Tragedy involves
conflict, the battle between two sides, whose respective stances must be
represented with equal accuracy and conviction if the play is to be effective.
Bad behaviour will take centre stage, represented with accuracy and artistic
as well as psychological credibility. This fact alone introduces into the play
a degree of openness and ambiguity that no amount of authorial intention
can hope to dispel for good. As I said earlier, I find wholly unpersuasive
the proposition that Seneca must secretly have meant his tragedy to be a
systematic refutation of the philosophical positions that are advocated in
his prose. But for all the reasons explored above, his choice of the tragic
form is inevitably perilous and ambivalent. A full recognition of the doubleedged powers of poetry, a recognition which could derive directly from the
theoretical principles of Stoic poetics, would have perhaps recommended a
different course of action. For instance, it could have supported an attempt
at poetry along the lines of Cleanthes Hymn to Zeus, where the pleasurable
impact of the medium is put to the service of an impeachable moral lesson.
252
Epilogue
253
But to resort to the psychagogic qualities of poetry in order to represent

negative passions, to create powerful phantasiai of winning evil, is to take
a step too far on the path to collusion with the enemy.
Senecas tragedies do precisely what most tragedies, or at least good
tragedies, must do: they present a forceful display (phantasia) of contrasting forces and passions and ask the audience, brought to ecstasy (
), to examine their feelings and assumptions. This examination will be
all the more difficult, and all the more rewarding, the more the poet will
have been able to make a compelling case for the very forces that constitute
tragedy. This position is perfectly in line with what our Stoic sources have
to offer on the subject of poetics, but they cannot go further than that. As I
have tried to show, when they try to explain how, exactly, the fearful myths
presented by tragedy can produce steering away (

) rather than
incitement (

), they do not offer anything more than a suspiciously circular argument. This is true in a particularly poignant way in the
case of tragedies, such as many written by Seneca, which do all they can
to blur the possibility of a clear-cut ethical reading, by presenting figures
such as Medea and Atreus as deeply connected with the fascinating tension of poetic creation, for instance, or by depriving characters such as the
satelles and the chorus in Thyestes of much of the poetic credibility and ethical consistency which might mould them into powerful counter-examples.
The audience, at least an audience of proficientes, will only be able to resist
the wicked allure of the various forms of tragic passion on the strength of
previously held moral convictions, which can thus be tested and perhaps
strengthened. Only in this very restricted sense does tragedy preserve an
educational function, the only function which Stoicism is ready to grant
it. But the risks implicit in writing tragedy are considerable, all the more
so, as I just emphasized, in the case of Seneca. What we ultimately face is
the impossibility of Stoic tragedy. For sapientes will have no interest in it,
and proficientes are as likely to be deceived by it as they are to draw useful
precepts. As Seneca admits in one of his Letters to Lucilius (29.1), for one
must not speak the truth to a man unless he is willing to listen (verum . . .
nulli . . . nisi audituro dicendum est).
Where do these theoretical considerations leave the present reading of
Thyestes? Above all, I hope they might allow an appreciation of this and
other plays that is less focused on the philosophical truth they supposedly
encode, less predicated, that is, on the dubious assumption that a final
message can indeed be ascertained. I am not advocating a free-floating
indeterminacy as much as I am trying to place tragedys complex signifying
strategies squarely at the centre of the readers and the critics attention.
254
There is no code to be broken: rather, there are emotions to be experienced

and negotiated, rational intimations to be assessed.
Everything in Thyestes points to future evil. The alternae vices of the
House of the Pelopidai admit no other solution. What we have watched is
but a fragment, a segment of a longer, more painful story. The curtain comes
down to announce not the end, but to be continued. The intersecting
layers of dramatic action that structure the play offer an illusion of order
and enclosure just as they remain ultimately at the mercy of an exterior,
non-human force, a Fury that has appeared and disappeared to motivate
the action and can return, as she indeed will, at any moment. In this respect
everything, even Atreus undoubted triumph, is momentary and elusive.
Thematically and ideologically, the play is a celebration of nefas: it reaffirms its poetic excitement, its dazzling aesthetic quality, its right to be
foregrounded. But it shows, or at least hints at, more: that perhaps fas and
nefas, just as they cannot be taken as reliable guides in deciding a priori
what can actually be said or not be said in poetry, and what will or will
not be poetically successful, might not even constitute reliable moral definitions in themselves. There is no way in which we can escape the moral
dilemma that Seneca imposes on us when he celebrates Atreus deeds, since
he explicitly connects them to the appeal of poetry; but he goes further,
since he relentlessly undermines the superficial moral judgement expressed
by the chorus: Atreus does have his reasons; the murders he commits could
be seen as an archaic, fascinating ritual in which he tries to restore, by
horrendous retaliation, the integrity of the genos that has been obscenely
perverted by Thyestes. Once these elements of doubt are allowed to creep
into the otherwise neat system of Stoic morality, there is no way back; once
we start doubting what is fas and what is nefas, who is right and who is
wrong, the moral certainties which Seneca seems to offer in much of his
work begin to crack (perhaps they were a generous illusion all along). The
fact is, and Thyestes shows it splendidly, that the other world, the one of
passions, blood, revenge, hatred, deceit and darkness, has its appeal; its
horror is inextricably fused with voluptas, its pleasure and beauty (moral
or otherwise) forever disjointed. The sparagmos is now complete, since no
certainties, whether moral or poetic, survive the force of the tragedy intact.
Like Procne, we are left to storm ahead confusing right and wrong.
Atreus is, simply, too good not to be true. His overwhelming physical
domination of the play in every conceivable aspect triggers the audiences
emotional response to a degree which is unparalleled by any other character.
The issue, again, is not whether Seneca, in a round-table debate over Poetry
and Emotions with his fellow Stoics, would have defended him or not; in
Epilogue
255
all likelihood he would have argued that yes, the moral lesson of the play is
to be found in the choruss well-meaning purple passages about powers selfrestraint, or in the satelles sheepish advocation of a moderate, considerate
tyranny. Or again, as he defensively puts it in De vita beata, he would have
defended Thyestes on the basis that, after all, what you really need to do
is to try to be wise (repent and you will be saved). Poètes maudits are (at
least in their theoretical dimension) a very recent discovery, and there is
no point in asking Seneca to provide a satisfactory prototype. But Seneca
could no better control the implications and emotional provocations of his
play than any modern author can. And, without purporting to explore his
own personal ambivalences about the kind of writing he offers, I suspect
that he would have been the first to admit (perhaps sotto voce) that he
had intentionally stacked the cards against Thyestes, the satelles, and the
dully moralizing chorus. There is too much pleasure in Atreus for the
author or the audience to be unaffected by it. Atreus is fulfilled, emotionally
and artistically, not to mention intellectually. He is, in his own words, a
(pagan) god who has scared away the pious gods of traditional religion, the
guarantors of a world order whom Thyestes ineffectually invokes even as
their power has been shattered by Atreus determined cruelty. He relishes
passions and he relishes pain, his enemies pain. There is pleasure to be
found in passions, and it is a pleasure that Thyestes and Atreus want us to
share with them.
Bibliography
Abbreviations of journal titles correspond to those used in the Oxford Classical

Dictionary.
Abel, L. (1963) Metatheatre: a New View of Dramatic Form, New York
Anliker, K. (1960) Prologe und Akteinteilung in Senecas Tragodien, Bern and
Stuttgart
Auerbach, E. (1959) Mimesis. Dargestellte Wirklichkeit in der abendlandischen
Literatur, 2nd edn, Bern (= Mimesis. The Representation of Reality in Western
Literature, Princeton, 1953)
Baar, M. (1990) Das Bild des Kaisers Tiberius bei Tacitus, Sueton und Cassius Dio,
Stuttgart
Baker, H. (1939) Induction to Tragedy, Baton Rouge
Baldo, G. (1989) I mollia iussa di Ovidio (ars 2, 196), MD 22: 3747
Barchiesi, A., ed. (1992) Ovid, Epistulae Heroidum 13, Florence
(1994) Il poeta e il principe, Rome and Bari
Barchiesi, M. (1981) I moderni alla ricerca di Enea, Rome
Barnes, J. (1978) La doctrine du retour eternel, in Brunschwig (1978) 320
Barrett, W. S., ed. (1964) Euripides, Hippolytos, Oxford
Bartezzaghi, S. and del Gaudio di Jueli, S. (1988) The quest and the doubt. Knowledge and emotions in Oedipus: between Sophocles and Seneca, VS 501:
5767
Barthes, R. (1964) Essais critiques, Paris
Barton, C. A. (1993) The Sorrows of the Ancient Romans, Princeton
Barton, T. (1994) The invention of Nero, in Elsner and Masters (1994) 4863
Bartsch, S. (1994) Actors in the Audience: Theatricality and Doublespeak from Nero
to Hadrian, Cambridge, MA and London
Bate, J. (1995) Introduction to W. Shakespeare, Titus Andronicus, ed. J. Bate,
London
Bergren, A. L. T. (1983) Language and the female in early Greek thought, Arethusa
16: 6995
Bessone, F., ed. (1997) Ovid, Heroidum Epistula XII Medea Iasoni, Florence
Bettini, M. (1984) Lettura divinatoria di un incesto (Seneca Oed. 366ss.), MD 12:
14559
256
Bibliography
257
(1987) Bruto lo sciocco, in (various authors) Il protagonismo nella storiografia

