Aeschylus Agamemnon (Barbara Goward)

Aeschylus.
qxd 15/08/2005 15:50 Page 7
Preface
The ancient and appalling history of the House of Atreus, told

in different ways, was well-known to the original audience of
the Agamemnon. Its founder, Tantalus, king of Lydia, was a son
of Zeus and favourite of the gods. But he offended them – in
some versions by serving up to them the cooked flesh of his own
son Pelops as a test of their omniscience – and became one of
the four proverbial sinners perpetually tormented in Hades.
Pelops himself, his body reconstituted by Demeter, grew up to
win a deadly chariot race in Greece against Oenomaus, though
not without becoming embroiled in an act of treachery, which
resulted in an ineradicable curse. Two of Pelops’ sons were
Atreus and Thyestes, and in this generation the story becomes
one of fraternal rivalry: Thyestes slept with Atreus’ wife,
Aerope; Atreus expelled Thyestes, then, feigning reconciliation,
brought him back for a banquet. In ignorance Thyestes ate the
cooked flesh of his own sons before being shown their heads and
hands. Expelled again, Thyestes begot an avenger, in many
versions by incestuous union with his own daughter. The result
was Aegisthus. Atreus himself had two sons, Agamemnon and
Menelaus (known as the Atridae), who married two sisters,
Clytemnestra and Helen respectively. When Paris abducted
Helen, the Greeks mustered a force at Aulis under Agamemnon
and sailed away to fight under Troy’s walls, leaving
Clytemnestra alone. They were away for ten years.
This is the point at which the play begins.
7
Aeschylus.qxd 15/08/2005 15:50 Page 8
Acknowledgements
I would like to thank the following people: Ruth and Steven

Curson and Chris and Ben Goward, who read the manuscript
through and made it more lucid. Also, many infelicities (not to
say outright howlers) were picked up by Pat Easterling and
Susan Woodford, whose encouragement and thoroughly prac-
tical advice spring from a rare generosity of nature. Fiona
Macintosh was kindness itself. I am very grateful for all this
help, and for the support of Tom Harrison and Deborah Blake.
Whatever errors remain are indubitably my own.
I would also like to put in a word of thanks to all Greek
students at the City Lit, past and present, for the enjoyable and
sometimes inspirational discussions resulting from our patient,
pooled work on various play texts over the years.
8
Orientation: Aeschylus, Athens

and Dramatic Poetry
There is the sea. And who will drain its resources?

Clytemnestra at 958
Agamemnon is the first play of the Oresteia, our only surviving

ancient Greek trilogy. Neglected for centuries except as a diffi-
cult text for advanced study, over the last hundred and thirty
years, the play’s treatment of political and familial crisis has
inspired many stage productions in fresh translations and adap-
tations, so that many countries can now testify to its elemental
power in performance. Throughout this period of influential
and benchmark productions (of both single play and trilogy),
scholars have been publishing a wide variety of different critical
analyses. Yet, despite so much heavy traffic in both dramatic
and academic interpretation, Agamemnon still exudes its
mysterious aura of unassailable and inexhaustible richness.
The action, however, can be described in a single sentence:
‘Agamemnon returns home from Troy in triumph, but his wife
Clytemnestra murders him.’
Agamemnon is named after its returning hero, and has a
‘return home’ or nostos story shape. However, the eponymous
hero appears only in the central scene (810-974), and his part
would not have been assigned to the main actor. The other two
plays are named after their female choruses; Libation-Bearers
(sometimes left untranslated as Choephoroe) and Eumenides,
which means ‘The Kindly Ones’ and is a way of naming the
Furies ‘apotropaically’, without arousing their powers. In
9
Aeschylus: Agamemnon
these plays the story continues as follows: ‘in revenge,

Agamemnon’s son Orestes murders Clytemnestra, but is
pursued by the Furies for matricide. After receiving purifica-
tion in Delphi he is at last acquitted by a new court established
on the Areopagus in Athens, and the Furies are finally called
off by being given a new permanent home there.’ The lost satyr
play, Proteus, which rounded off the trilogy, dealt with
Menelaus and Helen in Egypt.1
Even when reduced to this scarecrow outline, it is clear that
Agamemnon lays the ground for a huge design, initiating
themes about crime, retribution and justice, and about conflict
between male and female and between family (oikos) and
society (polis). These conflicts are repeated again in the second
play and finally ‘resolved’ in the third; in some senses the
trilogy shows thesis, antithesis and synthesis. This book sets
out to explore some of the brilliant features of Agamemnon in
its own right: the interplay with contemporary Athens; the
range of stunning theatrical effects; the elusive role of the gods
in determining human destiny (as embodied in the fates of
Agamemnon and Cassandra, and as elucidated by the chorus);
the rich and startling language, dense with symbols, personifi-
cations and imagery; the play’s characteristic mingling of
memory and prediction, hope and fear.
Life and times

Aeschylus was born 525/4 BC to a eupatrid or aristocratic family
from Eleusis near Athens. He grew up during a period of
profound political and cultural development, both of which are
reflected in his plays. At the time of his birth, Athens was a
kingdom ruled by Peisistratos and his sons. To increase the
prestige of their city they had re-organised a local quadrennial
festival, the Panatheneia, into a pan-Hellenic event including,
besides athletic contests, contests for different kinds of poetry
(epic and choral lyric), music and dance. Here the young
Aeschylus, whose aristocratic education would anyway have
entailed memorising great chunks of Homer and lyric poetry,
10
1. Orientation: Aeschylus, Athens and Dramatic Poetry
would had have the opportunity to attend performances by

exponents from all over the Greek-speaking world.
Possibly the Peisistratids also re-organised another annual
spring festival, the Great Dionysia, establishing for it a contest
in tragedy, a new genre of poetry recently developed from choral
lyric (this development is described on pp. 18-22 below).2
Aeschylus probably watched these ‘proto-tragedies’ too, which,
before a theatre was built, would have been performed in the
market-place or agora. We can only guess at the nature of these
early attempts to dramatise myth: Aristotle wrote of ‘short
plots and ludicrous diction’3 while Garvie imagines them as
‘little more than a messenger speech with lyric content’.4
In 510, when Aeschylus was about fifteen, the Peisistratids
were expelled and, after a period of turmoil, a democracy was
instituted in 508. The first recorded democracy in the world, it
was a development springing from the crisis of the times.5 As
Cleisthenes’ reforms were gradually implemented, mass
citizen assemblies began to meet regularly on the Pnyx to regu-
late internal affairs and deal with foreign embassies. Mass
citizen juries began to determine the fate of litigants in the law
courts. Over the next decades these assemblies and juries
became a source of civic pride and, if the restructuring of the
Great Dionysia to include tragedy took place only now,6 tragedy
would be the third great institution innovated by Athenian
democracy.
Aeschylus began to submit tragedies for the festival in c. 498,
but his first victory (with an unknown entry) was not until 484.
Meanwhile Athens was facing a series of threats from Persia. In
490, King Darius had launched an attack, beaten off at
Marathon. Both Aeschylus and his brother took part in this
battle; his brother was killed. In 480, a second immense inva-
sion began. The Persian navy was defeated at Salamis, largely
by Athenian warships manned by citizen oarsmen – though not
before Athens itself, abandoned by the civilian population, had
been sacked. After the hostile army was seen off at Plataea in
the following year, it was possible for Athens to style itself the
saviour of the whole of Greece.
11
We now enter the period of the seven surviving plays (see

The Plays of Aeschylus, pp. 136-7 below): Persians (472) imag-
ines the outcome of the battle of Salamis from the Persian point
of view. It is the only surviving tragedy to take recent events
rather than myth as its subject matter (though we know of
others in the preceding two decades).
Seven against Thebes is dated to 468-7, but at some point
before then, Aeschylus is known to have made the first of at
least two trips to Sicily under the patronage of Hieron, the great
tyrant of Syracuse and patron of the lyric poets Pindar,
Bacchylides and Simonides. On this first visit he probably
composed the lost Women of Etna to celebrate the founding of
Hieron’s new city, Etna.
Back in Athens, Sophocles became a fellow-competitor, and
defeated Aeschylus with his first entry in 468 (play unknown).
By the latest date for Aeschylus’ next surviving play Suppliants
(467-56), Pericles had become a major force in the democracy;
in fact, as a wealthy aristocrat, he had funded the production of
Persians a decade or so earlier. It was with Pericles’ backing
that the Areopagus court was reformed by Ephialtes and an
alliance made with Argos, events reflected and alluded to in
Eumenides. In 458 Aeschylus won first prize for the Oresteia,
but not long afterwards returned to Sicily. Perhaps it was there
that his Prometheus plays, including Prometheus Bound, were
performed: there is no external dating, but internal stylistic
evidence suggests a date late in his life. The play may have been
finished by his son, Euphorion. Aeschylus died at Gela in Sicily
in 456/5.
His epitaph, recorded in the unreliable Life, refers not at all
to his poetry but to his prowess at Marathon. If not composed
by his sons, two of whom, in fact, succeeded him as dramatists,
the lines might reflect his own deathbed wishes:
This tomb covers the body of the Athenian Aeschylus, son of

Euphorion.
He died in grain-bearing Gela, but famous Marathon would tell
of his courage,
And so – with knowledge of it – would the long-haired Persian.
12
Aeschylus had experienced a good deal in his lifetime: kings

displaced by democrats, and radically new political and legal
systems introduced; he had also lived and fought through the
equivalent of a World War and faced the threat of imminent
slavery; he had seen his city in ruins. He did not live to see the
Acropolis, razed in 480 and left derelict in honour to the dead,
rebuilt with its splendid Periclean temples, nor the full growth
of the Athenian empire. Neither did he see much of the war
against Sparta which, within a generation, brought about the
decline of the city that could style itself, in the words attributed
to Pericles, ‘the education of Greece’. The Athenians may have
perceived Aeschylus as part of that education. It is said that the
Assembly awarded him a unique posthumous honour: tragedies
were composed in the first instance for a single performance at
the Great Dionysia, but in his case, public funding was provided
for further performances (this is not to say that revivals else-
where did not happen).
The political and social context

The private imagination that created Agamemnon reached its
public through a thoroughly political filter. The Great Dionysia
festival in Athens, where the tragedies were performed, was an
opportunity for the democracy to celebrate its achievements. Its
administration was entirely under state control. The chief city
magistrate (archon eponymos) had pre-selected the three tragic
playwrights whose works were to be shown. Under the city’s
‘liturgical’ system7 he had also chosen the citizens who were to
finance the choruses. The festival’s opening rituals included
street processions and sacrifice of bullocks. Then officials took
their places in the state theatre and a series of political cere-
monies and speeches ensued, including a parade in armour of
war-orphans raised at the state’s expense. Ten judges, one from
each tribe, had already been appointed. The huge citizen audi-
ence sat in blocks of seats probably assigned by tribe, with
non-Athenians in other designated sections, up to around
14,000 strong in all.8
13
The first contest was for choral song in honour of Dionysus

(called dithyramb), held between the ten tribes of Athens, with
one chorus of men and one of boys from each tribe, each chorus
composed of fifty singers. After the dithyrambic competition
came three sets of tragedies, one set performed per day and
each set concluded by a satyr play. Satyr plays, with their
invariable choruses of satyrs (male hybrids of horse and
human) headed by their father Silenos, were, like dithyramb,
closely associated with the cult of the festival god Dionysus;
their earthy dedication to lust and drunkenness, woven into a
mythical plot often connected to the preceding trilogy, allevi-
ated the serious mood. The final competition was between five
separate comedies.
Playwright competed against playwright for a single prize,
providing an interesting insight into the ideology of equality
projected by the democratic polis. Athenian society was innately
competitive; engaged constantly in war, it also structured its
peace-time institutions antagonistically. In the assembly, polit-
ical debate with speeches for and against was the standard
procedure leading to the vote; the law court system too was
adversarial. Agamemnon, with its scenes of lengthy open
debate, its technical legal vocabulary and its general concern
with establishing responsibility for wrong-doing and deter-
mining punishment, shows a dialectical engagement with the
city’s democratic processes. The trilogy as a whole famously
problematises the nature of justice itself, and is to end in the
institution of a civic process which directly mirrors the external
city and the recent reforms of its judicial system.
The tragedies themselves, however, contain little public self-
congratulation.9 At this sacred civic festival they are rather, as
Nussbaum suggests, vehicles of political reflection, encouraging
the audience towards self-examination and change.
Emotionally charged, the plays ‘challenge the audience to
inhabit (the ethical space) actively, as a contested place of moral
struggle’.10 The chorus of male citizens, addressed by
Agamemnon as andres politai, ‘gentlemen of the city’ (855)
helps the citizen audience identify with their fictional counter-
14
parts. Recent experience is mirrored in the play’s vivid descrip-

tions of the hardships of campaigning (326-37) and the grief of
bereaved families (437-55). The audience too have made the
Oresteia’s painful move from kingdom to democracy.
The setting of Agamemnon, however, is mythical Argos, and
Athens itself goes unmentioned. The use of a story remote in
time and place enables social norms to be explored and chal-
lenged without causing direct offence. One of the norms
challenged concerns the role of women, who in contemporary
Athens had no political or legal powers and a low social status.
They certainly did not act in the plays. Evidence whether
women were even permitted to go to the theatre at this date
remains inconclusive.11 In this context, it is remarkable that the
most powerful figure of Agamemnon is undoubtedly
Clytemnestra: this conundrum is considered in chapter four.
The audience is on holiday, but the festival mood created by
the initial procedures is solemn; the subject matter of tragedy,
like that of epic and choral lyric, is naturally appropriate to the
dignified setting. Foreigners ignorant of the tragic genre need
to be impressed by this native Attic product. The home audi-
ence, meanwhile, have been annually watching plays ever since
the revision of the festival programme and, not unlike regular
football fans, many are doubtless highly sophisticated specta-
tors, quick to spot stylistic and technical innovations. Each
individual probably does his own internal judging, forming
opinions on the excellence and weakness of each element –
acting, plot construction, dance, singing, language and so forth.
They are in general familiar with the story material
Aeschylus uses in Agamemnon: the king’s disastrous home-
coming would be well-known from Homer and other sources.
Nonetheless, Aeschylus rivets their attention by setting the
myth in a social context they can relate to, by creating novel
effects of suspense, and by making them wait for 1,343 verses
before the inevitable murder takes place. In the presentation of
this story, many effects of irony, suspense and surprise, which
seem to us to be the very stuff of theatre, are Aeschylus’ own
innovations, minted for this festival audience. Such effects
15
were to become almost standard in the hands of later theatre

practitioners.
Loss and survival

Before going any further, it is worth emphasising the almost
total lack of other contemporary texts, both literary and histor-
ical, which would otherwise help in the task of approaching
Agamemnon. A comparison with Shakespeare is useful here.
The vast proportion of his life’s output survives, and we are
thus able to describe a chronological development and changing
themes and techniques. Comparisons with fellow poets and
dramatists can be made. Abundant documents can cast light on
different aspects, such as contemporary theatres and theatre-
managers. Yet even here there are huge areas of uncertainty
and debate, such as accurate dating, multiple authorship, adul-
teration of the text, attribution of speakers and so forth. Even
so, to engage in these Shakespearian problems seems a luxury
compared to the efforts that must be made with our minimal
information about Aeschylus’ life and work.
What is more, the surviving texts (dating only from
Aeschylus’ mid-fifties, after he had been submitting tragedies
for at least twenty-five years) come down to us as copies copied
many times over, with likely accretions, omissions and errors.
Some of these imperfections are evident, others uncertain,
others no doubt invisible to us. Despite the painstaking work of
scholars, the text of Agamemnon is inevitably built on ‘decayed
and patched foundations’.12
Aeschylus was prodigally productive (see The Plays of
Aeschylus, pp. 136-7 below). Ninety play titles are attested in
the Suda (a tenth-century encyclopaedia), seventy in the unre-
liable Life, but of these only seven tragedies survive, and the
authorship of the seventh, Prometheus Bound, is debated
(although in fact normally discussed with the rest of the
Aeschylean corpus). It seems that Aeschylus often, but not
invariably, produced linked tetralogies, that is four connected
plays performed over the course of a single day, three linked
16
tragedies and a satyr play. Certainly Suppliants13 and less

certainly Prometheus Bound14 were, like Agamemnon, first
plays of linked tetralogies. From fragmentary evidence and
various attestations we are able to put other play titles together
and suggest other tetralogies; ones ‘about’ Odysseus, Achilles,
Ajax and Jason for example, in the sense that this one is ‘about’
Orestes (whence ‘Oresteia’) are fairly secure. Persians, however,
was part of a group of plays apparently without a central figure
or focus (Phineus, Persians, Glaucus of Potniae, satyr play
Prometheus Pyrcaeus).
With ten per cent or less of Aeschylus’ output surviving, and
that only from the last fifteen years of his life, we cannot
pretend that we have a representative sample of his work to
study. We cannot evaluate the play against the context of his
total output, far less against that of his fellow poets. It would be
unsafe, for example, to assume that every surmised trilogy of
Aeschylus made the same ultimately positive movement from
uncertainty and disorder to celebrate human achievement. The
fact is, we simply do not have the evidence.15
However, in the case of Agamemnon, the rest of the trilogy
can illuminate at least some of the problems of interpretation.
And in a more general way, a good deal of the play’s subject
matter and treatment, of its formal structure and of its poetic
texture, can be understood simply by realising that it is part of
a continuing tradition of poetry. Attic tragedy is, first and fore-
most, a synthesis of different kinds of poetry: the great
non-Attic genres of epic and lyric, together with the local
poetry of the political reformer Solon, archon of Athens 594/3,
whose iambic trimeter verse was taken over for the episodes
(see below).
Dramatic poetry seems to us qualitatively different from epic
and lyric because – obviously – it is performed by actors in a
theatre. But the difference between dramatic and earlier poetry
is not as great as might be thought. The culture of the archaic
and classical periods was a ‘performance culture’. There were
few written texts. The habit of private reading had barely
begun and the term ‘literature’ had yet to be invented. All types
17
of poetry were heard and seen in performance in settings such

as the festivals described above.
Choral lyric
Choral lyric, performed in costume with dance and accompa-
nying music, is tragedy’s essential source. The archaic period
(660-460) has been described as the great lyric age of Greece
and, during this time, any major festival in a community’s reli-
gious cycle (as well as its victories in battle or at games
festivals), was likely to be celebrated by trained choruses
performing in a public space. These choruses were composed of
representative single-sex groups from the community: men or
women, boys or girls as appropriate for the occasion who, elab-
orately dressed, sang and danced in honour of the god(s)
concerned. These costumed choruses, with their music, poetry
and dance, were perhaps the most thrilling peace-time spectacle
in the ancient world.
Choral lyric was written on commission by professional
poets such as Alcman, Stesichorus, Simonides, Bacchylides and
Pindar. Though the vast majority of it is lost, sufficient frag-
ments survive to give us a clear idea of the genre. The poems
ritually invoke a specific god and/or gods in general, together
with hopeful prayers, expressions of fear and anxiety, and
general religious and philosophical reflection, often expressed
as wise sayings (gnômai). As a way of interpreting the rela-
tionship between god and man, the poems also tell mythical
stories about heroes from myth. A narrative of almost epic
length may be related, or a familiar story may receive only a
fleeting reference.
Pindar, for example, was a slightly older contemporary of
Aeschylus, and his victory odes, composed for winners at the
great games festivals in Greece, have survived complete. For the
poem Pythian 4, Pindar was commissioned to celebrate a
victory at the Pythian Games of Arcesilas, king of Cyrene, a
North African city originally founded by colonists sailing from
the island of Thera. Accordingly, the poem makes a lengthy and
18
flattering analogy between the voyage of the colony’s founder

and the mythical voyage of Jason and his Argonauts to Colchis.
In a high emotional register, Pindar provides a full background
to the Argo expedition, details of the voyage, the direct speech
of Jason and Medea, and much moral advice.
These characteristics, taken from choral lyric, can be seen in
operation in the chorus of Agamemnon: the capacity for exten-
sive narration, for moral/religious reflection, and for strong
personal emotion. They serve drama wonderfully well: to put it
(much too) neatly, the narrative capacity becomes the telling of
the play’s background story, the religious elements give it intel-
lectual significance, while the emotional tone readily creates
suspense and ambiguity about the outcome of the plot.
One member of the chorus had always acted as
leader/conductor (the koryphaios). Eventually, another member
stepped out to impersonate the hero of the myth, using the first
person pronoun ‘I’ rather than the third ‘he’ to deliver a mono-
logue (which was, perhaps, as Aristotle suggests, initially
improvised). This figure would develop into the protagonist of
tragedy. When the koryphaios began to respond to this ‘I’ figure
also using the first person, we have what to us is the essential
situation of drama – a dialogue.
Myth now becomes something that choruses can enact as
well as narrate. ‘Thespis’ is the elusive figure traditionally
credited with adding prologue and speech (as opposed to
song) to what had previously been only choral performance,
but it is Aeschylus himself to whom the credit belongs for
many other innovations. Aristotle’s Poetics chapter 4, for
example, tells us that it was Aeschylus who increased the
number of actors from one to two, curtailed the role of the
chorus and gave speech the leading part. Aristotle credits
Sophocles with being the first to use a third actor, but
Aeschylus himself was quick to imitate this innovation, since
three actors seem required for Agamemnon 782-974 when
Agamemnon enters to Clytemnestra with Cassandra behind
him on his chariot (it is difficult to see how this could have
been otherwise arranged: there seems no unobtrusive
19
moment in the text when a non-speaking actor playing