classica, Genova, 71120
Bickerman, E. J. (1980) Chronology of the Ancient World , rev. edn, London
Billerbeck, M. (1988) Senecas Tragodien. Sprachliche und stilistische Untersuchungen,
Leiden and New York
Biondi, G. G. (1984) Il nefas argonautico. Mythos e logos nella Medea di Seneca,
Bologna
Bodei, R. (2000) Le logiche del delirio. Ragione, affetti, follia, Rome and Bari
Bomer, F., ed. (196986) Ovid, Metamorphosen, Heidelberg
Borgo, A. (1993) Lessico parentale in Seneca tragico, Napoli
Bowie, A. M. (1993) Aristophanes. Myth, Ritual and Comedy, Cambridge
Boyle, A. J., ed. (1983a) Seneca tragicus. Ramus Essays on Senecan Drama, Berwick,
Victoria
(1983b) Hic epulis locus: the tragic worlds of Senecas Agamemmnon and
Thyestes, in Boyle (1983a) 199228
ed. (1994) Seneca, Troades, Leeds
(1997) Tragic Seneca. An Essay in the Theatrical Tradition, London
Braden, G. (1985) Renaissance Tragedy and the Senecan Tradition. Angers Privilege,
New Haven and London
Bramble, J. C. (1974) Persius and the Programmatic Satire, Cambridge
Brecht, B. (1930) Das Moderne Theater ist das Epische Theater. Anmerkungen
zur Oper Aufstieg und Fall der Stadt Mahagonny, in Schriften zum Theater.
Uber
eine nicht-aristotelische Dramatik, Berlin and Frankfurt a.M., 1957
(quoted from the English translation The modern theatre is the epic theatre
(notes to the Opera Aufstieg und Fall der Stadt Mahagonny), in J. Willett,
ed., Brecht on Theatre. The Development of an Aesthetic, London, 1974,
3342)
(1936) Vergnugungstheater oder Lehrtheater, in Schriften zum Theater, above
(quoted from the English translation Theatre for pleasure or theatre for
instruction, in J. Willett, ed., above, 6977)
Brink, C. O. (196382) Horace on Poetry, 3 vols., Cambridge
Brooks, P. (1984) Reading for the Plot. Design and Intention in Narrative, New York
Brunschwig, J., ed. (1978) Les Stociens et leur logique, Paris
Brunschwig, J. and Nussbaum M. C., eds. (1993) Passions and Perceptions. Studies
in Hellenistic Philosophy of Mind, Cambridge
Burkert, W. (1983) Homo necans: the Anthropology of Ancient Greek Sacrificial Ritual
and Myth, Berkeley, Los Angeles and London (original edn, Berlin and New
York, 1972)
Calboli, G., ed. (1993) Cornificius, Rhetorica ad Herennium, 2nd edn, Bologna
Calder, W. M. (1970) Originality in Senecas Troades, C Phil. 65: 7582
(1975) The size of the chorus in Senecas Agamemnon, C Phil. 70: 325
(1976) Senecas Agamemnon, C Phil. 71: 2736
(197677) Seneca: tragedian of Imperial Rome, CJ 72: 111
(1983) Secreti loquimur: an interpretation of Senecas Thyestes, in Boyle (1983a)
18498
258
Bibliography
(1984) Seneca Thyestes 1016, C Phil. 79: 2256

(1989) The authenticity of Seneca, Thyestes 336338, Mnemosyne 42: 1079
Calderwood, J. L. (1971) Shakespearean Metadrama, Minneapolis
Canfora, L. (1991) Vera vocabula rerum amisimus, in M. Pani, ed., Continuità e
trasformazioni tra Repubblica e Principato, Bari, 1981, 1038
Canter, H. V. (1925) Rhetorical Elements in the Tragedies of Seneca, Urbana IL
Carrière, J. (1977) Le choeur secondaire dans le drame grec: sur une ressource meconnue
de la scène antique, Paris
Castagna, L., ed. (1996) Nove studi sui cori tragici di Seneca, Milan
Catenacci, C. (1996) Il tiranno e leroe. Per unarcheologia del potere nella Grecia
antica, Milan
Cave, T. (1988) Recognitions. A Study in Poetics, Oxford
Caviglia, F., ed. (1979) Seneca, Il furore di Ercole, Rome
(1996) I cori dellOedipus di Seneca e linterpretazione della tragedia, in
Castagna (1996) 87103
Caws, M. A. (1985) Reading Frames in Modern Fiction, Princeton
Cazzaniga, I. (1950) La saga di Itis nella tradizione letteraria e mitografica grecoromana, vol. i: La tradizione letteraria e mitografica greco-romana da Omero a
Nonno Panopolitano, Milan and Varese
Ciappi, M. (1998) Contaminazioni fra tradizioni letterarie affini di ascendenza
tragica nel racconto ovidiano del mito di Procne e Filomela (met. VI
587666), Maia 50: 43363
Cipriani, G. (1978) Una scena dellAtreus di Accio e Seneca, Rivista di Studi Classici
26: 21021
Cixous, H. (1986) Sorties, in H. Cixous and C. Clement, eds., The Newly Born
Woman, Minneapolis (original edn, La jeune nee, Paris, 1975) 63132
(1991) Coming to Writing and Other Essays, Cambridge, MA and London
Coarelli, F. (1983), Il foro romano, vol. i, Rome
Conte, G. B. (1985) Memoria dei poeti e sistema letterario, 2nd edn, Turin (revised
English edn in C. Segal, ed., The Rhetoric of Imitation. Genre and Poetic
Memory in Virgil and other Latin Poets, Ithaca and London, 1986)
Conte, G. B. and Barchiesi, A. (1989) Imitazione e arte allusiva. Modi e funzioni
dellintertestualità, in G. Cavallo, P. Fedeli and A. Giardina, eds., Lo spazio
letterario di Roma antica, Rome, vol. i, 81114
Courtney, E. (1993) The Fragmentary Latin Poets, Oxford
Dallenbach, L. (1977) Le recit speculaire: essai sur la mise en abyme, Paris (= The
Mirror in the Text, Chicago, 1989)
Dangel, J., ed. (1995) Accius, Oeuvres (fragments), Paris
Davis, P. J. (1993) Shifting Song: the Chorus in Senecas Tragedies, Hildesheim, Zurich
and New York
De Lacy, P. (1948) Stoic views of poetry, AJPhil. 69: 24171
De Ritis, R. (1991) Il tempo in Freud , Naples
De Rosalia, A. (1981) Echi acciani in Seneca tragico, Dioniso 52: 22142
Detienne, M. (1979) Dionysus Slain, Baltimore and London (original edn, Paris,
1977)
Dingel, J. (1974) Seneca und die Dichtung, Heidelberg
Bibliography
259
Doblhofer, E. (1960) Ovidius urbanus. Eine Studie zum Humor in Ovids