Cassandra could be replaced by a speaking one).
Aeschylus’ speed at picking up Sophocles’ innovation
suggests that tragedy at this time was an innovative form ener-
getically exploring its new medium. In Agamemnon, the chorus
is still the most significant element, and most scenes consist of
only one actor responding to the chorus; at the same time,
however, the dilemmas and decisions of individual characters
are explored, and a whole range of thrilling dramatic techniques
can be seen in place for the first time in recorded theatre
history.
Choral origin for drama immediately accounts for many of
Agamemnon’s features, not least the vast number of lyric
verses. Choral odes account for 550 of the play’s 1,673 verses; in
the episodes, almost exactly half of their remaining 310 verses
are also lyric responsions. Their odes are some of the most wide-
ranging and philosophically probing passages in the whole of
tragedy, but they are still dramatically-evolved descendants of
the ‘community’ lyric from which they sprang and carry out
similar functions. Their identity too remains true to their
origins. Like all tragic choruses, they are a single-sex group,
broadly representative of a particular life-stage, though
Aeschylus has added specific idiosyncrasies of characterisation,
such as the extremity of their age, and their paradoxical blend
of knowledge of the past and blindness to the future.
With its extensiveness and power, the chorus is in a very real
sense the backbone of the whole play. The koryphaios has much
the biggest role of all since, as well as performing the lyrics, in
many of the episodes he acts as sole respondent to whichever
actor is on stage.
Myth and epic poetry

Another genre of poetry that exerted influence on tragedy was
the earliest type of all, epic. Tragedy does not use epic’s metre
(the hexameter), but its subject matter – myth – is the staple
of its story-lines (in fact all ancient poetry of any length uses
20
myth). Most of Aeschylus’ surviving plays use myths which

had already appeared in Homer or the Epic Cycle (for which
see p. 45).
The great forest of myth inhabited by the ancient Greeks is
a unique feature of their culture. It is impossible to over-
emphasise its influence on all areas of their thought and on all
artistic creativity. Recently redefined in a nutshell as ‘a socially
powerful traditional story’,16 individual myths covered a multi-
tude of subjects: the creation of the universe, the emergence of
the Olympian gods, the origins of man, the founding of indi-
vidual cities, the tales of heroes and their encounters with gods,
their feats, adventures, family relationships and deaths. There
was no limit on the re-telling and adaptation of such stories.
Professional bards, and the rhapsodes who continued their
tradition, lived by travelling to festivals round Greece where
they gave epic recitations (as did lyric poets too, making use of
the same body of material). No doubt these traditional stories
were often modified to please local audiences. However, by the
beginning of the fifth century, the two great epics of Homer,
Iliad and Odyssey, had come to hold a uniquely influential posi-
tion in the Greek-speaking world, akin to that of the Bible in
the European Middle Ages and Renaissance.17 To learn Homer
as a boy was to receive a moral education: he was thought to
teach a man how to live and die courageously, how to relate to
his fellow men and to the gods. Homer worked through complex
and sophisticated narrative that is by any standard compelling,
suspenseful and extremely vivid.
Plato describes Homer as ‘first of the tragic poets ‘(Republic
10.607a). The Iliad in particular, with its emphasis on loss and
suffering, and its evolving focus on inevitable death for Achilles
as an individual and for Troy as a community, suggests itself as
ready-made for tragedy. The type of discourse it uses is signifi-
cant too: nearly half the Iliad is direct rather than narrated
speech, and there is good evidence (e.g. from Plato’s Ion) that
the professional rhapsode, an actor avant la lettre, imperson-
ated the characters and acted out their speeches in a highly
emotive and dramatic way.
21
Despite the hierarchical world the heroes inhabit, there is a

vast humanity in the poem which encompasses the fates of
women and children as well as the warriors; Homer shows
immense understanding of psychological motivation, and
focuses on human behaviour and reactions in detailed ‘scenes’,
such as the quarrel in council between Agamemnon and
Achilles in Book 1, the tender exchange between Hector and
Andromache on the wall in Book 6, the encounter at night
between Priam and Achilles in Achilles’ hut in Book 24, as well
as in rarer soliloquies, such as Hector’s (22.99-130), his despair
wonderfully conveyed in the broken syntax. All this powerful
material, ‘pre-shaped’ for tragedy, was available to Aeschylus.
Late evidence that he acknowledged a debt to Homer comes
from Athenaeus’ Deipnosophistae where Aeschylus is attrib-
uted with the remark that his tragedies were, ‘mere slices from
Homer’s great banquets’.18 Some scholars, surveying the vast
range of play-titles, have even suggested that Aeschylus had the
project of putting the entire epic corpus into tragic form.
The structure of Agamemnon

(For unfamiliar terminology see Glossary, pp. 153-4)
By the time Agamemnon was performed in 458, tragic poetry –
a new and swiftly-evolving form when Aeschylus won his first
victory in 484 – had crystallised. A structure had emerged.
Plays almost invariably began with a prologue, and then
consisted of a basic alternation between choral odes (stasima)
in a variety of different lyric metres and recited ‘episodes’ or
acts (epeisodia), in which up to three actors and sometimes the
chorus-leader spoke to each other in iambic trimeters (or, less
frequently, tetrameters), tragic poetry’s closest approximation
to plain prose speech. Another metre, the anapaest, frequently
though not exclusively used for entries and exits, was also
employed occasionally. These are the three ‘levels’ of tragic
poetry: (1) sung lyric, in a variety of different metrical patterns,
each with their own emotional effect, (2) chanted anapaests and
(3) recited/spoken trimeters.
22
When, within an episode, a stage figure sings and the chorus

sing in reply, this is technically called an amoebaeon or, if it is a
lament, a kommos. If one character sings but the other remains
speaking in iambic trimeters, this is called an epirrhematic
composition (in the Cassandra scene, the chorus and the
prophetess’ fluctuating emotions are brilliantly expressed
through their alternations between song and speech).19 It would
be wrong to suggest that tragic structure was at any time rigid:
in Agamemnon the neat alternation of stasimon and episode
breaks down after the fourth episode.
The play is acted by a maximum of three male actors called,
in order of importance, the protagonist, deuteragonist and
tritagonist. Female roles were impersonated by men. Pickard-
Cambridge suggest that Agamemnon requires three actors: the
protagonist plays Clytemnestra, ‘and the parts of Agamemnon
and Cassandra require two further actors (it is only in lines
782-974 that all three actors are on stage together): the parts of
the Watchman, the Herald and Aegisthus could be variously
assigned to the actors of Agamemnon and Cassandra’.20
Cassandra is the only role with song, but Clytemnestra has
some recitative (1462-1576).
The stage figures wear elaborate costumes and masks which
cover their entire face and chin and include a full head of hair.
When they sing they are accompanied by an aulêtês or double-
pipe player.
The Outline of Agamemnon on pp. 138-43 has a double
purpose. It provides a synopsis of the action while at the same
time indicating the play’s formal structure, the purpose being
to give anyone reading the play in translation some sense of the
different constituent types of poetry. Within the episodes the
metre is iambic trimeter (unless specified otherwise) and here,
if the chorus participates, it will be only the koryphaios
speaking, not the entire group.21
23
Theatrical Space
And now, my dear, step out of that chariot …

Clytemnestra at 905-6
Imagine you are a director. Holding the script of Agamemnon in

your hand, walk into the ancient theatre. Actors trail expec-
tantly behind you. Where will you arrange them? What do the
different areas of space suggest? How, in short, will you fit the
text to the performance space?
This chapter attempts to understand the way the theatre
worked when Aeschylus first staged Agamemnon in 458, and to
describe the theatrical genius with which he exploited it. It tries
to describe the experience of watching that first performance,
and plots the influence of some of the new special effects in
subsequent tragedies.
The physical theatre at the time of Aeschylus

The diagram shows a plan of the ancient theatre in Athens set in
context with surrounding buildings. The audience sits in the open
air in daylight around the orchêstra. Facing them is a skênê or
stage-building. The skênê has a double function. It is a storage
place for costumes and masks when the limited number of
speaking actors (maximum three) need to change roles, but it also
symbolically represents the place where the play is set. Here, the
skênê represents the palace at Argos, so the orchestra naturally
represents the area just outside it. This inside-outside model is in
fact used by all subsequent ancient drama, both Greek and
Roman, both tragedy and comedy.
In 458 the skênê was perhaps a very new invention, maybe no
24
2. Theatrical Space
more than ten years old: Aeschylus’ three early plays (Persians,
Seven Against Thebes, Suppliants) seem designed without one.
Before the skênê came into existence perhaps a small booth
stored costumes. The date of its introduction is one of many
contested features.
The skênê has a door, but the audience cannot see what
happens inside, only who comes in and out.
The ancient theatre in Athens. A = orchêstra; B = skênê; C = eisodos.

Modified from Cambridge Ancient History, Plates to vols V and VI (new
edn, 1994), ‘The Theatre’ by J.R. Green, ch. 7, p. 150.
25
We need to pause here. The inability to see ‘inside’ (contrast

the effortless movement of the film camera from exterior to inte-
rior) is not to be understood as a restriction. It is, in fact, a
feature wonderfully exploited by Attic dramatists, who use it to
create suspense and to pursue the crisis of knowledge on which
so many plots are based. Much of this usage can be traced directly
back to Agamemnon, the first play to create such a powerful
contrast between inside and outside. Aeschylus was probably the
first playwright to associate the external orchêstra space with
what is light, knowable, seeable, rational and (sometimes) mascu-
line, and to contrast it with the internal space inside the oikos,
which is dark, unknowable, unpredictable and (sometimes) femi-
nine, the place of entrapment and death. This symbolic use of the
skênê will be discussed further on p. 89 below.
At some point, the inability to show what had taken place
inside was surmounted by using a device known as an ekkyk-
lêma: a flat cart, which could be wheeled out of the skênê with
a tableau of actor(s) on it (e.g. Sophocles Ajax 348ff.; the device
is used satirically by Aristophanes at Clouds 184ff.). It must
necessarily postdate the skênê, since only when an inside has
been created does the problem of showing it come into exis-
tence. However, even after its invention, events inside still
regularly continue to be described first before the display
appears. It is not clear if the ekkyklêma existed for the
Agamemnon. When the audience see Clytemnestra and her
victims at 1372, it could be that stage-hands brought the bodies
out and laid them at her feet.
The two long eisodoi or parodoi (entrances) are the route by
which the twelve members of the chorus enter and leave the
theatre; whether it is because they are too numerous to fit into
the skênê, or because the convention had become thoroughly
fixed before the skênê evolved, they regularly process into the
orchêstra after the prologue, singing their entry-song (called
parodos). Their departure by the same route marks the end of
the play; there is no curtain, and the play’s entire closing
section is known as the exodos (departure). Exceptions to
conventional practice are always worth considering very care-
26
2. Theatrical Space
fully for their likely ‘shocking’ effect on the audience, and in

Eumenides Aeschylus twice uses his chorus of Furies in a star-
tling way: first, they enter (as the text indicates) fast asleep
from the skênê (could the ekkyklêma have been used for this?),
and secondly, they leave the orchêstra half-way through, to
mark the change of setting from Delphi to Athens. Normally
the chorus, once on, stay put; when they do leave it is almost
invariably to mark a change of place.
By logical convention, actors subsequently entering along an
eisodos are understood to be arriving from outside the city.
Since a fair amount of time elapses between first becoming
visible on the exposed eisodos and reaching the acoustics of the
orchêstra to speak, such entries are usually covered by ‘entry
anapaests’, from the chorus, in which they first, perhaps,
describe the entrant, and then address him/her. In Aeschylus’
hands, this extended entry is exploited to heighten suspense. In
Agamemnon there are two such entries: first, the Herald, whose
unexpected appearance and possible news occupy the chorus for
fourteen anapaests (489-502) and secondly, the long-awaited
Agamemnon, whose victorious return from Troy is given a
sensational fillip by having him enter in a war-chariot, together
with Cassandra, his prize of war. The chorus take twenty-seven
anapaestic verses (782-809) to chant Agamemnon in.
No gods appear in Agamemnon and the gods who come on
stage in Eumenides enter from along the eisodoi, so we do not
know whether the mêchanê or crane – another device which
clearly postdates the skênê – had been invented yet. It was later
used frequently by Euripides for divine epiphanies (and
satirised by Aristophanes, e.g. Socrates’ entry, Clouds 223ff.).
The actors wear elaborate costumes and masks, and the text
of Agamemnon indicates that this chorus also carried staves.
Mute attendants may accompany the stage-figures:
Agamemnon has his soldiers and Clytemnestra her slave-
women in the central scene of this play: an actor in such a
non-speaking role was called a kôphon prosôpon, ‘an empty
mask’. The orchêstra (though possibly containing a natural rock
outcrop of some kind) is quite bare and, apart from an altar, to
27
which offerings are often made in the course of a play, there are
few stage props. In Agamemnon, Cassandra tears off her fillets;
in the first half of the next play Libation-Bearers, the action is
focused around Agamemnon’s tomb and the offerings brought
to it. Because they are sparse, stage items easily become
symbols of great potency, like the ‘carpet’ in Agamemnon (see
also, e.g., Hector’s sword in Sophocles’ Ajax, or Orestes’ urn in
Sophocles’ Electra).
The use of theatrical space in Agamemnon

Such was the space and such the mechanisms for which the text
of Agamemnon was written. We attempt now to block out
possible movements and gestures and, in particular, consider
exits and entries which, in the bare ancient theatre, could
create great impact. What do we actually see, and what do we
understand from what we see? What symbolic dimensions are
created?
Taplin’s important studies1 emphasise that Aeschylus was
not only a poet. In fact the verb graphein (= to write) was not
used to describe what ancient playwrights did. The verbs used
to describe their activity were poiein (= to create, conceptu-
alise, produce), or didaskein (= to teach, in the sense of
directing the chorus and actors). Ancient playwrights did not
write a play for production by others. Early on, at any rate, they
realised their own plays themselves in a thoroughly practical
way, taking full responsibility for every aspect. They were direc-
tors, choreographers and producers for their own work, and
probably composed the music and acted one of the roles.
Aeschylus composed the text of Agamemnon with the physical
space of the ancient theatre always in mind.
The full meaning of the work, like the libretto of an opera, is
realised only in performance. Our ability to recreate the first
performance is hindered by more problems than lack of
evidence about the exact nature of the theatre in 458. As with
every Attic drama, Agamemnon comes down to us with no
musical notation, no record of dance movements and no stage
28
2. Theatrical Space
directions, so that each exit and entrance has to be inferred

from the text. Information from scholia (ancient notes on the
text) and hypotheses (ancient summaries of plots) cannot be
guaranteed, since it may well reflect Hellenistic productions.
Some English translators have not considered these problems,
and their invented stage directions can be unlikely. The pres-
ence or absence of Clytemnestra on the stage at several points
up to 855 is a particularly acute problem, sometimes made
worse by the fact that the attribution of speakers can be in
doubt. On the subject of Clytemnestra, Taplin convincingly
argues that it is not tragic technique to have ‘unnoticed flitting
to and fro’ for any character, let alone such an authoritative one
as Clytemnestra: greater authority is established by having her
enter at key moments from the skênê, speak, and immediately
depart. Such incisive movements into and out of the palace
better sustain the required sense that it is she who controls the
threshold. Modern interpreters in the theatre, however, may
make a different choice.
Despite major uncertainties, Aeschylus’ brilliant innovation
with the theatre space is not in doubt. Agamemnon makes bold
and powerful use of every single one of its elements, not least
its possibly very new skênê, to create imaginary dimensions and
effects that still shock and frighten.
Agamemnon has a cast of royal characters and a stately
chorus. But the first character we see is a common man,
watching and waiting on events just as the audience are waiting
on the play to unfold. A natural position for a guard looking out
for a beacon would be on the roof, and although the text does
not confirm this, it seems a very likely position and a natural
exploitation of the new skênê. The Watchman’s magnificent
opening sentence (1-7) makes it clear that the skênê represents
the palace of Argos, a palace which through his prayer is placed
at once in relation to the august assemblies and unending
movements of the divine powers above it. Then the direction of
the Watchman’s gaze shifts and, looking along the eastern
eisodos, he sees the beacon-fire which originated in Troy.
The Watchman’s speech thus orientates the audience into
29
the spatial world of the play. From their seats they ‘see’ the
silent House of Atreus, the space before it, the gods in the
cosmos above, and Troy to the East. They will know from which
eisodos to expect Agamemnon.
The chorus enters (anapaests 40-103), perhaps along the other
eisodos, which remains spatially unspecific. As their narrative
takes us back to the outset of the war, the immediate location
fades and we move up again to the circling flight of the eagles and
even higher above them (hupatos, 55, ‘highest’) to the indetermi-
nate god who sends a Fury to Troy. Aeschylus then contrasts this
free aerial movement with the chorus’ earth-bound debility. The
chorus, like the setting of the play, is restricted to Argos: ten
years ago they were already too old to go to Troy (73-82). Would
‘stiff’ choreography and sticks reflect their age?
At 84, the present time in Argos returns when they ask
Clytemnestra for news. There is debate about her presence
here. In Taplin’s opinion their request is made through the wall
of the skênê and she is not visible at this point. At any rate, she
makes no reply (the first instance of questions without answers,
of problematic communication; an important theme in the
play). The vertical axis again dominates their resumed narra-
tive; both in its opening frame, ‘persuasion still breathes down
upon me from the god’ (105-6), and again with the portent of
the eagles. Their meditation on the three generations of rulers
in Olympus (160ff.) sustains the dimension of height.
The chorus return us to the stage at the end of their parodos
by saying, ‘may it all turn out “as this ever-present sole-
guardian bulwark of the land of Argos desires” ’ (Denniston and
Page’s translation of 255-7). But this is another problem. To
whom are they referring? Clytemnestra (addressed with some
deference in the following iambic trimeters), or themselves? Is
the chorus ironically addressing Clytemnestra with the extrav-
agant and deceitful flattery of her own manner, or are they
innocently referring to her, at the same time uttering an inad-
vertent dysphêmia (ill-omened expression)? Or could they,
despite their extreme age, be referring to themselves as a
‘defence’? I emphasise this difficulty to make the point that the
30
2. Theatrical Space
problem is not a textual one but one of stage meaning:

Aeschylus’ direction of the original performance could have
removed any ambiguity by the use of gesture.
Clytemnestra’s Beacon speech takes us East and upwards
again as she lists the series of fire-topped mountains. Inactivity
on stage is counter-balanced by the vigour of the urgent leaping
flames, as Hephaistus sends them zigzagging westward through
space, connecting offstage Troy to onstage Argos. Rosenmeyer
well describes a ‘growth of excitement and significance’2 as the
description progresses. There is nothing simple or linear about
the moving light, which symbolically portends Agamemnon’s
return from Troy in victory, and which is totally under its
narrator’s control. Clytemnestra’s second narrative returns to
Troy and includes a warning. Immediately after delivering it
she then makes an incisive exit (350).
The first stasimon opens up and away in Troy. Zeus and
Night have thrown a net over Troy’s battlements; Zeus’ arrow
has flown accurately through the air to strike Paris. From 399f.,
the traffic between Troy and Argos is heavy: hither came Paris
from Troy; hence Helen slipped from Menelaus’ sight to come to
Troy. Hither now from Troy will come the urns bearing the
ashes of the dead.
From Troy, an unexpected Herald now arrives. The speech
announcing his entry (489-502) is traditionally divided between
Clytemnestra (489-500) and the chorus (501-2) but assuming
that she exited at 350, the chorus delivers the entire announce-
ment. Their speech not only covers the time it takes the Herald
to travel along the eastern eisodos, but also creates more
tension: he looks auspicious with his olive crown, but can he
truly resolve their anxieties?
The Herald makes polar vertical movements, kissing the
earth and praying to sun and gods on the rooftops. But the
chorus’ anxieties multiply. In the Herald’s naive account, it
becomes clear that the Greeks have indeed offended the gods in
their sack: the wreck of Menelaus’ ships and his disappearance
hint that, somewhere in the eastern distance, on one son of
Atreus the gods are already exacting punishment.
31
The most striking stage action of this scene (if Taplin is

right), is the telling moment at which the chorus are recom-
mending the Herald to go in to give the news to Clytemnestra
(585-6) – when the doors unexpectedly open to reveal her.
Clytemnestra is dômatôn kuna, ‘the watchdog of the house’
(607), and her control over who enters, and when and how,
which is established here, will become even more explicit later.
After her abrupt appearance, she delivers a single speech (587-
614) and disappears again.
In the second stasimon, the chorus yet again visit Troy to
consider the effect of Helen’s marriage to Paris. But from 750
space becomes indeterminate: their ‘old logos’ about the doom
that lies in wait for the generations of the prosperous could apply
to onstage Argos as much as to offstage Troy: there is no precise
location for atê (disaster, 710). Uncertainty is building fast.
Chariot entry
Now the chorus chant anapaests announcing Agamemnon’s
entry. They hail him, warn of possible unrest among the popu-
lace, speak of their own qualms in the past, confirm their own
deep loyalty, but hint at the disloyalty of others. All the first
part of the play has led up to this moment of return (most
recently through the Herald with his repeated hêkei, ‘he comes’
(522, 531). Agamemnon has been repeatedly declared victor of
Troy (e.g. 264f., 355f., 524f., 575f.) and light from darkness
(25f., 264, 522). But Aeschylus has also made the audience
aware of the king’s vulnerability to attack from every dimen-
sion of the theatre’s virtual space: from the gods above, from
the populace of Argos (not assigned any spatial dimension), and
from persons as yet unspecified within the palace.
Agamemnon enters magnificently on a chariot, though this is
not a new departure for tragedy. Cassandra is in the chariot
behind him, perhaps hardly visible. Attendants accompany him,
probably pulling his chariot, unless real horses were used. We
have no indication of how or when the chariot was removed.
Agamemnon’s chariot-entry spectacularly symbolises his
32
2. Theatrical Space
triumph over Troy. He reappears after ten years – but only to

disappear for good. When he leaves his elevated position to close
the scene, it will be to go barefoot and silent into the palace,
while his wife speaks. Aeschylus gives Agamemnon a stunning
symbolic entry and an equally stunning symbolic exit. Together,
entry and exit encapsulate his life’s destiny. Aeschylus had used
the contrast between a chariot entry and a movement on foot to
similar, but less powerful effect in Persians. There, the Queen
makes an initial chariot entry (Pers. 159ff.) but pointedly re-
appears on foot ‘without former luxury’ after learning of the
defeat (ibid. 607ff.); Aeschylus, with the skênê now in place, can
recast what is essentially the same feature.
Lure-murder
The greatest and most original coup de théâtre is the exact way
Agamemnon enters the stage-building. The last four lines of his
speech (851-4) seem to indicate that he is preparing to go inside
of his own accord. Then Clytemnestra, who begins speaking at
855, probably again appears unannounced at the door of the
skênê, blocking the way (as she did with the Herald at 587).
Like Agamemnon, she too has attendants, and at 908 she bids
them spread coverings (909) to create a purple-strewn passage
(910). ‘Do not … make my passage grounds for envy by
strewing it with garments’, Agamemnon responds, ‘Mortal as I
am, it is in no way for me to walk on beautiful embroidered
works without fear. … My fame shouts aloud without foot-
wipers and embroideries’ (920-7). Later he calls the fabric
sea-workings of the gods and wealth, weaving purchased with
silver (946, 949).
These cloths or clothes, over which Clytemnestra and
Agamemnon debate, have no mythical precursors. Their use
here shows Aeschylus’ original and highly developed sense of
theatre’s symbolic potential: they are surely the most sinister
‘prop’ in ancient tragedy.
From the abundant and dense language that describes them,
what can we understand about their physical appearance? Are
33
they one continuous piece, consisting of several looms-worth

stitched together, or a series of separate items of clothing such as
cloaks? To think of a carpet is to miss the point. Carpets are
made for walking on; Agamemnon’s reluctance and the fabric’s
high value indicate that this is not. The material is sea-worked
because it has been dyed using a crushed shell (Greek porphura,
Latin murex), a laborious and costly process producing shades in
the violet-scarlet range. It has been further worked by embroi-
derers and so, both in cost and labour, it manifests the wealth of
the house (for an excellent discussion, see Jenkins, 1985). The
colour is another disquieting aspect: it looks very possibly like
blood, although this notion is not developed in this scene.
The crisis of the play approaches. The audience knows that
Agamemnon logically cannot remain in his chariot. He must go
into the stage-building. But must it be over the ‘carpet’? The
stichomythia (931-45), with its military language, indicates a
verbal battle between husband and wife. The male victor is soon
vanquished by the woman’s clever words.
The language of 919-25, 935-6 and 946-9 all make it clear
that it is dangerous for Agamemnon to walk on the cloth, but
what did the action symbolise to the watching audience? They
perhaps remembered the ‘impious trampling’ imagery of the
chorus (367ff.), perhaps also Agamemnon’s morally dubious
earlier actions: the sacrifice of Iphigeneia, the pillage of the
army under his command at Troy. Now a literal trampling is
enacted, which might stand for all his previous morally ambiva-
lent actions. Now too Aeschylus shows him in the process of
decision, just as before the chorus had described him making
the decision over Iphigeneia.
At 950 Agamemnon makes the first reference to Cassandra,
perhaps only now noticed by the audience in his chariot. The
continuous presence of Cassandra established here creates a
visual continuity between the two phases of the play (pre- and
post-return). A victim from Troy is established in Argos.
The text indicates that, as Clytemnestra begins her speech at
958, Agamemnon begins his walk over the fabric, to disappear
inside by the end at 972. Rehearsal must have been necessary
34
2. Theatrical Space
to make his walk correspond exactly to the fifteen allotted lines.

His movement in this extravagant theatrical exit would be slow.
Maid-servants roll up the purple cloths behind him, adding to
the sinister sense that this is a one-way-only journey. It matters
that Clytemnestra’s voice dominates his silent walk. Finally,
after 974, husband and wife are to be imagined together inside
the palace, privately re-united for the first time in ten years.
If the skênê was indeed a recent invention, Aeschylus had
lost no time in exploiting it to the full in this splendid exit. He
makes use of the same scene-type in his next play, Libation-
Bearers. This time it is Aegisthus who is lured inside to his
death. In a previous scene Orestes, by entering in disguise and
announcing his own death, has managed to gain access to the
palace (Clytemnestra’s control over the threshold has slipped).
He waits inside while Aegisthus is enticed in ‘to hear the news’
by Orestes’ old Nurse, who has been primed for the purpose. As
Orestes (in the audience’s imagination) exacts his revenge
inside, the chorus is left alone onstage to reflect the tension
(841ff.). After the murder a tableau follows, showing Orestes
standing over his kinsman’s corpse.
Similarly-constructed scenes, in which the murder victim is
tricked or lured inside, leaving the chorus to chant or sing,
‘covering’ the invisible murder on the empty stage, may be
found in Sophocles’ Electra and in Euripides’ Medea, Hecuba,
Heracles, Electra and Orestes. A horrible irony is created in
each case: the audience watch the ignorant victim duped and
disappearing. They know that he/she will not reappear alive,
and that theatre convention debars intervention from the
omnipresent chorus. As helpless as them, the audience can only
wait to hear the cries from within.3
In the following stasimon the chorus is deeply disturbed –
and disturbed by its own disturbance. They cannot understand
why terrible anxiety should be their reaction to witnessing the
reunion of their victorious king and his wife. It is surely puzzle-
ment (rather than full realisation) that Aeschylus wished to
inspire in his original audience by the action of the previous
scene. Agamemnon’s exit looked like a kind of trap and clearly
35
had sinister significance, but its exact import (like that of all
effective symbols and symbolic actions, in fact) surely continues
to remain enigmatic: the audience will understand its meaning
more fully only after another doomed exit into the house –
Cassandra’s (1330). Meanwhile, in this stasimon the irrevoca-
bility of Agamemnon’s exit seems to emerge almost unheeded
at 1019ff., where the chorus ponder, ‘who by an incantation
could summon back dark blood, once it has fatally fallen before
a man?’ Here propar andros (‘before a man, at a man’s feet’)
suddenly seems to refer to the walk on the purple cloth.
After this lyric, the audience logically expects Agamemnon’s
murder to follow; though an onstage presence, Cassandra has
been given little significance as yet. At 1035 it is a surprise
when Clytemnestra re-enters and we realise the situation inside
the palace is as yet unchanged. The following scene is an
intriguing meeting of wife and mistress, two women with oppo-
site and exceptional powers over language – the weapon of one
is eloquence, the other, unbroken silence.
With the same veiled intent and using the same phrase as for
her husband, ekbain’ apênês têsde, ‘get down from this chariot’
(906, 1039), Clytemnestra attempts to get Cassandra into the
house. This time, however, she cannot initiate a dialogue.
Cassandra’s persistent silence implicitly contrasts with that of
Agamemnon. Having previously ‘won’, we now see
Clytemnestra defeated by the prophetess’ lack of response; this
is the first time we have seen her eloquence defeated.
Silence and the kôphon prosôpon

The sustained power of Cassandra’s fascinating silence must
make the audience now realise that she is too important to be a
mere kôphon prosôpon, a non-speaking actor. Having thus
created intense audience interest in a figure characterised by
silence, Aeschylus transfixes them by her extraordinary utter-
ance. So far, within the episodes no character has sung, and so
her song here is in itself a new element. But that is not all. The
moment when she opens her mouth is a coup de théâtre and
36
2. Theatrical Space
begins a significant alteration in the prevailing character and

direction of the play. Aeschylus gives his prophetess a new, fren-
zied, extraordinarily arresting and individual voice, and in the
course of her scene (1071-1330), the virtual space of the play
undergoes a radical shift.
Earlier on, we said there was a convention of having non-
speaking actors on the stage (kôpha prosôpa), often to create a
sense of status – companions, members of a retinue, etc.
Aeschylus here cleverly makes his audience at first think that
Cassandra is one such kôphon prosôpon. In Libation-Bearers he
pulls the same trick to even greater effect: Pylades is Orestes’
conventional companion, and silently accompanies him
throughout the play until the critical moment of matricide
arrives. Then Clytemnestra bares her breast and dares Orestes
to kill her, his own mother. As Knox describes it, ‘And Orestes
breaks. The command of the god Apollo, the unavenged blood of
Agamemnon murdered in the bath, Orestes’ own exile and
poverty, all the forces behind him – gods, family and his own
ambition – fail him at this supreme moment. He turns to the
silent figure who has followed him step by step throughout the
play and asks him a direct question: “Pylades, what am I to do?”
And now at last Pylades speaks. He has only three lines but
they are enough: “Then what becomes in the future of Apollo’s
oracles, what meaning in the sworn pledge of faith? Better
offend the whole human race than the gods.” It is the voice of
Apollo himself; these three lines seal Clytemnestra’s death
warrant. … The third actor in Libation-Bearers is used to
dominate the stage for one tremendous moment in which
mother and son hang on his words for life or death.’4
Aeschylus was famous for his silent figures, who are satirised
by Aristophanes at Frogs 911-13; we know that a muffled, silent
Achilles began two of the three plays of his lost Achilleia trilogy.
The silent presence of children at the end of Sophocles’ Oedipus
Tyrannus, and Deianeira’s shifting sense of Iole’s silent identity
among the captive women in Sophocles’ Trachiniae testify to
the eerie power of stage silence – of which Aeschylus was the
originator.
37
In Agamemnon, the significance of Troy, so important earlier,

now fades and is relevant only to Cassandra’s past life. At last
the physical identity of the skênê comes to the fore and with it,
the long-disguised significance of the house facing the audience.
We have reached the play’s core: the past, present and future
destiny of the House of Atreus.
Oikos (house) in Greek designates both physical building and
abstract sense of household, the sum of family members. It also
means ‘ruling house’, and substance of the house, in the sense of
its total wealth of land and material possessions, not least
including its offspring. The continuity of the house in the beget-
ting of children and in careful economic management is central
to Greek thought and stressed in many texts. Oikos can also
contrast with polis, as personal contrasts with political. Further,
oikos can mean inheritance, a meaning which illuminates the
chorus’ notion of a doom passing down the generations.5
This is the House of which the Watchman said that, if it could
find a voice, it would speak most clearly (37-8). Soon it will find
a terrible voice, virtually at least, as Agamemnon’s death-cries
resonate from within (1343, 1345). Meanwhile, we have seen
that Clytemnestra controls its threshold and that she made her
husband trample on its substance. Clytemnestra argued that
‘the house doesn’t know poverty’ (962); but already the chorus
have pointed to the vulnerability of prosperity (750f.). All the
action of the play so far has led into this house. Under
Clytemnestra’s control, the Herald entered, Agamemnon has
entered, and Cassandra too must finally go inside.
Cassandra’s function in this scene is to reveal the house’s
hidden horrors. As soon as her shrieks become comprehensible,
she asks what house Apollo has led her to, and the chorus
emphatically confirms it to be the House of Atreus (1087-8). The
audience consider afresh the skênê in front of them, which
Cassandra immediately describes as misotheon (an adjective
inextricably meaning both ‘hated by’ and ‘hating’ the gods); a
house that is ‘well aware of kin-murder … (next word corrupt)
… a human slaughter-house and blood-spattered ground’
(1092). In it are the slaughtered weeping babies and their roast
38
2. Theatrical Space
flesh, which their father ate, and in the bathhouse a new

terrible evil: ‘(Clytemnestra) stretching out hand after hand,
reaching them out’ (1110-1). ‘What Fury are you summoning to
raise its cry through the house?’ ask the chorus (1119).
The house remains dominant when Cassandra tries again to
communicate her visions in iambic trimeters. ‘A chorus that
chants in unison, but is not melodious, never leaves this house
… Look! A revelling band of Furies of the family waits in the
house, hard to send out, drunk on human blood to give itself
greater boldness. Sitting in siege on the house they sing of
primal atê … (1186ff.). Again at 1214ff., she asks the chorus to
‘see the children, like figures in dreams, their hands imploringly
holding their own flesh, assailing this house … they are utterly
clear’. As the scene approaches its end, the text indicates that
three times Cassandra goes up to its door, bravely walking into
her own destruction (1291, 1306, 1313). Addressing it as the
‘gates of Hades’ now has a real and powerful meaning. She is
repelled by the house’s smell of blood, ‘the vapour that comes
from a tomb’, but finally enters, speaking magnificent lines on
the human condition as she does so. The autonomous dignity of
her exit makes an ironic contrast to Agamemnon’s silent with-
drawal.
The chorus begins to reflect on what they have just
witnessed, but their understanding is still dawning when there
is a shock: Agamemnon’s death-cries heard through the wall of
the stage-building (1343 and 1345; the chorus begins its reac-
tion on 1344). We have, perhaps, been expecting these cries ever
since Agamemnon’s exit at 974, but the powerful interest of the
long Cassandra scene has diverted our attention, so Aeschylus
brings it about that the king’s cries fall on our ears with a
painfully renewed sense of the inevitable.
Offstage cries
This is the first known example of an offstage voice being
projected through the wall of the skênê, and another new
theatrical technique to be attributed to Aeschylus. The main
39
effect, of course, is to make the murder palpably immediate to

the audience. At Libation-Bearers 869, Aegisthus similarly cries
through the wall.
Sophocles and Euripides further developed this feature of
offstage cries in lure-murders, always to the end of making a
murder seem more real. At Euripides Medea 1271ff., the little
boys cry out offstage (in iambic trimeters), and the chorus reply
in dochmiacs, picking up their vocabulary. The close verbal
interaction makes their lack of physical intervention particu-
larly horrible. In other plays, when a stage figure remains on
stage as well as the chorus (as for example, at Sophocles Electra
1398ff., Euripides Orestes 1246ff.), a vivid chain of communica-
tion can be created, as the one nearer the skênê can give a
blow-by-blow account of ‘what they hear’ to the other.
Sometimes even further immediacy is created when a line or
lines of verse are divided between speakers on either side of the
skênê wall, as when Electra eggs on her (offstage) brother to kill
Clytemnestra (Sophocles Electra 1410ff.; such a divided line is
called antilabê). None of these effects, of course, could take
place before the skênê evolved.
Offstage voices, whether only imagined or actually heard by
the audience, went on to be quite commonly used outside the
context of a lure-murder. There are, for example, two instances
in Euripides’ Hippolytus which both create a tremendous
immediacy: first, at 565ff., in a scene charged with horror and
tension, Phaedra, with her ear to the wall, gives a blow-by-blow
report to the chorus of ‘Hippolytus’ reactions to the Nurse’;
later in the play at 776, the offstage Nurse cries out for help
(this time audibly) – Phaedra has just hanged herself.
The cries are heard, and from 1346-1371, Aeschylus capi-
talises wonderfully on the chorus’ conventional inability to ‘go
inside’. Instead, they break down into their constituent
members, and in the twelve two-line speeches of this section, an
array of contradictory suggestions and proposals are made to
meet the crisis. We can only imagine the chorus’ movement at
this point – certainly a vivid onstage portrayal of political chaos
is created. Before they are anywhere near deciding how to
40
2. Theatrical Space
respond, Aeschylus introduces his next shock: Clytemnestra

appears.
Tableaux
If early tragedy, as is likely, contained a great deal of message
narrative, the original audience might well have expected a
messenger to emerge at this point. A servant could easily fill
this role (as at Libation-Bearers 875f.) They are perhaps star-
tled to see Clytemnestra herself appear so quickly from the door
of the stage building – yet she has done so already four times
previously (258, 587, 855, 1035). All these entries were theatri-
cally powerful, but none more so than this fifth one in which
her naked power is finally revealed. Holding a weapon, she
stands over the corpse of Agamemnon, wrapped in a net, and
that of Cassandra. Whether positioned on an ekkyklêma or not,
the tableau visibly represents the fulfilment of the play’s action.
Words are temporarily redundant.
There is a mirror tableau in the next play when Orestes
appears triumphant over the body of his mother. By using an
identical tableau, Aeschylus gives the audience visual evidence
that the successive murders are both reciprocal and identical;
they understand that the House of Atreus is doomed to repeat
its bloody acts.
Tableaux are found in subsequent plays and do not always
display corpses – there are, for example the sleeping Furies in
Eumenides and the grotesque sight of Ajax sitting on a heap of
slaughtered animals in Sophocles Ajax. Their effect is always to
make the audience gasp.
In the agôn which follows Clytemnestra’s speech of triumph,
as she and the chorus attempt to establish the real responsi-
bility for the crime, the visual focus stays on the figures from
the tableau – the murderess and her major victim. But other
dimensions are activated as well: the House is falling under a
hail of blood (1532ff.); Hades below is Agamemnon’s destina-
tion (1528f.), where Iphigeneia will welcome him (1555f.).
Eventually both chorus and Clytemnestra come to agree that
41
the daimôn or alastôr (spirit or avenging spirit) of the House

was active in the murder, and at this point the tragedy seems
about to end, in an uneasy compromise. Aeschylus now gives us
his final surprise entry: Aegisthus, with henchmen. He enters
from outside (see 1608), perhaps symbolising his return from
exile and showing clearly that he has played no active part in
his cousin’s murder. Clytemnestra’s verses 1567-1576 would be
delivered as he becomes visible, but there is no usual entry-
announcement; perhaps he slinks in. His final exit, however, is
into the palace with Clytemnestra – a life’s ambition visibly
achieved and a marked contrast to Agamemnon’s earlier
‘defeated’ exit.
Aegisthus gives a lucid account of Thyestes’ banquet (1583-
1602). From a new standpoint he fully confirms the accursed
(1602) history of the House, previously described in
Cassandra’s visionary glimpses and thrashed out by
Clytemnestra and the chorus. His naked power is brutal and
almost provokes a stage fight (1650-1); sufficiently so, at any
rate, to demonstrate that the chorus is too weak to resist.
Though they invoke the daimôn to bring Orestes back for
revenge, Clytemnestra persuades them to leave. The text of the
ending is defective, but Aeschylus does not seem to have
written the usual exit anapaests for them. Is this another inno-
vation, to make their departure piece-meal and disjointed, in
accordance with their diminished status? As the play ends,
Clytemnestra and Aegisthus go into the palace to begin their
new, oppressive regime.
42
The Story: Myth and Narrative

Technique
I’d like to hear your story again and admire it from beginning
to end …
Chorus to Clytemnestra, 318-19
This chapter turns away from Aeschylus’ wizardry in the theatre

to refocus on his genius as a story-teller. To get a perspective on
his narrative strategies, we shall first survey earlier versions of
the myth of Agamemnon’s murder, and then discuss in some
detail the way Aeschylus re-worked it for the stage.
Literary precursors: earlier treatments of the myth

The Trojan War background of Agamemnon comes essentially
from the Iliad, with some interesting differences of detail. In
Homer, Agamemnon has a son, Orestes, and three daughters,
Chrysothemis, Laodice and Iphianassa; no Iphigeneia as such
(and no Electra). In the opening book, Agamemnon remarks
that he prefers his concubine Chryseis to his wife (Iliad 1.113-
15; at Ag. 1439 Clytemnestra seems to echo this when she refers
to Cassandra as ‘a Chryseis’). Aulis is the muster point for the
outgoing ships and the place where a significant portent
occurred which received interpretation from Calchas (2.299-
330), but there is no adverse wind, nor daughter-sacrifice. On
the Trojan side, Cassandra is cast as merely one of Priam’s
daughters with no prophetic powers. She is briefly mentioned
only twice: her betrothed dies in battle (13.361ff.); she is the
43
first to see Priam’s return from the Greek camp (24.697-701).

The context seems to indicate keen eyesight, not foresight.
The Odyssey is the major source for Agamemnon. In fact it
contains a bewildering variety of accounts of Agamemnon’s
return and murder, and of the revenge taken by his son. The
divergency of these accounts, with their different details and
emphases, are best accounted for as varying according to the
speaker – his point of view within the text, and his persuasive
intention towards his listener. The level of Clytemnestra’s
participation in her husband’s murder varies considerably.
In the Odyssey, the triad of husband, wife and son
(Odysseus, Penelope and Telemachus) are frequently compared
with the triad of Agamemnon, Clytemnestra and Orestes to
create an alternative narrative model.1 In Books 1-4 (the
Telemachy), there is a concern to rouse the young Telemachus
into action on behalf of his wronged father, and Orestes is
brought in to be a positive role model for this purpose. Athena,
Nestor and Menelaus (quoting Proteus) all narrate what looks
like a plain tale of male revenge in which Clytemnestra’s role
is insignificant: Aegisthus is the sole villain in all these
accounts. He commits adultery with Clytemnestra and
murders Agamemnon on his return at a feast. Agamemnon
dies ‘like an ox at the stall’ (4.535, repeated 11.411). In
revenge, Orestes rightly murders him: the killing marks the
satisfactory end of the story.
Homer creates no sense of a generational conflict. Thyestes’
name is mentioned, but there is no mention of the strife of the
previous generation (known to us from other sources), when
the wife of Atreus was seduced by Aegisthus’ father Thyestes,
and in return Atreus perpetrated an horrific revenge at a
banquet, at which in ignorance Thyestes ate his own murdered
children.
In the second half of the Odyssey, two scenes in the under-
world make more of Clytemnestra’s role and bring us closer to
Aeschylus’ own plot-line: when Odysseus visits Hades,
Agamemnon’s shade tells him that Aegisthus and his own wife
both contributed to his murder; Clytemnestra personally killed
44
3. The Story: Myth and Narrative Technique
Cassandra in the same coup: Clytemnestra’s bigger role here

can be seen as motivated by Homer’s need to have Agamemnon
make a negative comparison between his own faithless wife and
Odysseus’ loyal Penelope.
While in the Underworld, Odysseus makes a comment which
for the first time in Homer’s poem seems to predicate the exact
story-line of Agamemnon:
Alas, right from the start, how much all-seeing Zeus has been a
relentless foe to the House of Atreus, working his will through
women’s counsels. It was for Helen’s sake that so many of us
died, and it was Clytemnestra who hatched the plot against you
her absent lord.
Odyssey 11.436-9; for a similar idea see also 24.24-9
It is all here: a generational saga in which the House of Atreus

figures in a problematic relationship to Zeus. And in fact the
chorus’ words at Ag. 1453ff., as they juxtapose the deadly effect
of the two sisters, seem a reverse echo of Odysseus’ second
remark:
By a woman’s hand (i.e. Clytemnestra) he perished. Alas, fren-

zied Helen, how very many lives you destroyed under the walls
of Troy …
Treatments of the story after Homer and before Aeschylus,

which survive only in a fragmentary form, show developments
in the already fluid story-patterns found in the Odyssey. The
Epic Cycle was a series of post-Homeric poems that formed a
vast sequence covering the entire heroic age. Six of them
appeared to fit round the Homeric poems to create an enor-
mous cycle including events before, during and after the
Trojan War which Homer himself did not relate. The poems
themselves (inferior in quality to Homer’s) survive only in
tiny fragments, but in the fifth century AD they were
summarised in prose by Proclus, and fragments of this work
have survived.2 His epitome of one poem, the Cypria, which
was a ‘prequel’ to the Iliad, provides the earliest source for
45
the information that Iphigeneia faced sacrifice at Aulis. This

sacrifice was brought about by the need to conciliate Artemis,
who had been angered by Agamemnon’s boast that he
surpassed the goddess in archery. However, at the last minute
the goddess substituted a deer and the girl was saved (as in
the story of Abraham and Isaac). This appears to have been
the more canonical version of the story – certainly it is the one
Sophocles uses in Electra and Euripides in Iphigeneia in
Tauris and Iphigeneia at Aulis – and raises the question
whether Aeschylus then made a bold innovation in having
Iphigeneia actually sacrificed.
In the Cypria the figure of Cassandra has developed too: for
the first time she is as she appears in Agamemnon: a prophetess
foretelling the doom of Troy, and Agamemnon’s prize.
In the later archaic period, lyric poets seem to have promoted
Clytemnestra to either joint participant or primary instigator in
Agamemnon’s murder. Tiny surviving fragments from
Stesichorus’ two-book lyric Oresteia suggest this, and Pindar’s
epinician ode, Pythian XI,3 glancingly treats Clytemnestra as
the only murderer, asking the question whether her motive was
anger at Iphigenia’s death or passion for Aegisthus. A beautiful
but baffling painting on the Boston krater, dated to approxi-
mately 470 BC, shows Agamemnon enveloped in a diaphanous
net: Aegisthus kills him with a sword, while Clytemnestra
follows up with a double axe. March4 suggests that this painting
might have been inspired by a now lost poem (possibly by
Simonides) treating the story along these lines. Note the two
murder weapons on the vase. It is interesting that
Clytemnestra’s weapon in Agamemnon seems to vary: at 1262
and 1528 a sword is implied; but at 1149 and 1520 the weapon
is an axe.
As to the site of the murder, we do not know who originated
the shift from banquet to bath, but since the bath was the place
where, in epic tradition, women on their own attended naked
and vulnerable men after their travels, the altered location is
exactly right for a female murderer.
Drawing what conclusions one can from the scant evidence,
46
it seems that by the time Aeschylus came to compose

Agamemnon, he had a variety of sources to choose from. The
story was clearly well known in various versions, but later ones
had increased Clytemnestra’s involvement in her husband’s
murder. Either Aeschylus himself or an earlier lost source had
given her much greater motivation by having Agamemnon
sacrifice their daughter.
Whatever the precedents, however, story tradition was
always fluid, and it would be a mistake to think that Aeschylus
was in any way obliged to follow the most recent casting of the
story. As a creative poet he was bound to the basic outline, but
not restricted to the detail of any previous version.
Furthermore, because he was dramatising it, he needed to find
new ways of treating narrated story material to make it work
for the stage. A similar analogy would be the strategy needed by
a script-writer to adapt a novel for cinema or television. It may
be that Aeschylus was the first dramatic poet to realise the need
for, and fully effect, this different strategy, and the brooding
physical presence of the House, the symbolic walk along the
carpet, the visionary horrors of Cassandra must all spring from
the realisation that the audience can be powerfully affected by
what it sees as well as what it hears. Aeschylus was the first
poet to dramatise this story, and Agamemnon is testament to
his genius at transforming epic narrative into dramatic art.
Suspense
If Aeschylus had followed the almost invariable technique of
Euripides, an opening prologue-speaker delivering a program-
matic speech would have outlined which particular story-line
his play was adopting and – given the extensive history of the
House of Atreus over many generations – where it would end
(the telos or terma). The clear exposition of Hippolytus’
prologue speech makes an ideal comparison. Aeschylus’ narra-
tive strategy works in exactly the opposite way and he
deliberately does neither of these things: his purpose thereby is
to create suspense.
47
Suspense has been variously defined. One usefully simple

definition is ‘an emotion … arising from a partial and anxious
uncertainty about the progression or outcome of an action’.5
Chapter 1 outlined the play’s action: ‘Agamemnon comes home
from Troy victorious, but his wife Clytemnestra kills him.’ That
is the ‘story’ of the play, an abstraction, giving the events in
chronological sequence. As the audience waited for the play to
begin, they undoubtedly knew this story. How then can
suspense be aroused?
Theoretical models, sophisticated enough to account for the
complex ways audiences respond, can supply some insights into
the way this works. The ‘audience morphology’ of Rabinowitz6
makes a distinction between an authorial and a narrative audi-
ence. The authorial audience consists of people with a similar
historical and cultural background to the author himself. They
are thus capable of responding to new elements in a story or to
any contemporary allusion, as well as being able to evaluate the
work objectively as it unfolds (what we might call the ‘sophisti-
cated’ audience). The members of the narrative audience, on
the other hand, are in a liminal state, their own lives
temporarily arrested as they give their attention to the action
before them. This audience temporarily assents to forget what
it knows, responding on a minute-by-minute basis to the
unfolding action, and allowing itself to be unaware of the
outcome (‘naïve’ is not the right word to describe this essential
engagement).
Rabinowitz’s model is of course theoretical only, since real
individuals react simultaneously in both ways. It is not
uncommon to feel objective admiration for a fine performance
while at the same time being involved enough in the drama to
be physiologically affected by it, to the level of altered breathing
and accelerated heart-rate. The value of the model is in
providing a basis for the disturbing and often contradictory
range of effects Aeschylus can arouse, as his audience shifts
uncomfortably from an external to an internal view of events on
stage, from a largely intellectual to a largely emotional
response.
48
The model also makes it clear that within each member of

the audience discrepant levels of awareness are in operation:
they both know and do not know. And Aeschylus makes
discrepant awareness, different levels of knowledge, a key part
of his strategy to arouse suspense. In this play discrepant
awareness operates at every level, between audience and stage-
figures, and between individual stage figures. With their
knowledge external to the play, the audience knows more than
the chorus. Yet Clytemnestra knows more about the future than
both the chorus and the audience: only Cassandra knows more
than Clytemnestra: ironically, this will not save her life.
Discrepant awareness creates rich audience involvement, and
all the abundant varieties of dramatic irony, seen perhaps most
prominently in the exchanges between Cassandra and the
chorus, where irony is by turns grim, grotesque, pathetic, even
momentarily funny.
A different model, this time directed towards the story, can
also aid our understanding of suspense. Russian Formalists in
the early twentieth century initiated a distinction between
story and plot (subsequently reformulated in various
different ways by structuralists and post-structuralists).
‘Story’ consists of the basic story material arranged as a
sequential chain of events (e.g. the action described at the
beginning of the first chapter of this book), whereas ‘plot’ is
the particular ordering and presentation of these events
constructed by a specific author, which may be very differ-
ently cast. They argued that the divergence between these
two elements of story and plot, in the experience of reading or
watching, created an essential effect of ‘defamiliarisation’, by
means of which familiar material became strange and
suspenseful again.
The audience knows in outline the outcome of Agamemnon’s
return (the ‘what’ of the story), but they do not the ‘how’ of the
plot: they do not know which (if any) particular version
Aeschylus is adopting. And Aeschylus deliberately creates a plot
full of subterfuge (via the chorus and his heroine), one in which
the chronological sequence of past events – the only key to the
49
causality of the murder – is wildly obscured and disordered. His

audience is thoroughly ‘defamiliarised’ and kept in a state of
uncertainty to the highest degree and for as long as possible.
Aeschylus planned the three plays of the Oresteia to have a
massive temporal span, extending from the mythical fall of
Troy and the return of Agamemnon, through Orestes’ matri-
cide and pursuit by the Furies to the historical establishment of
the court of the Areopagus. The most obvious connective thread
between the plays is revenge and reversal, and the resolution of
the conflicting claims to Justice (Dikê). But it is typical of his
narrative strategy not to reveal any of this at the outset, and to
begin instead with an ignorant, expendable figure whose sole
function is completed at l.22, once he has seen the signal fires.
Yet the fires indicate victory and Agamemnon’s imminent
return, and so the play gets under way. In the following scenes
this return becomes gradually nearer and more real: the signal
fires described by Clytemnestra become the specific geography
of the Beacons; her words also re-create the fallen city; then the
Herald gives his independent eyewitness report of Troy’s fall.
Then Agamemnon himself returns.
But the slow forward movement is more complex than this.
Each scene in fact has a similar shaping: good news arrives,
bringing Agamemnon nearer; but the positive forward move-
ment is always followed by negative suggestion;7 layers of
discrepant awareness develop, gaps about the past and fresh
uncertainties about the future emerge, and by the end of the
scene there is as much to fear as to hope for. But the stage
figures do not reveal exactly what they fear so much.
Meanwhile there is a thematic unravelling of any certainty
achieved, which we see when the chorus express their confi-
dence in Clytemnestra’s report at 351f. only to fall prey to
uncertainty again at 475f.. All the characters are reluctant to
give a full account of what they know; they merely warn or hint,
then refuse to speak and fall silent. The Watchman begins this
process (36-7):
As for the rest, I’m silent. Big ox on my tongue.
50
Similarly, Aeschylus created a Herald (not present in traditional

accounts) who first of all confirms the fall of Troy with his
eyewitness account, but then inadvertently confirms that the
feared desecration of holy places at Troy has occurred. Only
when prodded by the chorus does he reluctantly reveal more
bad news – the shipwreck of the returning fleet and the disap-
pearance of Menelaus. We may believe, though no character
articulates the thought (least of all the Herald himself), that the
gods are already exacting punishment.
In these opening scenes Aeschylus seems concerned to
thematise the human experience of waiting in suspense, with
all its emotional baggage – hopes and fears supported by a
wealth of unsupported and contradictory conjectures. We have
to remember how well-known the story was, and therefore the
boldness of this strategy of enforcing a return to complete
ignorance.
Aeschylus had made an even more startling, almost perverse,
use of the same technique fourteen years earlier in his histor-
ical play Persians, where, by shifting the setting to the Persian
court, the outcome of the battle of Salamis could similarly
become the object of intense expectation. ‘Waiting for a hero’s
return’ is part of nostos-structure (a plot centred round home-
coming), and the focus on those waiting ultimately derives from
the second half of the Odyssey, but its development in tragedy
is due to Aeschylus. Libation-Bearers too begins with waiting
(Electra waiting for Orestes), and Sophocles and Euripides
imitated this shaping in their own Electra plays. It also occurs
in Sophocles’ Trachiniae and in several Euripidean plays
opening with suppliants desperately awaiting a rescuer (e.g.
Heracles, Andromache and Supplices). Aeschylus’ strategy
here, as in so many other areas, influenced the shaping of many
subsequent tragedies.
Time
Aeschylus’ stage figures as they wait are enmeshed in the
passage of time. Another of Aeschylus’ suspense-arousing
51
strategies is to disturb the natural sequence of events in time,

opening gaps and creating that sense of defamiliarisation. As
described elsewhere,8 Aeschylus uses the chorus to set up a kind
of narrative pendulum which sweeps over the ‘now’ of the
present without stopping: instead we receive tantalising
accounts from different, unconnected points in the past. As the
mighty lyrics of the chorus keep returning us to the past, we
realise that it is only by understanding it that the future could
ever be predicted. Yet we are disabled by the chorus’ self-censor-
ship and incomprehension. Manifestly, some things are too
painful for them to admit, or too well known to mention (1106);
in fact they have a variety of excuses. Ultimately, their knowl-
edge of the past, wide-ranging though it is, fails to predict, far
less prevent, the murder.
From the moment the Watchman sees the beacon, the telos
or goal of the action is set to be Agamemnon’s return, and that
simple return seems to define the temporal extent of the play.
But the chorus then takes us back ten years to the setting out
from Aulis (40ff.) and, within that account, right back to the
earliest time of all, the creation of the world by Ouranos and
Gaia (160ff.). We are returned to the recent past when
Clytemnestra describes the passage of the Beacon fires and
speaks of the previous night’s events in Troy (281ff., 320ff.):
the chorus’ next ode begins with the same recent time as it
contemplates Troy’s fall (351ff.), but soon becomes a timeless
moral reflection before settling on yet another earlier past
time, when Paris seduced Helen and she came to Troy (399ff.).
Now the Herald speaks of more immediately recent events:
the victory (524ff., 551ff.) and the wreck of Menelaus’ fleet
(636ff.): this only spurs the chorus to go back yet further to an
earlier point in time, to consider the figure who named Helen
at her birth (681ff.). Even after Agamemnon enters,
Clytemnestra insists on telling him at length about her ten
years of waiting (855ff.).
The great Cassandra scene in itself has an extraordinary
temporal relationship to the present time of the play, seeming
to take place in ‘arrested’ rather than real time – between
52
Clytemnestra’s exit at 1068 and Agamemnon’s death cries at

1343. Its final effect is to distract us from the forthcoming
murder so successfully that when the death-cries are eventually
heard they come as a surprise after all.
The Cassandra scene has no ‘real’ present time but in it new
narratives from the past and future are at last offered, and into
the consciousness of the play for the first time comes the aware-
ness that the real telos is not Agamemnon’s return but his
murder on his return. The new narrative fragments, delivered
in a context of raw pain and frenzied emotion, alter the audi-
ence’s vision. Aulis and Troy as narrative loci fade away, and
the past now concerns the history of the physically present
House of Atreus: we have at last reached the true epicentre
where the telos is to be played out; for the first time we begin to
understand that this telos is the most recent link in a chain of
Atreid events all played out exactly here. It began with the
primal and original disaster (atê) of Thyestes’ adultery, and has
continued on through the feast of children, creating the
unremitting presence of Furies within. The telos itself, the
murder, is now also perceptible in horrible glimpses, as
Cassandra is possessed by Apollo and foresees it. Cassandra’s
narrative information, snatches of prophetic future time, is
presented in an utterly different way from anything in the
earlier scenes, but Aeschylus’ pendulum approach is still firmly
in operation: seized with her brief involuntary visions,
Cassandra presents her information achronically, indeed repet-
itively, and it is left to the audience to convert it into a narrative
sequence.
At 1279-85 material is at last introduced which could give the
audience an insight into the action of the next play: a new
revenge. The lines are a virtual synopsis of the action of
Libation-Bearers. However, we still have no sense of where the
trilogy as a whole is heading; Athens, the final destination, is
not mentioned at all.
The stream of narrative information that comes into the play
is incomplete, ambiguous, and disordered in time; by returning
to the past, it arouses a high level of emotion about what the
53
play’s outcome will be, but refuses even to define its goal until
the last possible moment. The narrative pendulum keeps
swinging even in the final scene, in which Aegisthus (virtually
deleted from the play until this point), fills in more detail from
the past in his account of Thyestes’ banquet.
Causality, guilt and motive

Establishing a chronology of events is an essential part of the
reader’s or audience’s decoding process, as is the establish-
ment of causality. In both these areas, not unlike a modern
detective novelist, Aeschylus’ strategy is to make uncertainty
proliferate – in this first play of three we should not expect
otherwise – as he manipulates our different levels of knowl-
edge. As for causality, both divine and human, very much is
suggested, of which only a little is finally confirmed. At the
level of theological explanation, Aeschylus indicates many
conflicting divine agents at work: the play shows an overall
descent into confusion (contrast the clarity of the deus ex
machina explanations at the end of the single plays of
Euripides). The chorus in their early Hymn affirm that Zeus is
the supreme causal agent for mortals but at the end, the
triumph of evil makes them feel that the god is at best incom-
prehensible, and at worst devoid of benevolence (1485-7). Only
the role of the daimôn of the House is at all clear.
This section sets itself to answer the question why
Agamemnon is murdered, and teases out three causal strands,
the purpose being to highlight Aeschylus’ deliberately baffling
strategy in this area.
1. Because he sacrificed Iphigeneia