Metamorphosen, Philologus 104: 6391 and 22335
Dodds, E. R., ed. (1960) Euripides, Bacchae, 2nd edn, Oxford
Dohm, H. (1964) Mageiros: die Rolle des Kochs in der griechisch-romischen Komodie,
Munich
Dover, K. J., ed. (1968) Aristophanes, Clouds, Oxford
duBois, P. (1988) Sowing the Body. Psychoanalysis and Ancient Representations of
Women, Chicago and London
(1991) Torture and Truth, New York and London
Duckworth, G. E. (1952) The Nature of Roman Comedy. A Study in Popular
Entertainment, Princeton
Dupont, F. (1985) Lacteur-roi, ou le thea tre dans la Rome antique, Paris
(1995) Les monstres de Senèque. Pour une dramaturgie de la tragedie romaine,
Paris
Edwards, C. (1993) The Politics of Immorality in Ancient Rome, Cambridge
Edwards, M. W. (1960) The example of Stoic ideas in the Aeneid , Phoenix 14:
15165
Eliade, M. (1954) The Myth of the Eternal Return, or, Cosmos and Reality, Princeton
Elsner, J. and Masters, J., eds. (1994) Reflections of Nero. Culture, History &
Representation, London
Fantham, E. (1972) Comparative Studies in Republican Latin Imagery, Toronto
ed. (1982) Senecas Troades, Princeton
Feeney, D. C. (1991) The Gods in Epic, Oxford
(1992) Si licet and fas est . . . Ovids Fasti and the Problem of Free Speech
under the Principate, in A. Powell, ed., Roman Poetry and Propaganda in the
Age of Augustus, London
(1999) Mea tempora: patterning of time in the Metamorphoses, in Hardie,
Barchiesi and Hinds (1999) 1330
Finkelberg, M. (1998) The Birth of Literary Fiction in Ancient Greece, Oxford
Fitch, J. G. (1981) Sense-pauses and relative dating in Seneca, Sophocles and
Shakespeare, AJPhil. 102: 289307
ed. (1987) Senecas Hercules Furens, Ithaca and London
Foley, H. (1980) The mask of Dionysus, TAPA 110: 10733
(1985) Ritual Irony: Poetry and Sacrifice in Euripides, Ithaca and London
Fordyce, C. J. (1977) Virgil, Aeneidos Libri VIIVIII , Oxford
Fortenbaugh, W. W. W. (1975) Aristotle on Emotion, London
Fowler, D. P. (1989) First thoughts on closure: problems and prospects, MD 22:
75122 = Fowler (2000) 23983
(1997a) On the shoulders of giants: intertextuality and classical studies, MD
39: 1334
(1997b) Second thoughts on closure, in Roberts, Dunn and Fowler (1997)
322 = Fowler (2000) 284307
(2000) Roman Constructions. Readings in Postmodern Latin, Oxford
Fraenkel, E. (1960) Elementi plautini in Plauto, 2nd edn, Florence
(1962) Beobachtungen zu Aristophanes, Rome
Frankel, H. (1945) Ovid. A Poet between Two Worlds, Berkeley and Los Angeles
260
Bibliography
Frede, M. (1986) The Stoic doctrine of the affections of the soul, in M. Schofield
and G. Striker, eds., The Norms of Nature. Studies in Hellenistic Ethics,
Cambridge and Paris, 93110
Freud, S. Works by S.F. are referred to in the text with a full title and the date of the
original German publication, and quoted according to the English Standard
Edition (SE), 24 vols., London, 195374
Frings, I. (1992) Odia fraterna als manieristisches Motiv: Betrachtungen zu Senecas
Thyest und Statius Thebais, Stuttgart
Galinsky, K. (1975) Ovids Metamorphoses. An Introduction to the Basic Aspects,
Oxford
Gallop, J. (1982) Feminism and Psychoanalysis. The Daughters Seduction, London
and Basingstoke
Giancotti, F., ed. (198889) Seneca, Tieste, 2 vols., Turin
(1989) Ignotus moritur sibi. Sul secondo canto corale del Tieste di Seneca, in
(various authors) Mnemosynum. Studi in onore di Alfredo Ghiselli, Bologna,
26191
Gigon, O. (1938) Bemerkungen zu Senecas Thyestes, Philologus 93: 17683
Ginzburg, C. (1992) Miti, emblemi, spie. Morfologia e storia, Turin
Girard, R. (1972) La violence et le sacre (= Violence and the Sacred, Baltimore,
1977)
Giua, M. A. (1975) Tiberio simulatore nella tradizione storica pretacitiana,
Athenaeum 53: 35263
Goldberg, S. M. (1995) Epic in Republican Rome, New York and Oxford
Goldhill, S. (1991) The Poets Voice. Essays on Poetics and Greek Literature, Cambridge
Goldschmidt, V. (1969) Le système stocien et lidee de temps, Paris
Gowers, E. (1993) The Loaded Table. Representations of Food in Roman Literature,
Oxford
Gredley, B. (1996) Comedy and tragedy inevitable distinctions: response to
Taplin, in Silk (1996) 20316
Grimal, P. (1975) Le theatre a` Rome, in Actes du IXe Congrès International de
lAssociation Guillaume Bude, Paris, 249305
Guastella, G. (1988) La contaminazione e il parassita: due studi su teatro e cultura
romana, Pisa
(2001) Lira e lonore. Forme della vendetta nel teatro senecano e nella sua tradizione,
Palermo
Gudeman, A., ed. (1914) Tacitus, Dialogus de oratoribus, 2nd edn, Leipzig and
Berlin
Hall, W. M. (1996) From Memories to Mental Illness. A Conceptual Journey, New
York and London
Halleran, M. (1985) Stagecraft in Euripides, London and Sydney
Hamon, P. (1977) Texte litteraire et metalangage, Poetique 31: 26184
Handley, E. W., ed. (1965) The Dyscolos of Menander, London
Hansen, E. (1934) Die Stellung der Affektreden in den Tragodien Senecas, dissertation, Berlin
Hardie, P. R. (1993) The Epic Successors of Virgil. A Study in the Dynamics of a
Tradition, Cambridge
Bibliography
261
(1997) Closure in Latin epic, in Roberts, Dunn and Fowler (1997) 13962
Hardie, P. R., Barchiesi, A. and Hinds, S., eds. (1999) Ovidian Transformations.
Essays on Ovids Metamorphoses and Its Reception, Cambridge
Harrison, S. J., ed. (1991) Vergil, Aeneid 10, Oxford
Haupt, M. (187576) Opuscula, Leipzig (= Hildesheim, 1967)
Heinze, R. (1928) Virgils epische Technik, 3rd edn, Leipzig (= Virgils Epic Technique,
Berkeley and Los Angeles, 1993)
Henderson, J. (1987) Lucan: the word at war, Ramus 16: 12264 (= Fighting
for Rome: Poets and Caesars, History and Civil War, Cambridge, 1998, 165
211)
Herington, C. J. (1966) Senecan tragedy, Arion 5: 42271
(1982) The Younger Seneca, in E. Kenney and W. Clausen, eds., The Cambridge
History of Classical Literature, vol. ii: Latin Literature, Cambridge, 51132
Herrmann, L. (1924) Le thea` tre de Senèque, Paris
Hershkowitz, D. (1998) The Madness of Epic: Reading Insanity from Homer to Statius,
Oxford
Hiltbrunner, O. (1985) Seneca als Tragodiendichter in der Forschung von 1965 bis
1975, ANRW ii.32.2, 9691051
Hinds, S. (1987) The Metamorphosis of Persephone. Ovid and the Self-Conscious
Muse, Cambridge
(1998) Allusion and Intertext: Dynamics of Appropriation in Roman Poetry,
Cambridge
Hine, H. M. (1981) The structure of Senecas Thyestes, Papers of the Liverpool
Latin Seminar 3: 25975
Hornby, R. (1986) Drama, Metadrama, and Perception, London and Toronto
Huet, V. (1994) Les images sacrificielles a` Rome, Paris
Hunter, R. L. (1985) The New Comedy of Greece and Rome, Cambridge
Hutcheon, L. (1984) Narcissistic Narrative. The Metafictional Paradox, 2nd edn,
London and New York
Irigaray, L. (1985) Speculum of the Other Woman, Ithaca and London (= Speculum
de lautre femme, Paris, 1974)
Irwin, J. T. (1975) Doubling and Incest, Repetition and Revenge: a Speculative Reading
of Faulkner, Baltimore and London
Jacobi, R. (1988) Der Einfluss Ovids auf den Tragiker Seneca, Berlin and New York
Jal, P. (1963) La guerre civile a` Rome. Etude litteraire et morale, Paris
James, H. (1997) Shakespeares Troy. Drama, Politics and the Translation of Empire,
Cambridge
Janan, M. (1994) When the Lamp Is Shattered: Desire and Narrative in Catullus,
Carbondale, IL
Janko, R. (1992) The Iliad: a Commentary, vol. iv, books 1316, Cambridge
Jocelyn, H. D., ed. (1967) The Tragedies of Ennius, Cambridge
(1995) Poeta and vates: concerning the nomenclature of the composer of
verses in republican and early imperial Rome, in Studia classica Iohanni Tarditi
oblata, Milan, vol. i, 1950
Johnson, W. R. (1987) Momentary Monsters. Lucan and His Heroes, Ithaca and
London
262
Bibliography
Joplin, P. K. (1984) The voice of the shuttle is ours, Stanford Literary Review 1:
2553 (= L. A. Higgins and B. R. Silver, eds., Rape and Representation, New
York, 1991, 3564)
Kern, S. (1983) The Culture of Time and Space 18801918, Cambridge, MA
Kerrigan, J. (1996) Revenge Tragedy. Aeschylus to Armageddon, Oxford
Kilgour, M. (1990) From Communion to Cannibalism: an Anatomy of Metaphors of
Incorporation, Princeton
Kinsey, T. E. (1979) The meaning of interea in Virgils Aeneid , Glotta 57: 25965
Knoche, U. (1941) Senecas Atreus. Ein Beispiel, Die Antike 17: 6076 (= Lefèvre
(1972) 47789)
Knox, B. (1964) The Heroic Temper, Berkeley and Los Angeles
Kraus, C. (1991) OO MEN ET AAIO: story and story-telling in
Sophocles Trachiniae, TAPA 121: 7598
Kremer, G. (1971) Die Struktur des Tragodienschlusses, in W. Jens, ed., Die
Bauformen der griechischen Tragodie, Munich, 11741
Lada, I. (1996) Emotion and meaning in tragic performance, in Silk (1996) 397413
Lana, I. (195859) LAtreo di Accio e la leggenda di Atreo e Tieste nel teatro tragico
Romano, Atti dellAccademia delle Scienze di Torino, Classe di Scienze morali,
storiche e filologiche 93: 293385
Lanza, D. (1977) Il tiranno e il suo pubblico, Turin
(1981) Lo spettacolo della parola: Riflessioni sulla testualità drammatica di
Seneca, Dioniso 52: 46376
La Penna, A. (1963) Tre poesie espressionistiche di Orazio (e una meno espressionistica), Belfagor 18: 181193
(1971) Estasi dionisiaca e poetica callimachea, in (various authors) Studi De
Falco, Napoli, 22937
(1979) Fra teatro, poesia e politica romana, Turin
Lapidge, M. (1978) Stoic Cosmology, in J. M. Rist, ed., The Stoics, 2nd edn,
Berkeley, 16185
Lefèvre, E. (1968) Review of Zwierlein (1966), Gnomon 40: 7829
ed. (1972) Senecas Tragodien, Darmstadt
(1976) Der Thyestes des Lucius Varius Rufus. Zehn Uberlegungen