Considered theologically, daughter-sacrifice must inevitably be
an offence liable to punishment. The chorus does not shrink
from describing the horror of the act. However, in this case the
sacrifice was specifically enjoined on him by the goddess
Artemis, as the only means to get the fully righteous expedition
54
under way. To add to the difficulty, Agamemnon has not

offended Artemis by a previous hybristic remark, as in other
versions of the story. There seems no reason for Aeschylus to
have omitted an account of his traditional offence other than to
weaken the causal link between crime and punishment, and
thus make his death, viewed as a divine punishment, less
readily explicable.
Aeschylus presents Agamemnon’s offence against Artemis as
purely symbolic, in the portent of the eagles devouring a preg-
nant hare and, as often remarked, a portent cannot be a cause,
only a sign. He is as yet innocent. Yet a sign might point to
future causes: if so, what are the referents implied by the preg-
nant hare? Innocent Trojans and Cassandra? Iphigeneia? Or,
from the past, Thyestes’ devoured children? The possibilities
cannot be restricted or fully clarified.
On the human level, the chorus’ lengthy description of
Iphigeneia’s death omits all reference whatsoever to
Clytemnestra, suppressing any idea of the obvious motive she
derives from it (perhaps another deliberate obfuscation on
Aeschylus’ part is that their lyrics generally prefer to focus on
her sister). The only possible link Aeschylus allows to be made
between Iphigeneia’s sacrifice and Clytemnestra’s response to
it are Calchas’ veiled words, 152ff., ‘for there awaits a fearful
resurgent crafty housekeeper, mindful Anger, avenger of her
children’. Yet, like the hares in the portent, it is not possible to
isolate a single referent standing behind the personified Menis
(= Anger: personified in the different context of Troy at 699ff.).
They might mean Clytemnestra, or the daimôn of the house.
Over 1,200 lines pass before the play spells out a causal connec-
tion between Iphigeneia’s sacrifice and Clytemnestra’s murder
(1415f., 1430f.).
2. Because he committed sacrilege at Troy

Anxiety that the Greek army, under Agamemnon’s command,
might behave sacrilegiously in the sack of Troy is raised by
Calchas (131f.) and Clytemnestra (341ff.), and the chorus too
55
(461ff.) keeps the subject in our minds. The Herald unwittingly

confirms (525ff.) that the Greeks destroyed sacred buildings.
Then, as the fleet returns, Thracian winds (as at Aulis) blow
adversely, causing a storm in which Greeks, including
Menelaus, either drown or disappear. The storm might seem an
immediate and adequate punishment for the Trojan sacrilege –
there is no more direct reference to it in the play. However, the
subject of sacrilege, now disconnected from Troy, emerges
prominently in the carpet scene. We observe Agamemnon’s
anxiety to avoid trampling his own wealth and his concern to
be treated ‘like a human not a god’. We are perhaps invited to
reflect that as commander-in-chief, Agamemnon, not
Menelaus, bears prime responsibility for any act of sacrilege.
But the motive of divine anger is left vague and, contrary to
tradition elsewhere, no specific gods are spoken of as having
been offended by the sack.
3. Because he comes from an accursed family

No member of a previous generation is named in the earlier
part of the play. Only in the strongly generational terminology
of the chorus’ ‘old story’ about hybris in a family one day ‘beget-
ting’ new hybris (750f.) do we sense a fearful reference to
pre-existing motives for murder in the House of Atreus. But
then in the Cassandra scene the narrative kaleidoscope is
shaken, and new visions of past family members vividly appear
with motives for revenge: Thyestes’ children (and the Erinyes
singing of his adultery). Cassandra sees these visions discretely
but, in prophetic context, causally connected to a new one:
Clytemnestra murdering Agamemnon. The House is steeped in
continuing acts of bloodshed. But the chorus, while accepting
that her knowledge of the family’s past is accurate, simply
cannot connect it to Agamemnon’s forthcoming murder and are
blind to Clytemnestra’s role in it. We learn that Cassandra’s
inability to enlighten them is exactly the form Apollo’s punish-
ment takes. In fact they have just about made the causal
connection (1338f.) – but it is too late.
56
In the final two scenes of the play, guilt, agents and motives are
overtly thrashed out. To summarise, Clytemnestra first claims
the work is all her own, in revenge for her child. So (1) is relevant.
But she rejects the chorus’ idea that Helen had any part in it,
which perhaps means that she also rejects the idea that sacrilege
at Troy (2) is significant. However, when the chorus suggests that
the murder was also perpetrated by the daimôn or alastôr of the
House (3), this is a double causality she can accept. Restricting
her sights to the triangle of herself, Agamemnon and Iphigeneia,
she hopes to draw a line under events and make the daimôn
depart, an idea which the chorus rejects. They have a bigger
perspective and now understand a continuing, generational
cause for bloodshed in the house with Orestes as spearhead for
the future. This final viewpoint is of course confirmed when
finally Aegisthus spells out his own familial motivation.
In conclusion, several stories in Agamemnon interlock, with
varied perspectives which are not all made available to the audi-
ence until after the murder, though all culminate in his death.
Causal relationships are extremely complex and ‘ends’ turn out
to be ‘beginnings’ or, unexpectedly, the outcome of things long
past. Everything points two ways, both to the past and to the
future. Cassandra’s reference to Orestes indicates that even
Agamemnon’s death – the result of the curse, the fulfilment of
Calchas’ warning prophecy – will not end the story. At any rate,
the whole subject of Troy as a cause of his murder is done with
by the end of the play: the central theme of family revenge has
been revealed.
Narrative voices
All earlier versions of the Agamemnon story were in epic or
lyric form and so were third-person narratives (even if the
‘embedded’ characters sometimes spoke directly). In a very
natural way, the narrator’s voice was omniscient and his focali-
sation external to the story: ‘Homer’ could relate a council on
Olympus with as much authority as a domestic scene inside
Troy, or the passage of a spear through a body.
57
Drama, however, is differently constituted. It deletes the

third-person narrator (though sometimes allowing the chorus
this role on a limited and temporary basis) and with it the
natural sense of objectivity and omniscience that the narration
of past events acquires. All the focalisation is internal. In drama
there can only be different voices and varying perceptions; situ-
ations and events can never attain a final objectivity.
The tragic dramatist has a theoretical choice between
furthering his plot through action or through a narrative voice.
In fact, although scenes of violent verbal confrontation and
extreme physical pain and grief are common enough in the
Greek theatrical tradition, battles, murders or suicides are
conventionally narrated rather than shown (Ajax’s onstage
suicide is exceptional). What stage action there is takes the
form of ritual activity – the performance of lamentation, suppli-
cation, offerings or prayer – or is essentially symbolic rather
than effective: in Agamemnon, the Herald kisses the earth
(508), the chorus make ineffectual physical threats against
Aegisthus (1650f.), Cassandra flings away her prophetic
insignia (1264f.), Agamemnon famously walks on the carpet
(957f.). Because effective action is narrated rather than being
visibly acted out, and because there is no objective narrator, the
dramatist’s choice of narrating voice throughout his play has
crucial importance for irony, suspense and discrepant aware-
ness.
In general, voices in Aeschylus tend to communicate in
formally structured and self-contained speeches. As
Rosenmeyer remarks, ‘Aeschylean speech is self-absorbed,
isolated, marked off from what precedes and what follows by a
gulf of silence’.9 This is particularly noticeable in the central
scene. Agamemnon enters and speaks at 810, only responding
to the chorus’ address after twenty lines. If Clytemnestra is on
stage, he ignores her throughout. When she begins to ‘reply’ at
855 (an inappropriate word when there has been no interac-
tion), she markedly addresses the chorus, not her husband, for
twenty-one lines. Only when she invites him to step down from
the chariot do they begin to engage verbally.
58
Voice of the chorus

The choral voice in any play functions in an interestingly hybrid
way: they or the koryphaios relate to the stage figures ‘realisti-
cally’ during the episodes, but also form a non-naturalistic
continuo in the stasima between them. While they sing, the
stage-time of the play becomes indefinite, and this fact,
together with their group memory and general moralising,
seems at times to put them on a different narrative level from
the main action. But the choral voice is never ultimately omni-
scient, nor can the choral persona, in the manner of ‘Homer’,
remain unaffected by the action. In fact the chorus resembles
the external audience in being forced to suffer passively the
actions performed by the stage figures.
These contrastive choral tendencies were grist to Aeschylus’
mill. On the one hand he gives them a knowledge and an intel-
lectual capacity which make them heady narrators of the past
in the earlier part of the play, while on the other hand he
dramatically heightens their conventional inefficacy by charac-
terising them as aged and physically helpless, talking heads
without effective bodies. He also gives them the emotional
burden of waiting, the play’s major concern until Agamemnon’s
actual return at 782. Their voice creates the play’s prevailing
atmosphere of fearful suspense.
Choral odes get successively shorter as the pace accelerates,
until between the exit of Cassandra and the murder of
Agamemnon there is no space even for a vestigial ode, only a
few chanted lines. However, they begin with the longest ode in
surviving tragedy, in which they reveal themselves as the play’s
major narrators of the past. They stress their authority in that
role (104) and demonstrate a capacity to reflect on the complex
Olympian powers, particularly Zeus, at work in human affairs.
But as humans, their theology is inevitably deficient. They can
say only that it was ‘some Apollo or Pan or Zeus’ who heard the
outraged vultures (55-6). They may grant themselves authority,
but lack predictive powers, and do not narrate even the past
without leaving significant gaps. This past they open up feels
59
dangerous, and their narrative skirts it: they retreat into

expressions of fatality, and mingled hope and fear.
The entry anapaests (40-103) and parodos (104-263) will be
used to exemplify their idiosyncratic narrative voice with its
flood of memories, self-suppression, sudden temporal shifts and
unmarked movements from specific to general (even leading to
what Goldhill calls the ‘erasure of meaning’10). Their great rich-
ness of language, imagery and theological reference will be
considered in Chapter 5: my concern here is to show how the
choral voice behaves when it narrates the background facts of
the play.
They begin by describing the setting-off of the punitive expe-
dition against Troy, a cause ratified by Zeus Xenios (God of Host
and Guest). After eight lines the narrative is interrupted by a
simile of ten lines (49-59, excluded from discussion here), but
continues after it to 67: they are still describing the expedition
in the most general terms. At this point they suddenly break off,
expressing a sense of mingled futility and fatality: ‘That’s how
things stand now. Fate is coming to fulfilment. No sacrifice of
any kind will assuage intense anger’ (67-71).
We assume the anger is divine, but have as yet no cause to
assign it to: we wait to learn. But the narrative has been
thrown off course and is not continued. For the next eleven
lines (71-82) the chorus only describe their own extreme
debility: they are little more than dream figures wandering in
the daylight. Now they ask the (offstage) Clytemnestra what
news has arrived to produce the sacrifices they see around
them. Quite humbly they ask her to be ‘healer of this anxiety
of mine, which is sometimes malignant, but then from the …’.
The text is then irrecoverable.
Thus end the entry anapaests: the chorus ceased entirely to
narrate at 67 and finally resorted to asking for another’s narra-
tive – which was denied them.
The parodos opens with the topics of the anapaests restated
in greatly expanded form. They reassert their lofty authority
(104-6); the vultures simile is replaced by another ornitholog-
ical phenomenon, the portent of the eagles and pregnant hare
60
(109-20). The ritual cry that ends the strophe, ‘Cry sorrow,
sorrow, but let what’s good prevail’ (a refrain to be repeated
twice more), mirrors the closing of the anapaests in the way it
strains for a balance between positive and negative possibilities.
Their voice expresses the fearful yearning of this whole section
of the play.
Their alternation of good/hope and bad/fear continues into
their description of the portent, which is both ‘propitious’ and
‘faulty’ (145). In fact two dangers emerge: sacrilege at Troy
(131-3) and adverse winds causing the need for the sacrifice of
Iphigeneia (146-52). Calchas’ quoted words are moderately
clear until we reach the densely allusive language of 152-5 (see
p. 55), which may contain a veiled reference to Clytemnestra
(though we could understand the lines more generally as refer-
ring to the daimôn of the House). At any rate, the chorus’
reportage begins to reveal, though not with clarity, that present
and past are connected together by a motive of revenge.
At this rich moment in the Aulis-narrative, when we desire
clarification, the chorus instead break off to sing their so-called
Hymn to Zeus (160-83; ‘so-called’ because Zeus is not strictly
addressed; this is a meditation, not a hymn). One effect is quite
suddenly to raise the action of the play to the highest religious
and philosophical level. To paraphrase, the chorus says: ‘Zeus is
unknowable, but if I am to lose my anxiety, it is only to him that
I can refer. … He has put us mortals on a path so that we learn
through suffering’ (this is the first statement of the pathei
mathos theme, the idea of learning through suffering, which
lies at the heart of the entire trilogy). ‘But pain that recalls
suffering drips before my heart where sleep should be. Though
wisdom comes even to the reluctant, the favour of the gods is
surely a forceful one.’
Suffering (pathos, ponos), resistance, and knowledge
(phronein, mathos, sôphronein): the chorus’ voice longingly
searches for a pattern of human wisdom and understanding
derived from Zeus. At the same time we are made aware of their
human frailty, since their noble thoughts coincide with their
bodies’ experiences of insomnia and heartache.
61
How does the chorus’ credo, their affirmation of faith in

Zeus’ plan for mortals, square with the Aulis narrative that
then resumes? Critics agree that, at this point in the trilogy, it
can do little more than underline the high significance and
hugely problematic nature of the choices that humans must
confront when faced with conflicting obligations, for Calchas’
worst fears are confirmed, the wind is adverse, and
Agamemnon is faced with either abandoning the expedition
enjoined by Zeus, or of commanding the sacrifice of his own
daughter enjoined by Artemis. In two distinct strophes the
chorus relate first Agamemnon’s thoughts as he weighs each
side of the possibilities before making his decision (205-17), and
then describe the hardening of mind that follows it (218-27).
The king’s unshakeable desire to commit an abomination is
heightened by the vivid description (228-47) of his young
daughter, with her mouth gagged and her dress slipping off,
held over the altar like a sacrificial goat to have her throat cut.
Her eyes, though, are eloquent, and the chorus remembers how
she used to sing at her father’s feasts: once this was a loving
relationship. Here the chorus, with steady eyes, does not shrink
from giving us the full measure of the girl’s human worth, and
the brutality and injustice of her innocent death. Their narra-
tive leaves her poised over the altar – a brilliant symbol of all
the innocent deaths to come. The pregnant, equivocal phrases
with which they ‘shut down’ the narrative are by now familiar
characteristics (248-55):
As for the rest, I neither saw it nor speak of it. However, Calchas’
craft is not without fulfilment. Justice tilts down her balance
scales for those who suffer to learn. No mourning before
anyone’s dead. It will come clear … As for the rest, may it turn
out well.
This parodos is for many the greatest lyric in extant tragedy,

and it is hard to do justice to the achievement of the choral voice
here. By its statement of the pathei mathos theme it lays the
ground for the whole trilogy. It shows in detail how an ethical
62
dilemma arises in which a choice must be made between alter-

natives, the omission of either of which will create appalling
consequences. The choral voice explores this dilemma at length
and in detail, considering it both on the theological plane, as the
result of the contingent intersection of two divine demands, and
also humanly, from the inside of Agamemnon’s head. The choice
they describe is an extreme one, nonetheless it bears the hall-
mark of real complex choices in human life: Agamemnon can
freely choose (in the sense that he is not ignorant of fact nor
forcibly compelled into one direction rather than another), but
he is also bound by necessity, since the choice is not entirely
open but only between two dreadful alternatives. The chorus
gives the measure of how evil the alternative is that he has
chosen as they describe Iphigeneia’s death with a compas-
sionate humanity which spells out its horror; having already
criticised Agamemnon’s changed mental state after making his
decision, they now seem to sense that the act deserves punish-
ment. (We might have expected the sacrifice to be carried out
with reluctance, but Agamemnon now desires it; see Euripides
Iphigeneia in Aulis for the portrayal of an Agamemnon who
reacts to his choice very differently.)
All this the chorus have not only narrated, but narrated in
character – old, frail, wise, fearing the worst yet longing for the
good. Their story and their emotions, their humanity, work
powerfully on the audience, stirring both its intellect and its
feelings and engaging them in the ethical conflict. Nussbaum
writes, ‘the chorus look, notice, respond, remember, cultivate
responsiveness by working through the memory of events. The
presentness of the chorus … and their patient work, even years
later, on the story of that action reminds us that responsive
attention to these complexities is a job that practical rationality
can, and should, undertake to perform ….’11
Voice of Clytemnestra
The chorus narrate the past, supplying voices, emotions and
reactions of the characters involved, as well as their own.
63
However, their male voice has no power to narrate the present

or immediate future. This is the function of the female voice of
Clytemnestra, to whom, reluctantly, they are both politically
subject, and dependent upon for information. It is an unequal
contest: while they claim divine powers of persuasion (105ff.), it
is Clytemnestra who ultimately convinces. Much of the play
consists of the uneasy clash of these unbalanced, differently-
authorised, differently-gendered voices.
After her significantly delayed initial entry (cf. Euripides
Medea for a differently-structured but equally sinister delay),
the chorus greets her with heavy flattery (255ff.). In the
stichomythia, while their emotional reactions to her news
predominate, Clytemnestra is cool and hostile. ‘Do I speak
clearly?’(269), ‘It is of course the case, unless the god be
deceiving us’ (273), ‘You find fault with my intelligence as
though I were a young child’ (277).
The Beacon Speech (281-316) is delivered as the proof the
chorus requires. It and the following speech (320-50) now give
a first startling display of her voice’s power to engage and
persuade, as it presents a succession of vivid pictures to serve as
detailed proof. This is a significant prelude to her later speeches
of deception. Here, Clytemnestra’s ‘knowledge’ of distant
events is logically inexplicable and surely meant to be so, and
Fraenkel’s rationalising idea that her information is not ta
genomena (‘what actually happened’) but hoia an genoito (‘the
sort of thing that happens’) misses the point: Aeschylus delib-
erately undermines a logical foundation, leaving the voice to
manifest itself in all its eloquence and power.
At the same time, however, it is a voice that even at this
stage begins to give itself away. Like Calchas, she raises the
idea that Greek sacrilege at Troy could turn victory into
‘defeat’ (helontes authis antheloien an, 340). At first, this
reference might point forward to the shipwreck that overtook
Menelaus’ contingent, which the Herald will describe. But now
her precise syntax becomes obscure and the meaning blurred:
‘[I fear] lest some desire (erôs de mê tis) overtake the army,
overcome by gain, to sack what it should not … but if the army
64
comes home without a crime against the gods, the suffering of

the dead might be aroused, if it doesn’t encounter unexpected
misfortune’ (341-7).
The opening elision of the verb of fearing leaves strangely
prominent the noun ‘desire’, erôs. Despite a similar prominent
erôs at 540, we are surely roused to ponder on what suppressed
desire of the narrator might be hinted at here? And what does
she mean? As Denniston and Page analyse the lines, she seems
to be raising three dangers to the Greek army: sacrilege at Troy
against the gods; misfortune arising from the spirits of the
Trojan dead; and random disaster. It is possible, however, to
understand of the last two items that it is the suffering of the
dead Iphigeneia that might be aroused, and that the ‘unex-
pected misfortune’ might be Agamemnon’s murder.
She ends with ‘Such things you hear from me, a woman’, to
which the chorus can only respond, ‘Lady, you speak with good
sense like a sensible man’ (351). There is grim humour here in
this momentary truce between the differently-gendered voices.
For now, the chorus believes that Clytemnestra has given
‘reliable evidence’ (361). But by the end of their song they are
in doubt again about the reliability of women (475-87). In the
Herald scene, Clytemnestra delivers a single speech (587-614),
the first eleven lines of which are devoted to self-justification.
In the twelfth line (598), the question of Agamemnon’s return
– the only topic so far discussed with the chorus – is dismissed.
The mood of her voice alters, and her peremptory order (604)
makes it is clear that what follows is at least partly for the
Herald to report back to Agamemnon.
The voice now adopted is that of a woman excitedly waiting
to welcome her victorious husband – but it continues the tone
of self-justification. Furthermore, its eloquence quickly ceases
to ring true: by any standards, such protestations of utter
loyalty, consistency and chastity are excessive and immodest.
She herself points up the paradox by describing her words as, ‘a
boast, bulging with truth, not shameful (aischros) for a woman
to utter’ (613-14), and Rosenmeyer correctly, I think, finds not
so much dissimulation as aggression here.
65
The same tone is continued in her response to Agamemnon’s

speech; a curious blend of sarcasm, defiance, resentment and
self-extenuation. The formal opening address to the chorus
(855) creates the sense of a public defence-speech before a jury.
She will not feel shame (ouk aischunoumai) to use her voice in
public to speak of her philanoras tropous (a phrase of splendid
ambiguity, meaning both ‘husband-loving’ and ‘man-loving’ (in
the sense of ‘adulterous’) ‘ways’. Her subject from 858-94 is her
life in Agamemnon’s absence. As Agamemnon himself is to
comment, it is remarkably long, a tour de force for an actor,
fluent, excited and marked by much grim humour, heavy irony
and wild exaggeration. However, at the point of accounting for
Orestes’ removal to Phocis, we get the sense of careful prepara-
tion: two plausible reasons for it are carefully ascribed to the
absent Strophius, and she ends, ‘now that is a justification
which conveys no deceit’ (886). But the very denial of deceit
raises the thought, qui s’excuse s’accuse: the more she exoner-
ates herself, the more suspicion falls on her: her lies draw
attention to themselves as lies. As the speech continues, the
exaggerations become more incredible: after 895 they are
turned on Agamemnon himself: he is perceived in a fast-flowing
stream of images as a guard-dog, a ship’s forestay, a house-
beam, an only son, land to sailors, sun after storm, water to a
traveller.
By the end of this speech Aeschylus has fully displayed the
unstoppable, deceitful weapon that is Clytemnestra’s voice – an
extraordinary and original creation on the playwright’s part. In
the stichomythia of 931-43 she uses it in single combat to defeat
her victorious husband. The vocabulary here (machês, battle,
940, to nikasthai, defeat, 941, nikên, victory 942) makes it clear
that she is forcing a battle of words. Each line she delivers is a
calculated thrust, which eventually he cannot parry.
As he steps over the carpet (958-74), her voice takes on an
even more sinister tone. She dares to compare the infinite
resources of the sea to the wealth of the House (the chorus have
already delivered warnings on the subject of wealth, 381ff., and
more relevantly, 750). Then in three extraordinary swift synaes-
66
thetic images, Agamemnon is compared to a vine giving shade,

to warmth at the hearth in winter, to cool in the house at the
time the unripe grape (Iphigeneia?) is harvested. The final
three lines are climactic and play on the multiple meanings of
the verb teleô (= fulfil; bring to an end; pay one’s dues; perform
a sacrifice), and the adjective teleios (= of sacrificial victims,
perfect or unblemished; of actions, resolved upon, accom-
plishing or fulfilled; of a man, one who has full power or
authority; of a god, one who brings to fulfilment).
…when a teleios man walks the house. O Zeus Teleios, telei my

prayers. May you telein what you intend.
Thus in only a slightly covert way, Clytemnestra blasphemously

prays for her husband’s death to his departing back.
It seems appropriate to end this section on narrative voice by
observing that, quite apart from those of the stage figures, the
play is full of a vast variety of embedded voices, human, myth-
ical and animal, largely conveyed by the chorus. We hear the
voiced thoughts of Agamemnon as he decides to sacrifice his
daughter (206-17). Other voices prophesy (Calchas at 126-55
and 201ff., the man who named Helen, 681ff.). The voices of the
populace are heard as they deliver a eulogy for their dead (445),
or express resentment over the war (449, 456-7, 883). Vultures
scream (48, 57-60), a raven sings tunelessly on Agamemnon’s
body (1472), Cassandra laments like a nightingale (1142).
There is even a gnat that whines (892). Different song types are
frequently mentioned within choral song – usually to make the
point that such a song is inappropriate or undesired: the
Trojans’ wedding song for Helen turned into a lament (707ff.);
the Furies are a chorus drunk on blood singing an ugly song in
the rafters of the House (1186ff.); the chorus’ fear is similarly
presented as an unwanted song ‘they didn’t hire’ (amisthos
aoida, 979), going on endlessly inside their bodies. These
multiple voices vary in significance, some merely adding a
temporary aural depth. Others have much more significance: in
particular the range of prophetic voices, coming from different
67
points in the past, help drive the action on towards inevitable

disaster. As a whole, they greatly contribute to the richness and
ambiguity of the play.
68
Gods and Humans
It’s the gods I beg for release from these sufferings

Watchman, 1
Gods
The problem of suffering is timeless, and resonates across
different religions and different historical epochs. Agamemnon
starts with the Watchman saying, ‘I beg the gods for release
from suffering’. His longing for help from the gods is the
emotional starting point for a theological questioning which
drives right through the trilogy. Less than two hundred lines
later, the chorus respond to the Watchman’s prayer when they
make the tangential but related proposal that suffering has a
divine purpose. Zeus has ordained for human suffering to be
more than a meaningless passive ordeal: it is rather a process
leading to learning (176-9, repeated 249-50). So through
suffering, they suggest, will come the Watchman’s longed-for
release (though the audience have probably already forgotten
him) in the form of progress into a better state of affairs.
This is the famous pathei mathos, ‘creed’ of ‘learning
through suffering’, as it is usually translated, though it is worth
saying that pathos can also mean ‘experience’ (passivity as
opposed to activity; being done to rather than doing) as well as
outright physical suffering, while mathos means both the act of
learning and the understanding or wisdom that results from it.
The ‘creed’ is only ten words long, and the chorus leave as an
open question whether the suffering can be merely observed or
has to be directly experienced in order for learning to take
69
place. Through observation, pathei mathos invites the audience

to think that they may learn something valuable from watching
the play. Through direct experience, within the play-world of
Agamemnon itself it is hard to find that the suffering characters
have acquired any useful knowledge: the chorus’ realisation of
the true state of affairs comes too late to prevent the king’s
murder and the palace coup; Agamemnon dies too quickly;
Cassandra had nothing to learn because she knew it all already;
mathos does not provide her with a means of escaping a horrible
death. In fact, the form of her punishment by Apollo thematises
the whole problem of mathos in the play. If, in her final pitiful
lines (1327-30), she expresses awareness of a new under-
standing of the human condition, it is a very bleak lesson.
Nonetheless, the powerful idea of pathei mathos remains in the
play’s ether, waiting to find resolution.
Agamemnon must be taken as a serious religious drama. It is
full of gods, whether Olympians such as Artemis, Apollo and
Zeus, or daimones (lesser divinities). Human characters
constantly attempt contact with them: Clytemnestra sets up
sacrifices through the city, the Herald and Agamemnon both
pray on their return; Agamemnon attempts to propitiate
Artemis; Aegisthus claims the gods have acted on his side. As
prophet and prophetess, Calchas and Cassandra in their
different ways both have special relationships with Apollo. Gods
receive a stream of invocations, supplications, propitiations.
The chorus, quoting Calchas, explore the meaning of the god-
sent omen, they describe the human sacrifice of Iphigeneia to
Artemis and, in their odes, lengthily ferret through past and
current events in an attempt to understand the divine forces at
work in them. Gods, although unseen on the stage until
Eumenides, densely inhabit the thought-world of the play. They
are identified as the underlying cause for everything that
happens; they inspire fear, but are also the only hope that
things might get better.
Greek religion was not inherently of a kind to support an
optimistic world-view. It was a blend of polytheism and
monotheism. Many gods existed, but Zeus was preponderant.
70
4. Gods and Humans
There was a deeply-rooted belief that the gods were the givers
of both good and bad. They demanded worship (timê) from
humans, and when denied due respect (aidôs, sebas) bore a
grudge (phthonos). The individual, or his family, would be
punished. The gods, however, did not interfere from the outside
with the course of nature (by e.g. visibly overthrowing cities),
but worked through the world’s natural processes, both inani-
mate and animate, including human agency and human
passions:1 however intricately brought about, Aeschylus makes
it clear that the gods work through Clytemnestra to bring about
the punishment of Agamemnon.
Greek religion was not anthropocentric. Man was not created
by god(s) nor of central concern to them. Zeus, though ‘the
father of gods and men’ was not a benevolent ‘father in heaven’;
gods governed the world in their own interests. However,
perhaps by analogy with early kings, Zeus had acquired the
function of administering justice in the world. This was
conceived of as the maintenance of a kind of balance. Action
was followed by reciprocal action and ‘the doer suffers’ (1564),
pathein ton erxanta; pathei mathos is a development of this
traditional view. It is important that justice (dikê) had, from the
time of the earliest natural philosophers, acquired the extended
meaning of ‘the order/balance of the universe’: dikê in this
sense perhaps underlies the Watchman’s opening description of
the regulation of the seasons by the rising and setting of the
stars.
Something like this was the religious belief of Aeschylus and
his audience. Unlike other religious traditions (such as
Christianity), no canonical texts existed and there was no body
of faith requiring absolute belief. Priests performed sacrifices,
prophets prophesied and seers interpreted omens; they did not
preach doctrine. No barrier stood in the way of attempting to
make sense of the world by rational explanation. ‘What
happened in the world depended ultimately on the gods, and
their purposes were usually inscrutable to human minds; that
did not mean that it was irrational, but that the reasons that
governed it usually remained mysterious’.2 If the Eleusinian
71
mysteries and the cult of Orphism offered their initiates a

significantly different body of belief which included a mystical
afterlife, that did not prevent Greeks from looking at their gods
with their eyes wide open.
In Agamemnon, Aeschylus is clearly not reflecting contem-
porary religious practice in any direct way: after all, the play is
set in Argos but makes no reference to the important cult of
Hera there.3 His apparently idiosyncratic theology is in essence
inherited ready-made from existing poetic tradition, which had
regularly depicted the world as a place where humans co-
existed with dense numbers of unseen gods and forces. Poetry’s
function of exploring the relationship between gods and men
had been well established by Homer and Hesiod: given that
philosophers and theologians in our modern sense had not yet
come into being (together with new prose genres of philosophy
and theology), poetry is the natural medium for the most
serious religious and philosophical reflections of the age. The
Oresteia converts that poetic project of religious and intellectual
exploration into superb drama.
Although Agamemnon shows the influence of both Homer’s
and Hesiod’s religious concepts, this discussion focuses on
Hesiod. In the ancient genre of ‘wisdom literature’, Hesiod had
written two didactic hexameter poems dating from the second
part of the eighth century, Works and Days and Theogony.
Works and Days (mostly moralistic farming advice) is addressed
to the poet’s brother Perses, who is involved in a law-suit, and
the poem contains a running opposition between hybris and
dikê (‘insulting behaviour’ and ‘justice’): this opposition,
heavily modified, finds its way into Agamemnon. Theogony
(‘Birth of the Gods’) includes an account of the two generations
of gods before Zeus gained ascendancy. It is a terrible story of
castration, child-eating and parricide (features shared by the
House of Atreus). Hesiod, however, draws a line under the past
and firmly tells his audience that now Zeus controls the
universe in accordance with Dikê, and ‘it is not possible to go
beyond or to deceive the will of Zeus’: the chorus of Agamemnon
take very much the same line at 160-75. In both Hesiod and
72
4. Gods and Humans
Aeschylus, Zeus’ supreme power in the universe – over humans

and other gods – is stressed as a hard-won evolution that will
now endure forever.
Daimones
Hesiod’s world is full of daimones. These are lesser gods, which
we would now categorise under separate headings (such as
‘concepts’, ‘personifications’, ‘drives’ or ‘forces’), or allocate to
different disciplines (physics, biology, ethics, psychology, mere
superstition). Daimones might be children of Zeus but might
also predate him, offspring of earlier, primeval gods such as
Earth and Darkness. They have a powerful and often negative
effect on human life, ‘For earth is full of evils, and sea is full of
evils’.4 Zeus himself has ‘thirty thousand spirits, watchers of
mortal men, and these keep watch on judgments and deeds of
wrong as they roam, clothed in mist, all over the earth’.5
Aeschylus’ play-world is similarly inhabited. We find for
example: justice (dikê), ruin (atê), excess (koros), insult (hybris),
Furies (Erinyes), envy (pthonos), anger (mênis), sleep (hypnos),
beauty (charis), hope (elpis), persuasion (peithô), the driver-
astray (alastôr), as well as simply, ‘the/a daimôn’.
The concepts here have been inconsistently capitalised,
reflecting the fact that in Aeschylus they are not consistently
personified entities. The list is not exhaustive: the co-existence
of wealth (ploutos) and prosperity (olbos), given serious sepa-
rate consideration by the chorus (381ff. and 752ff.) despite their
apparent similarity,6 indicates human incapacity either to
systematise daimones or to restrict their proliferation.
Giving a one-word English translation for these concepts
risks ignoring their fluid nature. Each term merits scholarly
analysis, and some few (hybris, atê and peithô, for example)
have received it. The results of research yield valuable insights
into the ways the Greeks understood themselves in relation to
gods and to natural/universal processes.
They reveal very different habits of thought. As a simple
example, Sleep was understood as a daimôn or external force
73
that assailed an individual from outside, (rather than as the

result of an internal system shut-down, as we currently under-
stand it). Homer describes Sleep as the ‘twin brother of Death’,
presumably because both are states of unconsciousness and
may look similar to an observer. An Attic red-figure calyx krater
by Euphronios (c. 515 BC) depicts a scene from the Iliad in
which the brothers, fully armed but also bearing wings, are
about to lift the body of Sarpedon out of battle following Zeus’
instructions (at Iliad 16.667f.).
Sleep, then, has the capacity to be both external and internal,
to be a living force or personification (to the extent that it can
be depicted with armour and wings and have kin in the shape
of a brother, Death), but it can also be treated as a bodiless
abstraction in the modern manner. Other daimones have
different kinds of fluid qualities: for example, hybris, ‘insulting
behaviour’, carries the general sense of a transgression of the
boundaries which divide man from god or equally, man from
man (i.e. hybris can be either a theological or a social offence).
Hybris in Athens was also a legal offence with a concrete
meaning equal to ‘assault’. But studies show that the term can
be used in a purely interior context without any implication of
action at all: hybris can be committed merely by an attitude of
mind. Atê too is similarly both external and internal, the same
word embracing both the mind-set that precedes ruin (folly or
delusion sent by the gods), as well as ruin itself, the word which
most inadequately translates it.
Daimones were ‘good to think with’ – ethically, psychologi-
cally, politically and scientifically. Poetry often rationalised by
speaking of them in generating groups: one daimôn ‘begot’ or
‘engendered’ another in a causal sequence (Gilbert Murray
referred to this phenomenon as ‘the inherited conglomerate’).
For example, describing the power of the wealthy in Athens, the
archaic poet and law-giver Solon (archon 594/3) had produced
the famous formula, ‘Excess engenders Hybris, when much
Prosperity accompanies men whose minds are not adequate’ (fr.
5.9-10). Expressed as a formula, we have the statement: Excess
+ [Prosperity and inadequate men] = Hybris, and every Greek
74
4. Gods and Humans
child knew that the next step was Hybris + something else =
Atê: later sophisticates were to see such statements as hardly
more than platitudinous old saws. Nonetheless, the formula
encapsulates generations of observation of human behaviour,
and there is wisdom here.
In Agamemnon Aeschylus’ elderly chorus are repositories of
this already ageing tradition and many of their more arcane
utterances become less baffling once we appreciate the typical
way their thought is expressed. For example at 381-6 they sing,
‘there is no defence of Wealth against Excess for the man who
has kicked the altar of Justice into invisibility: patient
Persuasion forces him, insufferable child of Doom who-plans-in-
advance’ (381-6). We might paraphrase, ‘If a man does not use
his wealth with justice, slowly but surely he will be persuaded
into some disaster.’
The extended meditation on the ‘offspring Prosperity begets’
(751-71, bedevilled by a corrupt text from 766 onwards) also
deserves attention: ‘there’s an old story among mortals that
when a man’s Prosperity has come to full growth it bears
offspring (interest) and doesn’t die without issue; it says that
insatiable sorrow grows from good fortune in a family. But I am
single-mindedly different from other people. For it is the
impious act which begets after it more impious acts resembling
their own kin – the destiny of houses that judge rightly is to
have lovely children. But Hybris is wont to engender new
Hybris among evil mortals, at some time or other, whenever the
appointed day of interest (birth) comes round, together with an
irresistible, unconquerable daimôn, unholy Boldness of black
Ruin for the house – children like their parents.’
Here the chorus (consistently conflating the language of
‘begetting’ with the monetary sense of ‘getting interest’)
debates whether prosperity in itself leads to sorrow (by some
law of natural reversal), or whether a family’s prosperity might
not continue forever, provided it behaves with justice (their own
preferred view). Hybris, however, once it has come into exis-
tence, engenders further Hybris leading eventually to an
inevitable disaster. These reflections are positioned strategi-
75
cally: delivered just before Agamemnon returns, they could not,

with all their teasing ambiguity, arouse more speculation about
the king’s potential standing with the unseen powers.
Like his daimones, Aeschylus’ Olympian gods are also
complex conceptions. The yin and yang of Aphrodite and
Artemis in Euripides’ Hippolytus are very simple by compar-
ison: at Hipp. 1328ff. it is even carefully explained that,
although gods may get even by punishing the other’s favourite,
their custom is to avoid direct opposition: all of them ultimately
respect Zeus. There is no such neat regulation in Agamemnon
where, on the one hand, Zeus Xenios (God of Host and Guest)
despatches the sons of Atreus to punish Paris’ theft of Helen in
a righteous expedition (60ff.) but on the other, Zeus’ daughter
Artemis, quite independently angered by the prospect of inno-
cent young lives to be lost, causes opposing winds to blow which
will only die down – illogically – on the loss of another young life
(the sacrifice to her of Iphigeneia). Why does Aeschylus’ Zeus
not intervene to curtail his daughter’s actions? This topic has
already been considered in the light of Aeschylus’ strategy of
obfuscation, but it is also a theological problem deliberately left
in all its baffling inscrutability.
Just a little clarification comes towards the end, when the
alastôr or daimôn in the House is openly debated by the chorus
and Clytemnestra. They agree that it has played a part in the
murder, but while Clytemnestra now hopes to make it depart
(1568ff.), the chorus state that ‘Dikê is sharpening her sword
for another deed of blood’ (1535-6) and they hope the daimôn
will bring Orestes back.
Here, however little we feel that Clytemnestra’s hopes can be
fulfilled, it is nonetheless significant that divergent possibilities
are presented. This is not at all in the manner of Hesiod: despite
some puzzling lessons in his stories of Pandora and the Five
Races of Men, by the end of Theogony the world has apparently
completed its processes of creation and change: all that seems
required of humans, as in Works and Days, is obedience to a
closed system. By contrast in Agamemnon, the world is
presented as still unknown, but open to conjecture and the
76
4. Gods and Humans
possibility of change. Characters subject the gods to their best

intellectual efforts: nothing is fixed or certainly known, but
some new mathos may emerge. Pathei mathos contains the posi-
tive idea that by trial and error, by getting to grips with this
chaos of conflicting powers, rational understanding can take
place – and man can improve his lot.
Zeus and Dikê

Elsewhere in surviving tragedy, Aeschylus showed great
interest in the nature of Zeus. Supplices, first play of his
Danaid trilogy, contains mighty hymns to Zeus, implying that
he would have had an important role at the end. Likewise
Prometheus Bound shows a Zeus who appears in this play at his
worst – a seducer and gaoler – but whom we know finally came
to terms with Prometheus (although we do not have the final
play of this trilogy).7 In a fine article, Herington8 tentatively
suggested that in his last phase Aeschylus’ concept of Zeus
might have changed from an anthropomorphic to a more philo-
sophical and abstract deity. In support he quoted a surviving
fragment from Heliades (‘Daughters of the Sun’: it is unfortu-
nate that the play cannot be dated):
Zeus is Air, Zeus is Earth, Zeus is Heaven, Zeus is All and what-
ever is beyond the All.
Certainly the Zeus of Agamemnon is cast in similarly philo-

sophical terms as panaitios, panergetas, ‘responsible for
everything, doing everything’ (1486) and the Hymn to Zeus,
‘whoever he is, and whatever his name is’ (160-1), considers
him in the same abstract terms. When, in the following lines,
the chorus sweep over his terrible forebears without naming
them (167-72), it is as if they are deliberately making the point
that the anthropomorphic era of the gods is over. ‘Now Zeus is
the name that counts’ (to paraphrase 174-5).
Aeschylus presents Zeus as a semi-abstract cause in the
world, who continues his association with dikê: when the stage
77
figures make their various claims to dikê, they appeal to him.