zu seiner
Rekonstruktion, AAWM 9
(1985) Die philosophische Bedeutung der Seneca-Tragodie am Beispiel des
Thyestes , ANRW ii.32.2, 126383
(1990) Die politische Bedeutung von Senecas Phaedra, Wiener Studien 103:
10922
Leigh, M. (1995) Review of Boyle (1994), Bryn Mawr Classical Review 6.6: 4702
(1996) Varius Rufus, Thyestes and the appetites of Antony, PCPS 42: 17197
Lemmers, J. (1931) Die Doppel- und Halbchore in der antiken Tragodie, dissertation, Paderborn
Leo F., ed. (187879) Seneca, Tragoediae, Berlin
(1897) Die Composition der Chorlieder Senecas, Rh. Mus. 52: 50918
(1908) Der Monolog im Drama: ein Beitrag zur griechisch- romischen Poetik, Berlin
(1912) Plautinische Forschungen, 2nd edn, Berlin
Bibliography
263
Leonard, M. (1999) Irigarays cave: feminist theory and the politics of French
classicism, Ramus 28: 15268
Lesky, A. (192223) Die griechische Pelopidendramen und Senecas Thyestes,
Wiener Studien 43: 17298 (= Gesammelte Schriften, Bern and Munich, 1966,
52040)
Lindskog, C. (1897) Studien zum Antiken Drama, Lund
Littlewood, C. (1997) Senecas Thyestes: the tragedy with no women?, MD 38:
5786
Long, A. A. (1985) Stoics on world-conflagration and everlasting recurrence,
Southern Journal of Philosophy 23 suppl.: 1338
Luce, T. J. (1991) Tacitus on historys highest function: praecipuum munus
annalium (Ann.3.65), ANRW ii.33.4, 290427
Ludwig, W. (1965) Struktur und Einheit der Metamorphosen Ovids, Berlin
Malaspina, E., ed. (2001) Seneca, De clementia libri duo, Alessandria
Mansfeld, J. (1979) Providence and destruction of universe in early Stoic thought,
in M. J. Vermaseren, ed., Studies in Hellenistic Religions, Leiden, 12988
Mantovanelli, P. (1984) La metafora del Tieste. Il nodo sadomasochistico nella tragedia
senecana del potere tirannico, Verona
Marchesi, C. (1908) Le fonti e la composizione del Thyestes di L. Anneo Seneca,
Riv. Fil. 36: 70104 (= Scritti minori di filologia e di letteratura, Florence,
1978, 493527)
Marienstras, R. (1985) New Perspectives on the Shakespearean World , Cambridge
Marshall, C. W. (2000) Location! Location! Location! Choral absence and theatrical space in the Troades, in G. W. M. Harrison, ed., Seneca in Performance,
London, 2751
Marti, B. L. (1945a) Senecas tragedies: a new interpretation, TAPA 76: 21645
(1945b) The Meaning of the Pharsalia, AJPhil. 66: 35276
Masters, J. (1992) Poetry and Civil War in Lucans Bellum Civile, Cambridge
Mastronarde, D. J. (1970) Senecas Oedipus: the drama in the word, TAPA 101:
291315
ed. (1994) Euripides, Phoenissae, Cambridge
Matte Blanco, I. (1975) The Unconscious as Infinite Sets. An Essay in Bi-Logic, London
(1988) Thinking, Feeling, and Being. Clinical Reflections on the Fundamental
Antinomy of Human Beings and World , London
Mayor, J. E. B. (1879) On licentia poetica, Journal of Philology 8: 2602
Mazzoli, G. (1970) Seneca e la poesia, Milan
(198687) Funzioni e strategie dei cori in Seneca tragico, Quaderni di cultura e
tradizione classica 45: 99112
Meltzer, G. (1988) Dark wit and black humor in Senecas Thyestes, TAPA 118:
30930
Mendelsohn, D. (1992),
: dithyrambic language and Dionysiac
cult, CJ 87: 10524
Merkelbach, R. and West, M. L., eds. (1967) Fragmenta Hesiodea, Oxford
Michel, A. (1969) Rhetorique, tragedie, philosophie: Senèque et le sublime,
Giornale Italiano di Filologia 21: 24557
264
Bibliography
Miola, R. S. (1983) Shakespeares Rome, Cambridge