Though Aeschylus does not follow Hesiod by making Dikê Zeus’
daughter (a role he reserved for Athena in the third play), he
connects dikê to Zeus’ plan for learning through suffering: the
Hymn to Zeus itself contains no use of the word or its cognates
but, before the end of the ode, the chorus sing ‘Dikê weighs out
understanding to those who have gone through suffering’
(pathousin mathein, 249-50; the translation is Fraenkel’s);
scales (from Homer’s poetry) belong to Zeus as judge and main-
tainer of balance in the world. At the end of the second
stasimon, the chorus conclude ‘Justice directs everything to its
end’, Dika … pan d’epi terma nomai (781).
Does the Oresteia actually show Zeus evolving a new dispen-
sation of dikê for humans? This used to be the established view
(perhaps by unexpressed analogy with the Old Testament God
giving way to the New, with his benevolent plan for humans). In
recent decades this has become an area of intense debate and
triggered valuable new analyses. The theory of an ‘evolving
Zeus’ was first seriously attacked by Lloyd-Jones (1971), who
argued that Aeschylus’ Zeus was no less anthropomorphic and
no more developed or developing than the Zeus of the epic
poets. His forceful book, The Justice of Zeus, remains influen-
tial. Since then, critics have produced a range of nuanced views
in support of either side of the argument. Winnington-Ingram,
for example, adopts a version of the traditional positive view.9
He begins by making the sideways move of showing that
previous literary tradition provided no clear conception of the
Furies; Aeschylus virtually invented them for the purposes of
this trilogy. He notes that they, who are chthonic forces (= gods
of earth), and Zeus, the supreme Olympian, are almost always
referred to in the trilogy in close proximity: that is because they
are working together. If at first, the shifting retributive
contexts to which they are linked are puzzling, the Furies are
finally revealed as the consistent supporters of Zeus in the move
from malign vendetta to a final, more benevolent, rule of civic
law. He concludes, ‘the drama, as the religion, of Aeschylus –
and the two are hardly separable the one from the other – is
78
4. Gods and Humans
centred on a Zeus who is conceived as the upholder of a just

moral order’.10 This is a well-argued reading of the trilogy with
which many are in agreement.
At the opposite extreme from Winnington-Ingram, writers
such as Cohen11 have argued that the trilogy reflects, ‘a cosmic
and political order neither moral nor just but tyrannic in the
sense that its ultimate foundations are force and fear’, an
emotional conclusion that seems based on an anachronistic
view of ancient society. The largely negative results of Goldhill’s
study of dikê have also been deservedly influential. Arguing
against a naively positive reading of the trilogy, he impressively
demonstrated the way the word dikê, a key concept of the polis,
fuses religious, political and legal aspects of the city’s life. By
analogy with the word ‘right’ and its cognates (to be right,
righteous, rightful, etc.) and the many ways in which it can be
used (the right to work, political right, to right a wrong, the
right answer and so forth), he showed how the word always
carries an irreducible excess of meaning. His conclusion is that,
just as in Antigone, all participants in the Oresteia trilogy
invoke dikê with some justification: the word cannot yield a
unitary meaning, and the problem of assessing claim and
counter-claim remains.12
In a recent introduction to the trilogy, Burian sensibly
restores the balance.13 He argues that for Aeschylus the
concept of dikê has an application wider than the Athenian
polis and means ‘a fundamental, natural principle, enforced by
necessity, on the grandest possible scale. The principle is
balance, and its enforcement takes the form of retribution to
redress imbalance. Retribution comes by what agency and at
what time it will, but it comes as an equal and opposite reac-
tion to whatever has overstepped its proper bounds.’14 He
shows how in Agamemnon Aeschylus uses the Trojan War as
an opening paradigm for a cycle of destruction in which each
retributive act of justice creates a further cause for retributive
action (the abduction of Helen produces the punitive expedi-
tion which produces the sacrilege at Troy which produces the
punishment of Agamemnon …), a pattern which is then fully
79
dramatised through the family members of the House of

Atreus in the rest of the trilogy. The natural cycle of dikê
cannot ever be overturned: but by the end, through the agency
of Zeus (and other gods and mortals), it has been transformed
into a civic process.
We have only the first play under consideration. The early
exposition of Zeus’ pathei mathos plan makes us concerned with
the justice of Zeus, but there is as yet no visible evidence of it,
only a series of apparent contradictions. By the end of
Agamemnon, we have witnessed the separate claims to justice
of Clytemnestra and Aegisthus, and the chorus’ counter-claim.
The chorus conclude that Justice has been polluted (1669) and
their final thoughts cast the future only in the form of further
retributive acts: ‘for it abides on the throne of Zeus that the
doer suffer. That is the law’ (themis, 1563-4).15 They can see no
further than the old law of endless retribution, pathein ton
erxanta (1658). There is no mathos here, only suffering, in the
despair and annihilation of the young and innocent caught in
this cycle (war victims, Iphigeneia, Cassandra). The play has
ended in the triumph of evil and Zeus’ reign seems pitiless.
Harsh coercion dominates the imagery: entanglement, taming,
subjugation, the snare, the yoke, the curb, the bit.
Humans: the problem of ‘character’

Until all too recently it was assumed that characters in Greek
tragedy could be simply ‘read off’ as with a modern novel or,
even, as in real life. This produced results that now amuse. The
great English textual scholars of the last century produced
famously naïve reasons why Agamemnon yields to
Clytemnestra’s request to walk on the carpet: Fraenkel claimed
that Agamemnon was a ‘great gentleman’, while Denniston and
Page, in corrective mode, asserted that, ‘it is … certainly not
because he cannot say no to a lady: it is simply because he is at
the mercy of his own vanity and arrogance …’.
The naïve idea that Greek tragedy deals with ladies and
gentlemen now makes us smile. It should also warn us to be
80
4. Gods and Humans
aware of our own pre-suppositions. It might be that we adopt a

post-structuralist stance and believe the concept of an
autonomous self has collapsed, to reveal only a network of
acquired discourses. This stance has certainly informed Gould’s
reading of Clytemnestra (below). It is more likely, however, that
we are still influenced by the long tradition of psychological
realism, to which a Freudian vocabulary and series of view-
points has more recently been added. We might unconsciously
expect that a playwright is concerned to convey individual
complex character to his audience: Hamlet, Othello, Lear are
readily interpreted as ‘tragedies of character’. But we should
not approach ‘character’ in ancient tragedy with the same
expectations. There is, to begin with, no word in Greek that
equates to ‘character’, its nearest equivalent, êthos, being an
important but much less inclusive concept. When Aristotle
discusses êthos in tragedy, for example, the concept is restricted
to, ‘that which reveals a moral choice’ (Poetics 50b7-8), and is of
secondary importance to plot – Aristotle’s view is that in
tragedy it is action (praxis) that counts, not an internal state of
mind. And although he regularly discusses the leading figure in
a tragedy, he has no word for ‘hero’ in our sense.
Furthermore, ancient tragedy has a stylised, public artifi-
ciality that works against developing the kind of audience
empathy we are used to experiencing in, for example, the
cinema, with its close-ups and voice-overs. In tragedy, masks,
formal costumes and bare orchestra remove ‘clues to inner
complexity’ (Easterling); actors deliver ‘set’ speeches; the
discontinuous mode of presentation (rhêsis followed by lyric
etc.) interrupts any feelings of identification we have made. We
do not see Aeschylean characters in their private world (as
arguably, we sometimes do in Homer, and again in Sophocles
and Euripides). There are no scenes of affectionate intimacy
(as, for example between Hector, Andromache and Astyanax in
Iliad 6). Characters in Aeschylus do not seem to have any other
bodily function except that of speaking.
A further not inconsiderable difficulty is that if a character’s
actions are controlled not by their internal psychê (our
81
Freudian model) but by external daimones, what space or role

is left for character?
This has exercised critics. One response was Dawe’s extreme
statement that the characters of tragedy are, in fact, thoroughly
inconsistent. This is so because they are subordinated to the
playwright’s desire to create ‘dramatic effectiveness’ in scene
after scene: ‘Agamemnon surrenders … only because it was
dramatically necessary that he should do so’.16 Goldhill’s more
recent sociological critique has negative implications for char-
acter too: ‘the language of mind and attitude, which one might
think to be a prime way of developing a sense of character, is
constantly implicated in the wider markings of social
discourse’.17
Easterling’s important 1972 article cut through a swathe of
problems to do with character.18 She restated a sensible
approach to the problem of free will apparently posed by the
daimones, arguing that the daimôn of the House does not
replace human motivation, rather it shows a duality at work
which is not, in fact, at all alien to common experience. Indeed
it is paralleled in modern thought where, one the one hand
geneticists and psychologists battle for predominance to explain
human conduct while on the other, law courts consider us
agents capable of free choice between good and evil. But it was
in her use of the phrase ‘humanly intelligible’ that she provided
the basis for fresh insights.19 She argued against Dawe that
Aeschylus, ‘wishes us to believe in his characters in a deep and
serious way. He may not have been interested in character for
its own sake, but he was profoundly interested in his charac-
ters, whom he saw as paradeigmata of the human condition.’20
A fresh development in the discussion of Aeschylean char-
acter was made in 1978 by Gould. Rejecting the idea of drama
as being, in any sense, an enactment of the way people behave,
he argued that ‘Dramatic personality is like dramatic space, in
being ‘framed’ by a principle of limited existence. … The
language of dramatic persons does not give clues to or ‘express’
their personality, their inward and spiritual being: it is their
personality, and their being’.21
82
4. Gods and Humans
Dramatic personality, then, is entirely determined by the

physical and, particularly, verbal medium in which the stage
figures exist. Characters become intelligible through language.
It is unacceptable to consider the personalities of the stage
figures in Agamemnon as somehow free-standing entities with
an existence external to the drama (at its most extreme, the
‘how many children had Lady Macbeth’ position’).
Gould’s reading of Clytemnestra deserves paraphrase here.
He rejects the idea of Clytemnestra being required to ‘make
sense’. Rather, ‘… it is very much a question of aura. It is not
that we do not understand Clytemnestra, but that we feel her.
We have a very strong sense of her as a person, one who
emanates from the world of the play and conveys in herself
much of its signal atmosphere. And our sense of her is rooted in
the play’s language and in its forms. … One strand in
Clytemnestra is her inverted, monstrous sensuality … but [this
is] not merely the “character” of Clytemnestra: [it is] linked to
the poetry of the play, with every part of the “world” of
Agamemnon. … Through a host of intensifications of language,
we have a sense of the whole world of Agamemnon, not just the
“character” of Clytemnestra, as figuring a dislocated and
unnatural ordering of human experience. … The public nature
of [her] language and the formal modes of speech which are the
medium in which her personality exists … and the pervasive
metaphorical colouring of the whole language of the play …
ensure that we cannot quite detach Clytemnestra from the
play’s “world”.’22
This is an impressive insight into ‘character’ as totally insep-
arable from the language of the play. The arc of Clytemnestra’s
development, as she moves from ironic concealment of her real
power to its triumphant manifestation, and then to milder ways
in an attempt to retain the new status quo, is equally the arc of
the plot as it is of her ‘character development’. Her intellectual
brilliance is entirely the product of her thrilling, dissimulating
language. Through language too we understand that she is
motivated not by internal pain or turmoil (there is nothing of
the mater dolorosa about her), but by external concerns which
83
are also the themes of the trilogy – by rights or claims to be

defended.
Pairs
Rather than developing his characters as unique individuals,
moreover, Aeschylus tends to consider them in conceptual pairs.
Clytemnestra is Helen’s twin sister, and he seems concerned
that his audience should consider them together – Helen
receives far more prominence in the play than its action would
suggest. The chorus’ thought turns repeatedly to Helen as
cause of suffering in war (225-6, 402, 823, 1453), keeping up her
manifest destructive power in parallel with her sister’s hidden
potential – until Clytemnestra overtly puts a stop to it (1462).
Their life-stories are of course similar – each marry a king, each
is an adulteress. By giving both of them compound man-
epithets Aeschylus links them further together as women
dangerous to men: Helen is poly-anôr (62), ‘a woman of many
men/husbands’, hel-andros (689) , ‘man-destroyer’ and andr-
oleteira (1465), ‘man-destroyer’ again. Clytemnestra’s heart is
famously andro-boulos, ‘man-minded’(11), an adjective with
wonderfully unrestrictable meanings (though all of them
sinister) ranging from, ‘with a mind like a man’, to ‘mindful of
her husband’.
The figure of Hesiod’s Pandora probably stands not far
behind Aeschylus’ presentation of both sisters as a kind of
unitary ‘woman-harming-man’ concept. Twice Hesiod tells of
the creation of the very first woman out of clay by the gods
(headed by Zeus), as a punishment directed at man. The name
of this woman, Pandora, means, ‘the giver of all gifts’, and is an
ironic allusion to the plagues and miseries she brings, evils
concealed by her deceitfully beautiful exterior. Through the
deadly beauty of Helen and the deadly deceit of Clytemnestra,
Agamemnon reflects this view of women as a sex specifically
designed to harm men by deceit.
The twin Tyndarid sisters are married to the Atreidae
brothers. Grammar is forced to its limits at the outset of the
84
4. Gods and Humans
entry anapaests which first describe them, compressing them

into one entity: ‘a great adversary at law, Lord Menelaus and
Agamemnon, the mighty yoke of honour of the sons of Atreus,
twin-throned from Zeus and twin-sceptred …’. Menelaus’ sepa-
rate kingdom in Sparta is elided so that both brothers appear to
live together in their family house. In this way Argos becomes
the virtual site of Helen’s seduction as well as the place from
which vengeance for it is exacted; Menelaus’ loss at sea is
understood by the audience as a precursor of a doom awaiting
his brother.
Clytemnestra and Cassandra make another pair, polarised
along the axis of language. One of them lies and persuades
everyone, the other tells the truth and persuades no one.
Aeschylus also creates some links between Cassandra and
Iphigeneia as a pair of doomed virgins. Rehm23 notes that both
cast off garments before death; both deaths are perverted sacri-
fices, one determined by Artemis and the other by her sibling
Apollo. While Iphigeneia is ‘a picture straining to speak’ (242),
Cassandra sees that life is like a picture ‘blotted out by a
sponge’ (1329). Both sacrifices are contrasted with earlier,
happier, ritual occasions in their father’s presence (cf. 243-7 and
1277-8).
Female versus male: gender and gender-horror

Over the last four decades, anthropology, sociology and feminist
studies have all transformed our understanding of the role of
women in Greek tragedy.24 It should first be made clear that,
despite the titanic role of Clytemnestra in Agamemnon (and
Medea, Antigone, Electra, Deianeira, Phaedra all dominate
their tragedies too), the lives of real women in fifth-century
Athens were extremely restricted. All evidence about them
comes from men. Perhaps the best-known comment on contem-
porary women is the dismissive remark Thucydides attributes
to Pericles in his Funeral speech of 430 BC. Addressing the
female relatives of the dead, he says, ‘Your great glory is not to
be inferior to what god has made you, and the greatest glory of
85
a woman is to be least talked about by men, whether in praise

or blame.’25
The invisibility of women in the public arena is borne out by
legal texts that show that women were lifelong legal minors,
who exercised no political or financial rights. They did not
participate in the exclusively male democracy. They were
always under the control of a male kurios or guardian – first
father, then husband. In marriage there was an imbalance of
age: girls tended to be married at the age of puberty, while the
husband selected for them was likely to be about thirty.26
A feature that brings out their value in the eyes of men is the
epiklêros system: if her father’s household was left without a
male heir, the daughter became an epiklêros. She ‘inherited’ her
father’s property, in the sense that she acquired the right to
transfer the ownership of the family property to a male member
of the immediate family: in fact this meant that she was legally
obliged to marry the next-of-kin on her father’s side, even if this
meant leaving a marriage she had already made.27 The reason
for this was to produce a male heir for her father’s oikos. From
this we can see that an Athenian woman’s status was derived
entirely from her kinship with males, and that her sole positive
function was to produce a male heir for the oikos of her
husband, or if epiklêros, her father.
She thus, however passively, played a key role in the trans-
mission of property and, in less than a decade after the first
performance of Agamemnon, was to be vital too for the trans-
mission of citizenship (after 451/0 both parents had to be full
Athenians for their child to qualify).28 On both these counts,
female chastity was consequently paramount and, while men
spent little time indoors (and were not expected to be chaste),
respectable women lived their lives inside, confined to separate
women’s quarters with their children. Women left home to visit
neighbours, to assist in childbirth and – their sole public func-
tion – to participate in religious rituals for the benefit of the
polis. The symbolic dichotomy of male exterior and female inte-
rior, built into the structure of the theatre (see Chapter 2),
reflects a reality. Contemporary male rationalisations for this
86
4. Gods and Humans
treatment of women included the view that women were natu-

rally lacking in self-control and rationality; male intervention,
in the form of frequent sexual intercourse and pregnancy, was
the best treatment. Such views inform the presentation of the
disruptive virgins of tragedy, such as Cassandra, Antigone and
Electra, and also the immoderate passions of women separated
from their husbands such as Deianeira, Phaedra and Medea.
Anthropologists have made the following kind of analysis:
like many cultures, the culture of ancient Athens devalued
women in relation to men. While men exercised power at the
cultural centre, women were identified with, and reduced to,
the margins, linked to the irrational world of beasts, but
accorded some divine powers as well. Women’s fertility made
them a vital part of culture – but they were also what culture
was designed to tame or suppress.
Women, then, provided men with a sense of that Other
against which the male citizen defined himself. Men adopted for
themselves positive qualities, projecting out onto women what
they feared and hated. Slater’s (1968) psychological reading
viewed tragedy as expressing the fearful ‘double feelings’ of
men towards women.29 Tragedy, putting onto the stage female
figures of extraordinary power, exercised the male imagination
by dramatising its intractable dilemmas and basic fears: threats
to rationality and order, to the integrity of the family, and the
survival of the community. We should be clear that, even if
tragedy is often ‘about’ women, it is addressed to men. This
would still be true, even if new evidence were to show that fifth-
century women were allowed to attend tragic competitions, as
they did in later antiquity.
Greek thought, generally tending towards polarisation,30
certainly polarised gender. Cassandra’s prophecy of the revenge
murders to come in Libation-Bearers is cast in typically
gendered terms, ‘when woman falls for me, a woman, and man
falls in place of ill-wedded man’ 1318-19). Oresteia can certainly
be read as a drama of gender-conflict which is eventually
resolved in favour of patriarchy over matriarchy. This is
achieved through the figure of Athens’ presiding goddess
87
Athena who, although female in gender, was not born from a

female and in this, as in her perpetual virginity and her military
aspect, is a kind of honorary male. (The downgrading of the
female reproductive role Eumenides 666-71 is significant here.)
Clytemnestra at the beginning of the trilogy makes an inter-
esting contrastive pair with Athena at the end: negative, then
positive, both are women with minds like a man.
The conflict of gender we noted between the chorus and
Clytemnestra in Chapter 3 can now be discussed further.
Frightening issues of gender pervade Agamemnon. The figure
of Clytemnestra monstrously combines both male and female
attributes within one body, which also includes a terrifying and
explicit sexuality. She embodies the perverted world of the play,
climactically displayed in her speeches of triumph (1372ff. and
1431ff.), when the chorus note her bloodshot eyes (1428-9). The
combination of, on the one hand, her masculine abilities to kill
and to control events by argument and persuasion, and, on the
other, her feminine skills of flattery and deceit, is profoundly
disturbing. Cassandra can only use a stream of beast and
monster images to describe her (1228ff.).
Clytemnestra’s relationship with Aegisthus inverts their
gender roles. She is the man and he the woman: while she
guards the threshold, he is an oikouros, a stay-at-home (1225).
She is a lioness to his mere wolf (1258-9), he is a woman (1625)
in the chorus’ contemptuous address. Goldhill well notes her
prominent use of the word aischos, ‘shame’ or ‘modesty’ and its
cognates (614, 856, 1373). The word, commonly used of sexual
behaviour, indicates the kind of behavioural control men exer-
cised over women in Greek society: Clytemnestra boldly turns
the male concept of female ‘shame’ inside out when she
proclaims that she is not ashamed (1373).
Killing belongs to the male domain. ‘The Female is slayer of
the Male’ (1232-3) expresses the play’s central horror, a murder
which is also an act of gender inversion: in the scene with
Cassandra, the notion of a female threat to Agamemnon is
precisely what the chorus cannot grasp (1251), and after the
murder, the fact that a woman perpetrated it is an additional
88
4. Gods and Humans
horror. The chorus had devoted much of their dealings with

Clytemnestra in the earlier scenes to doubting that a woman
could even have the mental capacity to report information
factually and uninfluenced by fantasy (see Chapter 3). Now
they see her as a spider who has caught her husband in a web.
But Clytemnestra still demands not to be judged as a witless
woman (1401) and defends herself like a man, until gracefully
giving room for Aegisthus. Even then, she boldly asserts, ‘This
is a woman’s word, if anyone cares to understand it’ (1661).
Aeschylus deliberately kept the theme of perverted blood-
shed caused by a woman in our minds from the play’s outset
by the many references to Helen and the War, which was
‘because of a woman’ (62, 823), or ‘through the theft of a
woman’ (402, prominently placed at the end of a strophe). The
first stasimon from 403ff. paints a picture of Helen’s arrival at
Troy where she brought ‘destruction in the place of a dowry’
(406: note the inversion, which recalls the irony of Pandora’s
name). In the first part of the second stasimon, the
‘oxymoronic’ concept of Helen as a femme who is also fatale in
the vivid sense of a ‘woman causing death’ is twice extensively
developed, first by the etymology of her name (681ff.) and
then through the lion-cub analogy. Even after Clytemnestra’s
declaration of herself as Agamemnon’s murderess, the chorus
suggest that Helen, the archetypal woman-bringing-death, is
the ultimate cause (1455ff.).
In Greek society, women lived largely indoors. The appear-
ance of an unaccompanied female from the skênê into the
orchêstra often receives explicit motivation in the text.31 The
reason given is usually the absence of the male kurios. Chapter
2 alluded to the symbolically-gendered configuration of the
theatre, whereby, in some plays, the dark, unseen interior could
be identified with the female domain, the bright exterior with
the male.
This potential is put to brilliant use in Agamemnon.
Clytemnestra stands at the threshold into the interior, control-
ling the passage into it. She persuades Agamemnon to go inside
in a very particular way: over the red carpet, which seems to
89
represent infinite possibilities of danger. Then Cassandra spells

out the particular horrors of this interior. Then Agamemnon
meets his death there, inside the woman’s domain.
The powerful effect on the imagination produced by
Aeschylus’ horrific gendered space is vividly conveyed by Hall.
Seeing the House as a ‘monstrous mother’ she writes, ‘a psycho-
analytical critic might even suggest [that it] becomes itself an
enormous, toxic, lethal womb. It disgorges the bloodied corpse
of Agamemnon, killed like a defenceless baby in the amniotic
fluid of his homecoming bath; he is dragged alongside
Cassandra, stillborn or aborted in a sinister parody of a
multiple birth, through the vulva-like doors of the palace into
the harsh daylight of Argos.’32
Many plays present corpses from the inside of the skênê.
Only in this play does the gendered symbolism of the theatre
space create such a powerful effect of ‘gender-horror’.
90
Language, Speech and Silence,