Moatti, C. (1997) La raison de Rome. Naissance de lesprit critique a` la fin de la
Republique (IIe-Ier siècle avant Jesus-Christ), Paris
Monteleone, C. (1980) I modelli di Seneca nel prologo del Thyestes, Giornale
Italiano di Filologia 32: 7782
Moretti, G. (1995) Acutum dicendi genus. Brevità, oscurità, sottigliezze e paradossi
nelle tradizioni retoriche degli Stoici, Bologna
Most, G. (1992) disiecti membra poetae: the rhetoric of dismemberment in Neronian poetry, in R. Hexter and D. Selden, eds., Innovations of Antiquity, New
York, 391419
Moulton, C. (1977) Similes in the Homeric Poems, Gottingen
Murray, P. (1996) Plato on Poetry, Cambridge
Narducci, E. (1979) La provvidenza crudele. Lucano e la distruzione dei miti augustei,
Pisa
die Musik IV. Buch, Naples
Neubecker, A. J., ed. (1986) Philodemus, Uber
Newlands, C. (1995) Playing with Time: Ovid and the Fasti, Ithaca and London
Newman, J. K. (1967) The Concept of vates in Augustan Poetry, Brussels
Nilsson, M. P. (1932) Sonnenkalender u. Sonnenreligion, ARW 30: 14173
(= Opuscula selecta linguis Anglica, Francogallica, Germanica conscripta, Lund,
195160, 462504)
Nisbet, R. G. M. (1990) The dating of Senecas tragedies, with special reference
to Thyestes, Papers of the Leeds International Latin Seminar 6: 95114
Nisbet, R. G. M. and Hubbard, M. (1970) A Commentary on Horace Odes, Book
I , Oxford
Norden, E., ed. and comm. (1927) Virgil, Aeneis Buch VI , 3rd edn, Leipzig
Nussbaum, M. (1993) Poetry and the passions: two Stoic views, in Brunschwig
and Nussbaum (1993) 97149
(1994) The Therapy of Desire: Theory and Practice in Hellenistic Ethics, Princeton
OHiggins, D. (1988) Lucan as vates, Cl. Ant. 7: 20826
Orlando, F. (1971) Lettura freudiana della Phèdre, Turin
(1973) Per una teoria freudiana della letteratura, Turin
(1978) Toward a Freudian Theory of Literature. With an analysis of Racines Phèdre,
Baltimore and London (English translation of 1971 and 1973)
(1979) Lettura freudiana del Misanthrope, Turin
(1982) Illuminismo e retorica freudiana, Turin
(1993) Gli oggetti desueti nelle immagini della letteratura. Rovine, reliquie, rarità,
robaccia, luoghi inabitati e tesori nascosti, Turin
Owen, W. H. (1970) Time and event in Senecas Troades, Wiener Studien 83:
11837
Palmieri, N. (1989) Alia temptanda est via: allusività e innovazione drammatica
nellEdipo di Seneca, MD 23: 17589
Paratore, E. (1956) La poesia nellOedipus di Seneca, Giornale Italiano di Filologia
9: 97132
Parker, R. (1983) Miasma. Pollution and Purification in Early Greek Religion,
Oxford
Bibliography
265
Pasiani P. (1967) Attonitus nelle tragedie di Seneca, in (various authors) Studi sulla
lingua poetica latina, Rome, 113-36 = A. Traina, ed., Seneca. Letture critiche,
Milan, 1976, 194207
Pasquali, G. (1920) Orazio Lirico, Florence
Pedrick, V. and Rabinowitz, N. S. (1986) Introduction, Arethusa 19: 10514
Petrone, G. (1984) La scrittura tragica dellirrazionale: note di lettura al teatro di
Seneca, Palermo
Pfister, M. (1977) Das Drama, Munich (= The Theory and Analysis of Drama,
Cambridge 1988)
Picone, G. (1976) Il significato politico di alcuni anacronismi nel Thyestes di
Seneca, Pan 3: 617
(1984) La fabula e il regno: studi sul Thyestes di Seneca, Palermo
Plass, P. (1988) Wit and the Writing of History, Madison
Pratt, N. T. (1948) The Stoic base of Senecan Drama, TAPA 79: 111
(1983) Senecas Drama, Chapel Hill
Price, S. R. F. (1980) Between man and god: sacrifice in the Roman imperial cult,
JRS 70: 2843
Prince, G. (1977) Remarques sur les signes metanarratifs, Degres 5: e1-e10
Putnam, M. C. J. (1995) Virgils Aeneid. Interpretation and Influence, Chapel Hill
and London
Quint, D. (1993) Epic and Empire, Princeton
Regenbogen, O. (192728) Schmerz und Tod in den Tragodien Senecas, Vortrage
der Bibliothek Warburg 7: 167218 (= Kleine Schriften, ed. F. Dirlmeier,
Munich, 1961, 40962)
Reinmuth, O. W. (1933) Vergils use of interea, AJPhil. 54: 32339
Rengakos, A. (1995) Zeit und Gleichzeitigkeit in den homerischen Epen, Antike
und Abendland 41: 133
Richlin, A. (1992) Reading Ovids rape, in A. Richlin, ed., Pornography and
Representation in Greece & Rome, New York and Oxford, 15879
Richter, G., ed. (1902) Seneca, Tragoediae, Lipsiae
Rimell, E. V. (2002) Petronius and the Anatomy of Fiction, Cambridge
Roberts, D. H. (1989) Different Stories: Sophoclean narrative(s) in the Philoctetes,
TAPA 119: 16176
Roberts, D. H., Dunn, F. M. and Fowler, D., eds. (1997) Classical Closure: Reading
the End in Greek and Latin Literature, Princeton
Robin, D. (1993) Film theory and the gendered voice in Seneca, in N. S.
Rabinowitz and A. Richlin, eds., Feminist Theory and the Classics, New York
and London, 10221
Rosati, G. (1983) Narciso e Pigmalione. Illusione e spettacolo nelle Metamorfosi di
Ovidio, Florence
Rosenmeyer, T. G. (1989) Senecan Drama and Stoic Cosmology, Berkeley
Rossi, E. (1989) Una metafora presa alla lettera: le membra lacerate della famiglia.
Tieste di Seneca e i rifacimenti moderni, Pisa
Rozelaar, M. (1976) Seneca. Eine Gesamtdarstellung, Amsterdam
Russell, D. A., ed. (1964) Longinus, On the Sublime, Oxford
266
Bibliography
Salvadore, M., ed. (1999) M. Terenti Varronis fragmenta omnia quae extant. Pars 1,
Hildesheim
Schawaller, D. (1987) Semantische Wortspiele in Ovids Metamorphosen und
Heroides, GB 14: 199214
Scheid, J. (1988) La spartizione sacrificale a Roma, in C. Grottanelli and N. F.
Parise, eds., Sacrificio e società nel mondo antico, Rome and Bari, 26792
Scheid, J. and Svenbro, J. (1996) The Craft of Zeus: Myths of Weaving and Fabric,
Cambridge, MA
Schetter, W. (1965) Sulla struttura delle Troiane di Seneca, Riv. Fil. 93: 396429
(= Zum Aufbau von Senecas Troerinnen, in Lefèvre (1972) 23072)
Schiesaro, A. (1985) Il locus horridus nelle Metamorfosi di Apuleio, Maia 37: 21123
(1993a) Il destinatario discreto. Funzioni didascaliche e progetto culturale nelle
Georgiche, MD 31: 12947
(1993b) Review of J. OHara, Death and the Optimistic Prophecy in Virgils Aeneid
(Princeton, 1990), C Phil. 88: 25865
(1996) Felicità, passione e libertà nel De vita beata di Seneca, in Seneca, Sulla
felicità (trans. A. Schiesaro), Milan
(1997) The boundaries of knowledge in Virgils Georgics, in T. N. Habinek and
A. Schiesaro, eds., The Roman Cultural Revolution, Cambridge, 6389
Schmeling, M. (1982) Metathea tre et intertexte. Aspects du thea tre dans le thea tre,
Paris
Schmitz, C. (1993) Die kosmische Dimension in den Tragodien Senecas, Berlin and
New York
Seaford, R. A. S., ed. (1984) Euripides, Cyclops, Oxford
(1987) The Tragic Wedding, JHS 107: 10630
Segal, C. (1982) Dionysiac Poetics and Euripides Bacchae, Princeton (expanded edn,
1997)
(1986) Language and Desire in Senecas Phaedra, Princeton
(1994) Philomelas web and the pleasures of the text, in I. J. F. de Jong and
J. P. Sullivan, eds., Modern Critical Theory and Classical Literature, Leiden,
New York and Cologne, 25780
Seidensticker, B. (1978) Comic elements in Euripides Bacchae, AJPhil. 99: 30320
(1985) Maius solito: Senecas Thyestes und die tragoedia rhetorica, Antike und
Abendland 31: 11636
Sharrock, A. (1994) Seduction and Repetition in Ovids Ars Amatoria, Oxford
Shelton, J.-A. (1975) Problems of time in Senecas Hercules Furens and Thyestes,
California Studies in Classical Antiquity 8: 25769
(1978) Senecas Hercules Furens. Theme, Structure and Style, Gottingen
Sifakis, G. (1967) Studies in the History of Hellenistic Drama, London
Silk, M. S., ed. (1996) Tragedy and the Tragic. Greek Theatre and Beyond , Oxford
(2000) Aristophanes and the Definition of Comedy, Oxford
Slater, N.W. (1985) Plautus in Performance. The Theatre of the Mind , Princeton
Smolenaars, J. J. L. (1998) The Vergilian background of Senecas Thyestes 64182,
Vergilius 44: 5165
Staley, G. A. (1981) Senecas Thyestes: quantum mali habeat ira, GB 10: 23346
Bibliography
267
Stam, R. (1992) Reflexivity in Film and Literature. From Don Quixote to Jean-Luc
Godard, 2nd edn, New York
Stanley, K. (1993) The Shield of Homer. Narrative Structure in the Iliad , Princeton
Steidle, W. (1941) Zu Senecas Troerinnen, Philologus 94: 26684
(194344) Bemerkungen zu Senecas Tragodien, Philologus 96: 25064
(= Lefèvre (1972) 4909, 28691)
Sutton, D. F. (1986) Seneca on the Stage, Leiden
Taplin, O. (1977) The Stagecraft of Aeschylus. The Dramatic Use of Exits and
Entrances in Greek Tragedy, Oxford
(1996) Comedy and the tragic, in Silk (1996) 188202
Tarrant, R. J., ed. (1976) Seneca, Agamemnon, Cambridge
(1978) Senecan drama and its antecedent, Harv. Stud. 82: 21363
ed. (1985) Seneca, Thyestes, Atlanta
Tieleman, T. L. (1992) Galen and Chrysippus. Argument and Refutation in the De
placitis Books IIIII , Utrecht
Tigerstedt, E. N. (1970) Furor poeticus, Journal of the History of Ideas 31: 16378
Timpanaro, S. (1981) Un nuovo commento allHercules Furens di Seneca nel quadro
della critica recente, Atene e Roma n.s. 26: 11341
Tochterle, K., ed. (1994) Seneca, Oedipus, Heidelberg
Too, Y. L. (1994) Educating Nero: a reading of Senecas Moral Epistles, in Elsner
and Masters (1994) 21124
Traina, A. (1981) Seneca, Thyest. 713s. Mactet sibi o sibi dubitat? Un recupero
esegetico, Maia 33: 1513 (= Poeti latini (e neolatini) III , Bologna, 1989,
16770)
(1987) Lo stile drammatico del filosofo Seneca, 4th edn, Bologna
(1990) Turnus, in Enciclopedia Virgiliana, Rome, vol. v. 1, 32436
Tricomi, A. H. (1974) The aesthetics of mutilation in Titus Andronicus ,
Shakespeare Survey 27: 1129
Wagenvoort, H. (1933) Ad Senecae Herculum Furentem, Mnemosyne 60: 1708
Waith, E. M. (1957) The metamorphosis of violence in Titus Andronicus,
Shakespeare Survey 10: 3949
Walker, B. (1969) Review of Zwierlein (1966), C Phil. 64: 1837
Wallace-Hadrill, A. (1987) Time for Augustus: Ovid, Augustus, and the Fasti , in
M. Whitby, P. Hardie and M. Whitby, eds., Homo Viator. Classical Essays for
John Bramble, Bristol, 22130
Webb, R. (1997) Imagination and the arousal of the emotions in Greco-Roman
rhetoric, in S. Morton Braund and C. Gill, eds., The Passions in Roman
Thought and Literature, Cambridge, 11227
Weinstock, S. (1948) A new Greek calendar and festivals of the sun, JRS 38: 3742
Whitman, C. H. and Scodel, R. (1981) Sequence and simultaneity in Iliad N , ,
and O, Harv. Stud. 85: 115
Wilamowitz-Moellendorff, U. von, ed. (1909) Euripides, Herakles, 2nd edn, Berlin
(1919) Griechische Tragodien, 4 vols., Berlin
Willbern, D. (1978) Rape and revenge in Titus Andronicus, English Literary
Renaissance 8: 15972
268
Bibliography
Williams, G. (1978) Change and Decline. Roman Literature in the Early Empire,
Berkeley, Los Angeles and London
Willink, C.W. (1983) Prodikos, Meteorosophists and the Tantalos Paradigm,
CQ 33: 2533
ed. (1986) Euripides, Orestes, Oxford
Wills, J. (1996) Repetition in Latin Poetry. Figures of Allusion, Oxford
Woodman, T. (1992) Neros alien capital: Tacitus as paradoxographer (Annals
15.367), in Woodman, T. and Powell, J., eds., Author and Audience in Latin
Literature, Cambridge, 17388
Zanker, P. (1988) The Power of Images in the Age of Augustus, Ann Arbor
Zecchini, G. (1986) La Tabula Siarensis e la dissimulatio di Tiberio, ZPE 66: 2329
Zeitlin, F. I. (1965) The motif of corrupted sacrifice in Aeschylus Oresteia, TAPA
96: 463508
(1980) The closet of masks: role-playing and myth-making in the Orestes of
Euripides, Ramus 9: 5177
(1990) Thebes: theatre of self and society in Athenian drama, in J. J. Winkler
and F. I. Zeitlin, eds., Nothing To Do with Dionysos. Athenian Drama in Its
Social Context, Princeton, 13067
(1996) Playing the Other. Gender and Society in Classical Greek Literature, Chicago
and London
Zielinski, T. (1901) Die Behandlung gleichzeitiger Ereignisse im antiken Epos,
Philologus Supplementband 8: 40749
Zissos, A. and Gildenhard, I. (1999) Problems of time in Metamorphoses 2, in
Hardie, Barchiesi and Hinds (1999) 3147
Zumwalt, N. (1977) Fama Subversa: theme and structure in Ovid Metamorphoses
12, California Studies in Classical Antiquity 10: 20922
Zwierlein, O. (1966) Die Rezitationsdramen Senecas, Meisenheim am Glan
(1976) Versinterpolationen und Korruptelen in den Tragodien Senecas, WJA
2: 181217
(1986) Kritischer Kommentar zu den Tragodien Senecas, Stuttgart
ACCIUS (Ribbeck2 )
198201: 132
199202: 142
205: 101
2068: 102
2202: 85
226: 196
6345: 81
AESCHYLUS
Agamemnon
9445: 151
10967: 85
11923: 141
15878: 142
15981601: 136
166273: 67
Prometheus
187: 30
Seven at Thebes
87170: 166
Supplices
225: 101
ARISTOTLE
Poetics
1449b: 245
1449b 248: 245
1449b 26: 188
1461b: 245
1462a: 245
1462b: 245
CALPURNIUS
Eclogue
3.5960: 132
CATULLUS
61.214: 200
85: 51
CICERO
De natura deorum
3.689: 99
De officiis
1.28.97: 161
De optimo genere oratorum
1: 136
De oratore
2.194: 22
3.219: 132
De republica
1.65: 118
Paradoxa stoicorum
25: 39
Philippicae
1.36: 116
Tusculanae
1.64: 22
4.38: 132
CLEANTHES
Hymn to Zeus
SVF 1.486: 23, 128
AUCTOR AD HERENNIUM
4.41: 131
DIODORUS
1.1.5: 44
15.1.1: 44
BRECHT
The Modern Theatre Is the Epic
Theatre: 2478
Theatre for Pleasure or Theatre for
Instruction: 2467
ENNIUS
Fragmenta scenica
73 Ribbeck2 : 200
Annales
3324 Skutsch: 100
269
270
EURIPIDES
Antiopa
177227 Nauck2 : 9
Bacchae
1: 56
196: 134
231: 135
269: 134
326: 134
332: 134
436: 135
451: 135
47980: 134
491: 134
508: 134
58692: 138
6559: 134
656: 134
7313: 135
842: 138
8456: 151
848: 135
861: 138
977: 135
98991: 135
1019: 135
1021: 135
1113: 135
11201: 135
118991: 135
12812: 134
1345: 135
Danae
324 Nauck2 : 234
Electra
699728: 141
Hecuba
1: 56
524: 49
Hercules furens
82274: 30
84354: 30
857: 30
85873: 31
8656: 31
8723: 31
897: 31
Hippolytus
513: 48
5871: 201
877: 221
Ion
5: 56
7880: 48
Kressai
4669 Nauck2 : 141

Medea
3556: 212
376: 17
37680: 17
8078: 17
1384: 210
Phoenissae
50910: 130
553: 53, 130
554: 130
Troades
1: 56
1133: 200
117881: 199
HESIOD
Catalogus mulierum
182 MerkelbachWest: 9
Theogonia
278: 24
HOMER
Iliad
3.1267: 1
3.236: 124
5.750: 226
6.42930: 197
12.292300: 124
17.617: 124
17.5402: 124
20.615: 226
21.388: 226
22.18893: 100
22.396404: 200
22.4013: 200
22.402: 124
Odyssey
4.3359 = 17.12630: 124
11.2234: 230
11.2605: 9
14.192: 115
HOMERIC HYMN TO DIONYSUS
7.44: 123
HORACE
Ars poetica
69a: 120
9b13: 120
18990: 65
445: 132
Carmina
1.1.36: 59, 116
1.1.356: 59, 116
1.16.15: 124
1.18.711: 76
2.20.14: 116
3.1.23: 9

3.19.14: 51
3.25.13: 51
3.25.1718: 51
Epistles
2.1.67: 131
2.1.21113: 10
2.2.123: 132
Epodes
7: 35
Satires
2.1.10: 51
HYGINUS
Fabulae
87: 97
88: 97
ISIDORUS
Origines
8.7.3: 121
LIVY
Ab Urbe condita libri
21.45.8: 91
22.57.6: 85
39.15.9: 51
[LONGINUS]
1.3: 128
1.4: 127, 128, 231
7.3: 23
8.1: 128
8.2: 115
8.4: 128
9.2: 128
9.6: 226
10.3: 129
10.6: 128
12.2: 130
13.2: 129
14.3: 129
15.1: 131
15.9: 128
18.12: 131
34.4: 132
LUCAN
Bellum Civile
1.2: 98
1.95: 35
1.20412: 124
7.5526: 41
7.556: 41
10.1847: 218
LUCRETIUS
1.79: 98
1.1023: 194
1.4048: 100
5.878924: 120
MANILIUS
Astronomica
5.462: 136
NAEVIUS
57 Ribbeck2 : 123
OVID
Amores
1.1.1: 716
2.2.44: 40
3.7.51: 40
Ars amatoria
3.61132: 77
3.61718: 77
Fasti
1.478: 39
3.4712: 27
6.56: 29
Heroides
12.212: 81
Ibis
187: 28
Medea
2 Ribbeck2 : 51
Metamorphoses
1.64950: 73
2.641: 30
4.767: 151
4.157: 198
4.173: 28
4.4845: 9
5.1649: 123
6.412674: 70
6.424: 71
6.469: 73
6.46974: 79
6.4734: 73
6.498: 79
6.51926: 734
6.524: 74
6.5378: 80
6.5448: 74
6.548: 74
6.556: 74
6.565: 73
6.571: 79
6.574: 74
6.575: 74
6.5769: 745
6.5816: 75, 237
6.582: 75
6.585: 76
271
272
6.5856: 76
6.586: 81
6.587: 133
6.588: 76
6.6012: 179, 180
6.611: 77
6.61113: 801
6.61819: 81
6.6212: 200
6.635: 81
6.6367: 80, 123
6.655: 82
6.65960: 82
6.6635: 8990
7.6971: 151
7.767: 151
7.173: 28
12.5961: 9
14.56872: 146
15.86: 123
15.1309: 137
Tristia
4.1.413: 29

Pseudolus
524: 136
584: 136
1063: 136
12434: 136
1295: 136
1301: 136
Stichus
73972: 136
Trinummus
12: 29
13: 32
706: 116
PLINY
Panegyricus
3.1: 115
3.4: 115
PACUVIUS
120a Ribbeck2 : 9
PLUTARCH
Quomodo adulescens poetas audire
debeat
16d-e: 230
16e: 230
17d: 2301
18ab: 231
18d: 231
PHILODEMUS
De musica 28.114 Neubecker:
229
PROPERTIUS
2.34.66: 81
3.1.3: 9
PLATO
Gorgias
466c: 121
Phaedrus
245a: 22
Respublica
439e440a: 242
533e534e: 119
565d566a: 126
571b: 118
571cd: 94, 118
572b: 118
577b: 119
579bc: 119
603c605c: 119
605c607a: 119
QUINTILIAN
Institutio oratoria
3.8.45: 142
9.3.54: 131
9.4.137: 131
9.4.142: 132
PLAUTUS
Amphitruo
69: 116
Bacchides
925: 136
92578: 136
Poenulus
37: 116
SALLUST
Bellum Catilinae
52.11: 115
SAPPHO
31 Voigt: 129
SCHILLER
Letter to Goethe (26.12.1797): 2456
SCHOLIA TO ILIAD
3.1267: 1
SENECA
Agamemnon
12: 28
31: 88
334: 88
346: 2034, 215
38: 180

434: 181
44: 164
46: 181
536: 181
778: 164
241: 209
656: 185
71215: 203
725: 203
72831: 207
7528: 202
754: 203
81415: 216
8813: 137
8814: 137
901: 179
9678: 209
10967: 85
Consolatio ad Helviam
8.3: 217
Consolatio ad Marciam
1.4: 44
19.4: 25
De beneficiis
1.3.10: 24
1.4.5: 24
4.7.2: 217
5.2.4: 29
De brevitate vitae
2.2: 24, 53
16.5: 24, 216
De clementia
1.1.4: 160
1.5.4: 162
1.9.4: 153
1.12.34: 15960
1.15.5: 159
1.19.8: 160
2.1.4: 160
De ira
1.10.2: 50
1.20.4: 161
2.2.1: 232
2.2.36: 232
2.2.5: 233
2.3.1: 232
2.16.1: 124
2.35.4: 245
2.35.6: 53
De providentia
5.8: 216
De tranquillitate animi
1.14: 22, 534
17: 53
17.5: 22
17.7: 22
17.8: 22
17.10: 22
17.1011: 22, 53
17.11: 23, 24
De vita beata
2.2: 53
9.2: 53
15.5: 29
15.6: 29
26.6: 24
Epistulae ad Lucilium
7.2: 233
7.5: 234
29.1: 253
41.3: 86, 127
41.6: 127
47.10: 143
75.13: 147
79.6: 52
108.7: 23, 24
108.10: 229
108.1112: 234
108.26: 53
115.12: 233
115.15: 234
122.7: 215
122.8: 215
122.18: 214, 215
Hercules Furens
5: 215
618: 215
246: 21516
306: 183
4751: 185
4752: 1845
512: 185
64: 184
668: 184
67: 184
74: 184
757: 183
868: 183
968: 9
1005: 1834
2547: 185
520: 185
5203: 1856
5213: 186
523: 186
6906: 9
81317: 186
81321: 185
8989: 92
926: 185
273
274
1035: 179
1039: 92
Medea
12: 17
1317: 17
256: 90
2831: 212
40: 17
4552: 18
47: 17
523: 18, 43
166: 213
171: 18, 213
175: 212
176: 17
1923: 18
2456: 209
2723: 209
292: 164, 212
331: 213
4017: 212
414: 212
482: 209
489: 209
5357: 146
75964: 210
910: 18, 213
928: 209, 210
984: 105, 210
101213: 210
1017: 212
101820: 212
1019: 179
1021: 213
1024: 213
Naturales quaestiones
3.2930: 216
6.324: 29
Oedipus
15: 10
257: 10
826: 10
202: 11
211: 11
21314: 11
216: 11
230: 12
2338: 12
2368: 204
291402: 92
302: 204
303: 12
303402: 204
349: 204
36671: 204

3735: 205
390: 12, 226
3904: 12
3907: 226
392: 12, 226
4012: 227
41822: 227
4246: 123
4578: 123
458: 123
4868: 227
50929: 11
509658: 8, 54
530: 89
5301: 8
53068: 11
530658: 92
548: 8
550: 89
551: 12
5515: 12
552: 8
553: 12
554: 12
555: 12
561: 227
5612: 11, 87
5613: 8
5667: 227
567: 227
5678: 8
5713: 9
572: 227
57281: 89
573: 2, 12
576: 9, 87
5826: 9
5902: 9
5924: 9
5956: 9
61118: 9
61618: 227
626: 9, 11
66970: 19
7667: 11
768: 12
8648: 196
86870: 88, 204
9427: 2056
971: 12
977: 12
998: 179
999: 12
Phaedra
1712: 88

2623: 158
26773: 158
36086: 158
40630: 158
605: 29
6468: 199
896: 221
9913: 39
1180: 76
12389: 197
1249: 201
Phoenissae
319: 29
4079: 29
Thyestes
14: 171
16: 267
1121: 49
4: 28
911: 56
13: 28
23: 50
239: 38
245: 47
25: 165
256: 83
26: 38
27: 38
28: 38
289: 165
2930: 38
39: 38
46: 38
47: 38
478: 38
523: 43
5267: 38
547: 31
56: 38, 51
567: 52, 143, 165, 179
57: 38
589: 72
62: 38
623: 28
656: 38, 45, 60, 236
657: 178
66: 48
6883a: 178
701: 165
7880: 48
823: 165
83: 33, 178
836: 32
84: 33
86b101: 178
867: 27, 192

8990: 165
8995: 39
905: 28
92: 39
95: 28
96: 29
9699: 29, 51
96100: 28
989: 34
96101: 178
100: 29, 47, 149
1004: 1789
101: 29
1014: 178
102: 66
1034: 87, 138, 179
1037: 33
105: 66
1056: 48, 178
1057: 179
107: 33
1078: 87, 205
115: 204
1201: 172, 180
1226: 164
12275: 49
1279: 164
1325: 164
133: 164
1335: 165
135: 165
138: 165
13940: 165
1501: 165
159: 113, 165
167: 113, 165
1745: 166
176: 17, 131
1768: 46
17680: 131, 140
176204: 46, 66, 154, 159
176335: 49
17880: 80, 131
179: 93
192: 46, 131
1923: 18, 43, 46, 129, 155
193: 46
1936: 141
1956: 144
199200: 106
2014: 141
202: 143
203: 46, 50
2045: 154, 159, 161
275
276
2057: 161
20510: 157
206: 161
20710: 159
209: 161
21112: 161
212: 143, 151, 157
213: 155, 159
21517: 159, 160
216: 155
217: 156
219: 155
220: 81
2204: 1012
2214: 147
2224: 80
23941: 102
240: 72
244: 17, 91
245: 155
246: 155
2468: 96
24854: 50
249: 131
2504: 30, 46
2523: 143
2524: 34, 179
2534: 143
254: 502, 143
2558: 156
256: 130
257: 17, 81
2602: 51, 108, 130
2612: 51
2656: 50, 51, 98, 129
267: 130, 143
2678: 98
26777: 52
26778: 134
26786: 156
268: 50, 52, 121
26970: 81
2725: 130
2745: 143
2756: 180
279: 59
27980: 179
27981: 142
2812: 60, 81, 181, 236
2813: 131
285: 126
2856: 144
2869: 156
287: 156
2889: 112, 157

294: 107, 157
2945: 106, 156
295: 107, 157
299302: 113
3025: 106
303: 107
3057: 157
3067: 106
312: 175
31416: 1412
315: 114
316: 175
318: 157
320: 175
321: 175
32530: 102
32930: 103
3313: 102
3323: 175
333: 157
3345: 157, 175
335: 35, 159
336403: 49
3368: 175
339: 175
33941: 145
340: 145
3446: 138
3447: 175
350: 175
3537: 175
3658: 59
386: 35
396: 153
4047: 148
40420: 59
40490: 47, 49, 56, 64
4067: 152
407: 56, 57
40910: 116
410: 59
411: 162
41220: 149
41416: 57
416: 108
41820: 108
41920: 162
4212: 132
4238: 149
434: 110
4347: 108
4349: 149
436: 47
440: 47
446: 150

44670: 108
44951: 113
459: 204
469: 57
47682: 109
487: 56, 106
4879: 149
4889: 47
489: 29, 56, 149, 150
48990: 47, 58
491: 57, 99
4913: 135
491507: 49, 56, 64, 149
493: 57
496505: 99100
497505: 55
498: 79
504: 96
5045: 55, 100
505: 131
507: 55, 57, 149, 210
50845: 49, 56, 64
51214: 147
51216: 149
51314: 106
516: 150
517: 77, 115
520: 64, 150
5201: 79, 113, 150
5246: 137
525: 57
53343: 150
53943: 109
540: 109, 155
542: 109
5423: 155
5445: 111, 137
545: 65, 166
54651: 1667, 175
54676: 145
546622: 49, 59
553: 145
5524: 33, 38
5589: 167
5601: 167
562: 167
56871: 33
572: 168
5736: 168
588: 168
58895: 168
590: 169
594: 169
5967: 169
596622: 169
60714: 169
61314: 169
6212: 170
623: 170, 236
6235: 180, 236
623788: 49, 59
6245: 236
626: 170
633: 170
6348: 129
6356: 236
6358: 236
63840: 145, 170
639: 50
640: 145
64190: 8
649: 85
650: 85, 87, 88
6506: 103
651: 87
6578: 86
658: 103
659: 86
6667: 103
6678: 103
668: 86
6701: 86
6712: 86
673: 86
677: 86
6778: 86
681: 87
68290: 901
684: 40, 91
686: 91
687: 91
688: 91
689: 91
690: 170
691: 87
6912: 87
695: 91
6969: 87
7001: 87
70714: 80, 122
71314: 91
71516: 91
716: 170
719: 170
7301: 170
73241: 1223
735: 123
737: 123
743: 170
744: 129
277
278
7456: 170
7478: 170
7558: 88
75560: 103
757: 99
758: 177
759: 99
7657: 87
776: 180
7835: 171
789: 180
78991: 180
78993: 171
789884: 49, 59, 129
803: 172
80314: 172
806: 98
81314: 172
821: 172
82732: 172
831: 174
833: 174
837: 173
874: 173
87581: 173
87584: 173
885: 116, 143
8856: 98, 116, 180
8858: 59, 68
885919: 49, 59
888: 98, 152
8889: 116
889: 59, 179
8901: 143, 196
8912: 172, 180, 215
892: 180
8945: 60
895: 60
903: 60
9035: 60, 64, 96
907: 60, 96, 243
90910: 138
911: 136
917: 60
91718: 144
92069: 105, 129
9201004: 49, 64
922: 109
9224: 107
937: 214
93841: 10910
9424: 110
947: 138
9526: 40
9578: 115
95764: 107

962: 107
964: 110
967: 110
9702: 110
976: 110
9824: 113
990: 180
9901: 180
9978: 112
1004: 66
1005: 110, 111
10056: 134
1005112: 49
1006: 11
101112: 146
1021: 152
10212: 113, 150
102731: 112
10301: 82
10367: 40
1040: 113
10501: 196
1051: 97
105268: 58
10579: 91
105765: 96
105960: 9
1077: 152
10858: 146
108999: 211
10902: 196
10968: 59, 82, 116
10969: 104
1098: 103
10989: 5, 105, 211
11002: 104
1102: 105
11023: 152
110410: 144
110610: 104
1107: 105
1110: 105
111012: 65, 152, 173
1112: 96
111213: 1524
Troades
1645: 191
170: 185
2489: 191
279: 192
2867: 192, 193
336: 192
3512: 184, 192
360: 193
36070: 1923
362: 193

367: 193
368: 152, 193
4618: 199
46974: 195
501: 198, 202
5012: 198
5045: 55, 214
51921: 196
548: 9
551: 195
5947: 197
61314: 136
64262: 197
6468: 199
6535: 197
659: 198
65961: 195
662: 197
77185: 195
861: 200
861871a: 200
883: 137
8835: 137
10657: 2378
1068103: 240
107587: 238
10991100: 238
1103: 29
110410: 238, 241
111117: 200
1117: 200, 241
111828: 2389
11268: 242
112831: 239
1129: 241, 242
1132: 242
11367: 239
1138: 242
11402: 242
11434: 243
11438: 2423
1143b4a: 243
1147: 243
11478: 239
11601: 239
11624: 196
116577: 241
[SENECA]
Hercules Oetaeus
207: 185
1457: 179
1472: 179
SHAKESPEARE
King Lear
4.6.1238: 889
Titus Andronicus
2.1.14: 59
2.1.131: 76
2.1.1335: 76
2.3.923: 88
2.3.98: 88
2.3.98104: 88
2.3.265: 72
2.4.3843: 701
3.1.1335: 46
4.1.69: 73
5.1.636: 72
5.2.2830: 72
5.2.80: 72
5.2.1424: 114
5.2.1945: 73
5.2.204: 73
SILIUS ITALICUS
Punica
5.653: 91
9.4025: 146
SOPHOCLES
Ajax
66976: 169
6745: 169
693717: 167
Antigone
111554: 167
Electra
810: 56
7785: 48
Oedipus Tyrannus
158215: 166
1086109: 167
121315: 205
1214: 205
Oedipus Coloneus
176976: 67
1779: 67
Trachiniae
63362: 167
STATIUS
Silvae
5.1.57: 124
Thebaid
1.12: 145
STRABO
1.2.8: 231
SUETONIUS
Augustus
15: 85
Caligula
32.3: 153
279
280
Nero
21: 9
56.1: 153
Tiberius
44: 153
TACITUS
Annales
6.7.5: 44
Dialogus
18.5: 132
TERENTIUS
Phormio
1617: 116
ULPIAN
Digest
3.2.11.1: 102
VARIUS
1 Ribbeck2 : 142
De morte, fr.4 Courtney: 100
VARRO
De lingua Latina
6.30: 39
Fragmenta operum incertorum (Salvadore)
66: 2
VIRGIL
Aeneid
1.203: 225
1.25796: 220
1.483: 200
2.3: 225
2.27097: 195
2.499501: 185
3.302: 199
3.304: 199
3.31012: 199
4.344: 195
4.46573: 206
4.469: 9
4.660: 178
5.1: 187
5.750: 28
6.27381: 9
6.4851: 53
6.7780: 29
6.1001: 29
7.44: 31, 35
7445: 71
7.45: 35
7.286322: 32
7.312: 152
7.315: 217
7.317: 35
7.322: 195
7.335: 35
7.33540: 32
7.336: 33
7.337: 33
7.3735: 29
7.38590: 34
7.386: 34, 35
7.40674: 34
7.422: 28
7.4567: 34
7.552: 179
7.5524: 33
7.559: 179
7.56871: 33
7.580: 51
9.1834: 126
9.33941: 126
9.4469: 41
9.7928: 125
10.1: 187
10.58: 195
10.443: 60
10.4546: 125
10.51720: 94
10.7239: 1256
10.7913: 41
10.793: 41
10.843: 116
11.1: 187
11.812: 94
12.49: 125
12.156: 78
12.161215: 93
12.179: 93
12.313: 35
12.9459: 93
Eclogues
4.1: 31
Georgics
3.89: 227
3.10.20: 59, 116
3.266: 217
3.2847: 217
General index
Abel, L. 14
Accetto, T. 11415
Accius
Atreus 30, 84
Bacchae 9
Aeschylus
Agamemnon 845
beginnings in 56
closure in 667
revenge plots in 141
Alexander the Great
and Caesar 219
and Dionysus 97
anagnorisis 60, 61, 66, 111, 171, 248
analepsis 187
Atreus
and cannibalism 1267
and foreknowledge 105
and memory 18990
and nefas 193
and Nero 97, 153
and passions 1058
and paternity doubts 5, 88, 1012,
105
and Procne 803, 1334
and satelles 15464
and the sublime 127, 225
as animal 98
as Dionysiac character 1338
as god 97
as poet/director 556
as sacerdos 849, 1513
compared to a lion 1228
compared to a tigress 122
intertextual competence 11517
linguistic prowess 11113
logic of 945, 98105
metadramatic role 90
puns of 77, 11315, 133
audience, reaction of 1823
see also frames
Auerbach, E. 16
Augustus
and Dionysus 97
and time 21920
Bacchic/Dionysiac elements 97, 1338
and Augustus 97
and enthousiasmos 121
and inspiration 10
and sparagmos 845
and women 77
in Alexander the Great 97
in Horace 51
in poetry 31, 33
in Procne 1334
in Virgil 76
beginnings
in Aeschylus 56
in Euripides 56
in Sophocles 56
belatedness in Senecan tragedy 223
in Oedipus 223
in Phoenissae 223
bi-logic 143, 21011, 250
Brecht, B. 2458
Caesar 126
and Alexander the Great 219
and calendar reform 21819
in Lucan 124
Calder, W. 16
cannibalism 94, 126, 135
children, death of 2012, 212
chorus 16476
in Sophocles 167
in Thyestes 1446, 148, 1657, 16970,
1746
in Troades 201
Chrysippus 24
civil strife in Thyestes 1467
Cleanthes 24, 22943
281
282
closure
in Aeschylus 66, 67
in Greek tragedy 667
in Ovid 249
in Seneca 659, 957, 173, 189
in Sophocles 67
in Virgil 678
comic elements in tragedy 1367
critical spectatorship 244
deception 612, 73, 79
Democritus 22, 53
Dingel, J. 21
dismemberment 201
dissimulation 114
Dupont, F. 15
173
ekpyrosis
Eliot, T. S. 75
emotions, Stoic theories of 231
Ennius 222
enthousiasmos 225, 29, 53, 90, 93, 121,
231
epic
and tragedy 835, 2078
in Ovid 24950
in Senecan tragedy 249
Euripides 845
and metatheatre 223
and sacrifice 923
Bacchae 612, 76, 97, 1338
beginnings in 56
Hercules furens 30
Medea 17
Orestes 40
Phoenissae 130
extispicium 1012
frames (narrative) 14, 4564, 92, 182, 250
Freud, S. 1619, 42, 878, 94
furor 21, 25, 46, 47, 143
Fury/Furies 245, 30, 379, 84,
18890
General index
ictus in Stoic theory of passions 232, 233
incest 94
and poetry 72
in Agamemnon 204
in Oedipus 196, 2046
in Troades 197
intertextuality 2218
and Bacchic elements 2278
and defence mechanisms 225
and horror 225
and memory 224
in Oedipus 2258
in Virgil 222, 2267
iterum 27
Juno as metadramatic character 1834, 186
knowledge 12, 47, 578
Lef e`vre, E. 147
Leo, F. 30
lexis in Stoic theory of poetry 229
lion-simile as stylistic icon 127
and Caesar 124
locus horridus 85, 879, 103, 127
and sexual symbolism 878
[Longinus] 12730, 132, 226, 232
Lucan, Bellum Civile 40, 87, 98, 124, 146
and metapoetry 13
Lucretius 98, 100, 120, 218
maius-motif 31, 34, 35, 51, 70, 130
Matte Blanco, I. 42, 94, 143
Mazzoli, G. 24
metadrama and metatheatre 1315, 367
in Lucan 13
in Oedipus 54
in Ovid 13
in Thyestes 49, 54
Mezentius 73, 1246
mise en abyme 11, 1819, 31
gigantomachy 98
Gigon, O. 147
Girard, R. 93
Gracian, B. 115
Naevius 222
nefas 31, 369, 41, 43, 47, 50, 74, 76, 91,
193
Nero 97
and Atreus 153
Nussbaum, M. 244
Hesiod 245
Horace 120
Bacchic elements in 51
Epode 16 1, 220
hunting imagery in Thyestes 99100,
135
Orlando, F. 10, 42
Ovid
and Aeneid 146
Ars amatoria 77
irony in 112
Metamorphoses and metapoetry 13, 249
General index
passions 20
peripeteia 248
phantasia 1301
in [Longinus] 232
Picone, G. 53
Piscator, E. 246
Plato 22, 53
and bacchic inspiration 119
description of tyrants 94, 11820
Plautus Trinummus 32
play-within-the-play 64
Plutarch 22931
poetry
and the female body 8990
and tyrants 11820
as danger 72, 229
as power 778, 87
Posidonius 24
Procne 81
as tigress 123
Dionysiac features 1334
prolepsis
187
psychoanalysis
and time 1867, 211
in post-Freudian thought 94
regression 189, 207
and repetition 189, 214
in Phoenissae 206
repetition 35, 186, 224
and female characters 767
and regression 214
in Agamemnon 194
in Stoic thought 21617
repression/return of the repressed 39, 424, 55,
75, 76
revenge in Thyestes 141
sacrifice and murder 934
satelles 46, 102, 106, 139, 15464
scelus 11, 279
Schiller, F. 2456
Schmeling, M. 14
Seidensticker, B. 34
Seneca
and theory of enthousiasmos 225
dating of works 20
De clementia 11819, 15964
style of prose works 13
Senecas tragedies
Agamemnon 28, 68, 137
Hercules furens 27, 912, 1836
Medea 1619, 43, 90, 146
Oedipus 812, 1820, 54, 87, 89
Phaedra 92
283
Phoenissae 1589, 206

Thyestes passim
Troades 92, 137, 18890, 1945, 1979, 225,
23743
sententiae 29, 106, 157
Shakespeare, W.
King Lear 889
Titus Andronicus 703, 111
Shelton, J.-A. 181
Sophocles 10, 83
and metatheatre 223
Antigone 193
beginnings in 56
chorus in 167
closure in 67
Oedipus Rex 12
Philoctetes 65
Tereus 83
spectatorship 57, 60, 235
in Aristotle 245
in Thyestes 2357
in Troades 23743
Stoic theories of passions 22935
sublime 23, 53, 55, 12732
and intertextuality 12930
stylistic features of 1312
Suetonius 97, 118
sun 95
disappearance of 1702
reversal of path 220
Szondi, P. 246
Tacitus 44
and irony 112
Tantalus in Thyestes 278, 378, 40,
41
Tarrant, R. 167
TereusProcne episode
in Ovid 17980
in Seneca and Ovid 2631, 701, 79,
115
Thyestes 1407
and gods 152
and Horace 116
and rational understanding 10811
and Stoic ethos 1478, 151
and Tantalus jr 14950
and Virgil 11516
and women 90
lapsus of 1489
time
and alienating effects 1867
and Augustus 21920
and Caesar 21819
and political discourse 21819
284
time (cont.)
and undoing of the past
(Ungeschehenmachen) 211
cubist treatment of 188
in Agamemnon 1801, 2024
in epic 1878, 202
in epic theatre 183, 248
in Epistulae ad Lucilium 21416
in Euripides Medea 210
in Greek tragedy 177
in Hercules furens 1836
in Lucan 21718
in Lycophron 218
in Medea 20814
in Ovid 1878
in psychoanalytic theory 1867, 211
in Senecan tragedy 177220, 2489
in Thyestes 17883
General index
in Troades 190202
in Virgil 187, 195, 202, 2067
reversal of 21416
Turnus 146
as a lion 1245
tyranny and maius-motif 130
in Euripides Phoenissae 130
Varius 1423
vates 89, 12, 53, 87, 121
vestigia 12, 19, 101
Virgil 846, 146
Aeneid 7 and Thyestes 27, 324,
845
and Homer 222
closure in 678
Wilamowitz-Moellendorff, U. von 18

The Passions in Play PDF

Uploaded by

Copyright:

Available Formats

You might also like

The Passions in Play PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Passions in Play PDF

Uploaded by

Copyright:

Available Formats

This page intentionally left blank

THE PASSIONS IN PLAY

Cambridge University Press has no responsibility for the persistence or accuracy of

per mia madre

Flectere si nequeo superos, Acheronta movebo.

Poetry, passions and knowledge

The poetics of furor

Non quis, sed uter

Fata se vertunt retro

The poetics of passions

Intertextuality and its discontents

The quotation from W. H. Auden, Epitaph on a Tyrant (1939) on

The following published translations have been used in this work:

t r ag e d i e s, Loeb Classical Library Volume 62, edited and translated by

Mme Martin: Quelle est la morale?

Thyestes embodies a tragic conflict, and an even more tragic contradiction,

The Passions in Play

The Passions in Play

unusually emphatic role in the emotional dynamics outlined above. As

The Passions in Play

it is perfectly plausible that a play such as Thyestes could have political

The relative importance is debated; see p. 70 n. 1.

sign of Stoic orthodoxy, or even only of Stoicizing morality. We must give

Poetry, passions and knowledge

iterque populis Ditis ad superos datur

negat enim sine furore Democritus

Poetry, passions and knowledge

The Passions in Play

only to a non-rational, horrific form of Dionysiac inspiration,6 and both

Jocastas exhortation at 826 confirms that we are to consider Oedipus

Poetry, passions and knowledge

The standard treatment of mise en abyme is still Dallenbach (1977).

The Passions in Play

Poetry, passions and knowledge

The Passions in Play

While I cannot review in detail here the extensive critical debate on

Poetry, passions and knowledge

organization of the drama, especially as they focus on the authorial role of

See Dupont (1985), with Woodman (1992) and Bartsch (1994).

Hutcheon (1984) 23.

The Passions in Play

understanding of its ancient counterpart. In modern fiction, metanarration

It would be interesting to combine Auerbachs (1959) treatment of realism with an analysis of

Poetry, passions and knowledge

The Passions in Play

Medea seems to be aware of the essentially literary nature of her pursuit.

Poetry, passions and knowledge

other characters, presumably because his persistent trust in reason makes

The Passions in Play

Poetry, passions and knowledge

Interesting observations in Moretti (1995) and Too (1994).

The Passions in Play

follow, I address especially these attempts at resolution and connect them

Poetry, passions and knowledge

Although Seneca is not engaged here in an explicit declaration of poetics,

The Passions in Play

While, in this particular context, the orgiastic frenzy of the converted is

Mazzoli (1970) 556.

Its Greek equivalent "