Style, Imagery
As for the rest, I’m silent. Big ox on my tongue. But the house itself,
if it had a tongue, could speak out very plainly. In fact I speak freely
to those who get it, and for those who don’t I’m a blank.
Watchman, 36-9
The poetic texture of Agamemnon hits the reader between the

eyes, demanding attention in its own right. It is deinon: an
adjective that combines the notions of terrible, powerful,
wonderful and skilful. Inseparable effects of language and
imagery combine. Through dense accumulation of adjectives
(many of them coined or compounded), mixed metaphors,
verbal ambiguities and suggestive allusions that cannot be
pinned down, Aeschylus forges an idiosyncratic language-world,
one which continuously startles and arrests the listening audi-
ence, and forces them to grasp at complex and elusive
meanings. It is a deliberate strategy, and the Watchman indi-
cates here the discrepant awareness between those who realise
they must engage with it in order to ‘get it’ (mathein), and the
rest of us. Language, just like the range of theatrical and narra-
tive effects discussed earlier, is another heavy weapon in
Aeschylus’ arsenal, deployed to shock, promote defamiliarisa-
tion, and to convey a world where ‘fact and fantasy, act and
symbol, person and personification freely associate’.1
The Oresteia is particularly famous for its unparalleled
system of interlocking and recurrent images, mapped out in a
key analysis by Lebeck2 and described as the trilogy’s ‘single
91
most compelling feature’ by Zeitlin.3 These images expand and

develop throughout the trilogy, keeping pace with the stage
action. Sometimes they acquire symbolic power. They may even
culminate in being visibly embodied onstage. Language, style
and imagery thus work inseparably with the stage action to an
extent that has probably never been subsequently equalled;
because Agamemnon lays the ground for the trilogy, it contains
more than double the number of images in the other two plays.4
The rich and deliberately ambiguous texture, with its
multiple and often irresolvable layers of meaning, creates huge
problems for textual scholars grappling with a defective text.
The fusion of choral and civic language

Aeschylus puts two different worlds into a linguistic crucible.
The first, choral lyric (already discussed in Chapters 1 and 3), is
narrative and moralising, emotional and theological. Its texture
is allusively suggestive, richly permeated with metaphors,
symbols, gods and personifications, historical memories and
associations. It is a self-conscious language claiming wisdom
through the extrapolation of universal truths from traditional
myths and fables. Aeschylus shows the chorus fruitfully
engaged in this activity, but also exposes its limitations. He
pushes all these properties of lyric further than they had gone
before to serve his dramatic purposes.
The second language world reflects the developing Athenian
democracy. Citizens in the Assembly, Council and Law Courts
listened to their fellows attempting to state facts as effectively
as possible in order to persuade their audiences into a partic-
ular decision. They observed how linguistic efficacy or inefficacy
affected the final vote of debate or trial. They would have been
well aware that language could control the perception of reality,
and be in itself a powerful determinant of outcome. Goldhill’s
work5 amply makes the case that Agamemnon mirrors a society
intensely aware of, and fascinated by, language, dialectic and
rhetoric (even if the famous teachers of these subjects were yet
to reach Athens). These citizens are also the audience of the
92
5. Language, Speech and Silence, Style, Imagery
play, which has been constructed as a complex legal action

submitted to their judgment. A trial is staged in Eumenides.
Not surprisingly the text of the Agamemnon interplays with
legal procedure – not least in the use of contemporary technical
legal terms (e.g. ‘claimant-at-law’, 41; ‘alien resident’, 57), the
semi-legal structure of the epirrhematic agôn (contest) between
Clytemnestra and the chorus (1399ff.). Noteworthy too is
Agamemnon’s measured and programmatic political speech,
addressing the problems of a disaffected populace (830ff.).
Aeschylus welds together these two different types of
discourse, lyric and civic, old and new, connotative and denota-
tive. If Aeschylus, as Aristotle asserts, was the first playwright
to use a second actor and give more emphasis to speech as
opposed to song, then he originated that familiar dialectic in
tragedy between the more precise and contemporary language
of the episodes and the ‘traditional’ language of the odes. But
this juxtaposition is not so sharp in Agamemnon as it was to
become in later tragedy: the language of the episodes too can be
dense, ambiguous and allusive.
Inherent powers and dangers of language; silence

Agamemnon reactivates the archaic belief that words in them-
selves are not merely inert but active and ‘efficacious’.6 Hinted
at early on through the Watchman’s triple self-suppression (36-
9), the looming power of words heightens the potential
significance of every utterance in the play and cues the audi-
ence to listen with particular attention. Words, like daimones,
can have autonomous and independent life once released from
the speaker’s lips. The Watchman, the Herald and particularly
the chorus consequently speak with anxious awareness. The
chorus so frequently break off their narrative in the parodos
that it becomes a structural pattern.7
Proper names are a special case of language’s power. By a
kind of etymological punning, names are prophetic, precipi-
tating events inherent in their meaning (this archaic belief was
expressed in the Greek phrase onoma ornis, replicated in Latin
93
as nomen omen, ‘names are harbingers’). On hearing the

Herald’s news (681ff.), the chorus meditate at length on the
prophetic meaning of Helen’s name, of which the hel- prefix
carries the sense ‘destroy’, connecting it to the loss of resources
and life at Troy: she was hele-nas, ship-destroyer, hel-andros,
man-destroyer, and hele-ptolis, city-destroyer. They wonder who
had the wisdom to name her so appropriately (their own
wisdom on the subject has, with typical irony, come too late).
Similarly, in despair after the death of their king, they pause on
the phrase dia Dios (‘through Zeus’) and suggest that the god’s
name reveals his role as the agent of causality (dia is both an
oblique case of the noun ‘Zeus’ and a preposition meaning
‘through’, ‘because of’): ‘Alas! Through Zeus, the cause and
agent of everything that happens. For what is fulfilled for
mortals without Zeus? Which of these events is not wrought by
gods?’ (1485-8). In both instances, revealing the power inherent
in a name has contributed to the play’s overall search for under-
standing.
The chorus show that dysphêmia, the expression of anything
‘untoward’, is dangerous, and that it is safer to restrict oneself
to positive utterances or to keep silent (euphêmia). The idea of
silence as a safeguard, hinted at by the Watchman at 36, is
expressed more clearly by the chorus at 548, ‘For a long time
I’ve considered silence a remedy against harm’. Aeschylus’
chorus show their respect for silence in many ways: in their self-
imposed break-offs (e.g. 67-71, 160-83, 248-57), in occasional
failure to press Clytemnestra for clarification (e.g. 263), in their
strange first passage of stichomythia with the Herald (538-50)
where they do not fully clear up the Herald’s incomprehension,
and in their attempts to hush up Cassandra. At the end of the
play, denied their usual role of speaking the concluding lines,
they are literally and effectively silenced by the new regime.
The effectiveness of silence is dramatised in a different way
in the scene between Clytemnestra and Cassandra (1045-68):
Cassandra’s striking refusal to engage in words constitutes a
brief but telling victory. For the first time we see that it is
possible to foil Clytemnestra.
94
Words as weapons: persuasion and failure

Language is the power-play of Agamemnon, displayed as a sharp
or blunt weapon at the service of all the speaking characters.
Every episode shows language faltering and failing, but the goal
of language and the victory for the speaker, is always persuasion.
Twice personified (105-6, 385-9), persuasion (peithô) is a
dangerous power and an important theme of the play. The
chorus claim it (105-6) at the beginning of the mighty parodos;
in the first episode (264ff.) they suspect Clytemnestra of being
prematurely convinced of Troy’s fall without evidence and her
two speeches are the proof they require. In the following first
stasimon they sing with ambiguous reference, ‘headstrong
Peithô, insufferable child of Atê who plans in advance, forces (a
man into evil) and there is no remedy’. Agamemnon’s murder
takes place because of Clytemnestra’s dissimulating persuasion:
the stichomythia of 931ff., followed by the walk on the tapes-
tries, is a culminating enactment of evil persuasion, to be
contrasted in the next major scene where Cassandra, who does
not dissimulate, ineffectually spells out the painful truth: she
cries in despair, ‘I wasn’t persuasive at all, not at all’ (1212).
Both the chorus and Cassandra, in their different ways, coun-
terpoint Clytemnestra’s verbal victories by their linguistic failure.
The chorus, despite so much fearful anticipation, are simply
unable to put into words the fate that awaits Agamemnon while
Cassandra, who can verbalise it, cannot do so efficaciously. The ox
stands on their tongues too. While the chorus refuses to accept
both its own gift of prophecy and the truths offered by Cassandra,
Cassandra is of course expressly denied peithô by Apollo.
The third stasimon, sung after watching the events of the
carpet scene (975ff.), most vividly conveys the chorus’ linguistic
plight – in some of Aeschylus’ most striking (and difficult)
metaphorical language. I quote Denniston and Page’s literal
translation here:
Why does this terror persistently hover in front of my divining

heart? It plays the prophet, my song, though none has bidden or
hired it. Nor yet, to reject it (the terror), like dreams of doubtful
95
import, does confidence persuasive sit on the throne of my

thought.
In this passage their terror is cast as a song ‘prophesying’ to

them; their heart is likewise ‘divining’. The content of the
prophecy comes to them involuntarily and cannot be logically
rejected: they have some idea what the future holds. This
important sequence of ideas is soon repeated in different
phrases (990-1000; the translation here more loosely derives
from Denniston and Page’s suggestions):
Spontaneously from within me, my heart is chanting a Furies’

paean, one without a lyre, one which in no way has the welcome
confidence that hope provides. My gut feelings are telling me no
idle tale, my heart (I mean) going round and round in circles that
bring fulfilment. I pray that this may fall away from expectation
as falsehood into non-fulfilment.
The two passages form a ring (a feature of archaic poetry): in

the second, ‘spontaneously’ echoes ‘not bidden and not hired’;
‘heart’ and ‘guts’ echoes the earlier ‘heart’; a ‘prophet’s song’
becomes a ‘Furies’ paean’; in both cases confidence fails the
chorus but the spontaneous song inside their body cannot be
cast out. Thus twice, and at length, the chorus suggest that they
have an inner knowledge of the future. Yet nonetheless this is
how their ode ends (in another passage of strained and tangled
metaphors):
Were my destiny not prevented by gods from getting more than

its share, my heart, outstripping my tongue, would be expressing
these things. But as it is, it mutters in pain in the dark, not
expecting ever to accomplish anything in time, though my mind
is on fire.
The chorus’ intuition of the future hovers on the threshold of

consciousness, but is denied admittance. They are doomed to feel
in their heart all the anguish that the future holds, without the
capacity to use their tongues to speak of it; this is their destiny.
96
Cassandra’s verbal destiny is no less painful to her. The fact

that she is denied peithô is hammered home again and again.
She makes repeated efforts to say that Agamemnon will be
murdered, and the chorus show sympathetic interest in her
own history and full acceptance of her details about the
House’s past – in fact they respond positively in every possible
way except the important one: they avoid the truth with a
string of dismissive responses: she utters dysphêmia (1078),
they’re not looking for prophets (1099), the whole city knows
what she’s talking about (1106), she is riddling and obscure
(1112-13), they would not style themselves good interpreters
(1130-1), prophets never say anything good (1132-3), a baby
would understand her (1163-4).
Eventually Cassandra explains why she prophesies without
being believed. With unintentional irony the chorus earnestly
reply, ‘Well, we believe you’ (1212-13). It makes no difference.
Cassandra goes on trying, but even when she forces them to
hear, ‘I’m saying you will see Agamemnon dead’ (1246), they
can only respond with a further request for euphêmia and a
wish that it may not be so; their next question, ‘what man
contemplates this deed?’ (1251) only shows how far off the
mark they still are. The scene is a strange, protracted display
of failed communication, but Cassandra’s manifest agony each
time her visions overtake her, the poignant bird imagery
(1050, 1140ff., 1316), the terrible proximity of her death and
the bleak understanding of the human condition expressed in
her final lines (1327-30) make it one of vivid interest, stirring
profound pity.
Style
All poetry, through its use of vocabulary, metre, alliteration and
sound patterns, metaphor, paradox and ambiguity, intensifies
normal language. Considering these attention-seeking features,
the literary theorist R. Jacobson once defined poetry as ‘organ-
ised violence committed on ordinary speech’, a phrase strikingly
appropriate for Agamemnon.
97
Although the play includes passages of a delicate and

sensuous beauty (e.g. 72-82, 403-26, 737-43), its more typical
language is ‘shockingly’ forceful and jagged. Especially in lyric,
strings of lengthy compound adjectives, some of them newly-
coined for the context, unnaturally attach to a single noun and
impede any easy flow of comprehension; or maybe the opposite
effect occurs, making a single epithet stick out of its
surrounding context like a boulder in the sea (e.g. the sinister
warning of androboulon, 11, discussed on p. 84 above). Abstract
replaces concrete noun, and vice versa. Metaphorical language
is used with an astonishing freedom (in English literature,
Gerard Manley Hopkins is the only remote comparison one
could make); sometimes multiple metaphors combine to create
an effect not so much ‘mixed’ as ‘clogged’.
Traditional features of earlier poetry are often transformed:
in the Beacon Speech, the journey of the flame from Troy to
Argos might have been a merely stock geographical description,
familiar from the archaic period; instead the flames are made to
partake in a thrilling relay footrace, recalling the
lampadêphoria festival in Athens.8 Likewise, the simple
morality of the traditional fable/animal story turns into the
multivalent image of the ‘lion in the house’.9 So too the tropes
of choral wisdom are repeatedly undercut to expose multiple
ambiguity. The key opening simile of the vultures bereft of their
young10 is remarkable, among other things, for its abandon-
ment of the clear ‘as … so’ structure familiar from Homer; the
comparans and the comparandum (the two elements which are
compared) refuse to remain distinct.
Similarly, Aeschylus uses existing epic vocabulary, but may
also transform it in startling ways. Here we consider the
complexity of a single sentence, a key passage, difficult and
possibly corrupt (140ff.), in which Calchas is interpreting the
portent of the pregnant hare devoured by eagles by aligning it
with Artemis’ affection for all baby animals, and specifically
lion-cubs.
Homer had used the word hersê to describe new-born
animals (e.g. Odyssey 9.222); its literal meaning is ‘dew’, and
98
perhaps dew conveyed the drenched quality of young at the

moment of emerging from their mother. But to describe the
baby lions, Aeschylus employs a quite different word for dew,
drosos, coupling it with an invented adjective aeptos which
might mean ‘not-following’ and so in this context ‘incapable of
following the mother’. In this same sentence there are two more
unique words, philomastoi, ‘breast-loving’ (simple enough), and
obrikaloi, a word which has eluded derivation but in context
must also mean ‘animal young’. Translated as literally as
possible, the sentence reads: ‘So very well-disposed is the
Beautiful One (Artemis) to the not-following dews of ravening
lions, and delighting in the breast-loving obrikaloi of all wild
creatures that roam the land, that she demands to bring about
what is portended by these events (toutôn xumbola)’.
In this sentence the first two phrases carry the same simple
general sense, ‘Artemis likes the young’. Why repeat it twice in
such arresting and demanding language? We need to invoke the
principle expounded by Lebeck, that the greatest number of
meanings are likely to be compressed into the passages where
language and syntax are most difficult. In this instance,
perhaps the purpose is to flag up to ‘those who get it’ (39) that
‘the young’ has not merely a local but a wider significance, in
this trilogy where each innocent new generation is forced back
into a parental pattern of revenge and death (the brief mention
of lions helps to prepare for the centrally important lion-cub
story, 771ff.). Then, after the arresting expansion of these first
two phrases, the third is no less arresting by being, in contrast,
cryptically brief; Denniston and Page understand that here
Artemis is demanding from Zeus the death of Iphigeneia: the
killing of the mother hare acts as token or portent of this second
sacrifice too.
It is hard to imagine that the average audience member could
have grasped the import of this. However, just a little earlier
(134-6), carefully-chosen vocabulary created further striking
ambiguity which seems to reinforce this interpretation. Calchas
had explained, ‘for through pity, pure Artemis bears a grudge
against her father’s winged dogs, autotokon pro lochou mogeran
99
ptaka thuomenoisin’. This phrase is usually translated as ‘who

are sacrificing a hapless hare with its brood before the birth’.
Stanford, however, notes another possible translation: ‘sacri-
ficing a poor trembling victim, his own child, on behalf of the
host’. (Autotokon can be taken either as equivalent to autois tois
tokois, ‘children and all’ or as ‘own child’, while lochos means
both ‘armed band’ and ‘child-birth’). So perhaps for those (few)
who ‘get it’ Calchas has already pointed twice to the sacrifice of
Iphigeneia before the chorus describe it at 228ff.
Lesser kinds of linguistic violence employed by Aeschylus
include catachresis (misuse or misapplication of the usual
meaning of a word): Helen is polyanôr (62, literally ‘of many
men’), an epithet which normally means ‘populous’ or
‘frequented’: here he makes it mean ‘promiscuous’. Language
may also heightened by paradox and oxymoron, for example
Clytemnestra is a two-footed lion (1258), there will be a music-
less, banquetless sacrifice (150), Helen dared the undareable
(408), grace is violent (182).
Compound adjectives
The abundance of these is a particularly prominent feature of
Agamemnon.11 As many as 102 are hapax legomena (words
created for the specific context and never used again), and there
are another forty-four unique to Aeschylus but also present in
other plays. Many of these compounds are bold and startling
formulations. They occur in iambic trimeter as well as lyric:
Aeschylus often violates any pretence of ‘normal speech’ by
building a trimeter line out of four or even merely three
massive compound words (107 such verses out of a total of 876).
Jacobson’s ‘organised violence’ is apparent here.
Aeschylus’ compounds are coined for exact purposes within
their context (they do not create an ‘ornamental’ effect, as in
Persians). ‘Unbulled’ (ataurôtos, 245) is an extraordinary way
of talking about Iphigeneia’s virgin state (= ‘not yet sent to the
bull’, ‘unmated’). Looking back to her innocent childhood and
set within the phrase, ‘with pure voice’ (‘hagnai d’ataurôtos
100
audai’), Aeschylus creates an arresting juxtaposition of inno-

cence and brute force. In the context of her sacrifice, the epithet
also resonates with the additional sense of, ‘not (yet) treated
like an animal’ (= being sacrificed; at 232 she is held over the
altar ‘like a goat’). The coinage richly appeals to our human
sympathy and adds to the cruel picture of the sacrifice.
Other coinages compress a noun and adjective into a single
word, for example the evocative nuktiplanktos, ‘night-
wandering’, ‘wandering by night’ (12 and 330), used both for the
anxious Watchman’s bed (he cannot find the right place for it?),
and for the wandering of the exhausted victors of Troy in search
of food. Hêmerophantos (82), ‘day-visible’, ‘visible by day’ is used
to powerful effect in a metaphor for extreme old age: the chorus
say they advance on sticks, ‘no stronger than a child, a dream
that wanders visible by day’ (onar hêmerophanton alainei). The
phrase, hinting at the future inefficacy of the chorus, expresses
all the fragility of old age as it nears the vanishing point of
extinction. With its identical suffix oneirophantos, literally
‘dream-visible’, ‘visible in dreams’ occurs in a passage of the first
stasimon which forms a bridge between Menelaus’ grief at
Helen’s absence and the grief of the relatives of the war dead
(410ff.: sadly, some of the text here is irrecoverable): like the
previous passage, this one describes the movement towards final
invisibility and ends with a beautiful ‘fade’: ‘a ghost will seem to
rule the house … sad imaginings appear, haunting the dreams
(oneirophantoi) … (then) slipping through the fingers the vision
goes, not hereafter accompanying the winged paths of sleep’ (i.e.
not to be contacted even through dreams).
As already shown by the discussion of 141f., Aeschylus’
coinages often reveal strategic concerns which underlie the
trilogy as a whole. One area was already noted in chapter four:
the di- compounds, dithronos (‘twin-throned’, 43, 109) and
diskêptros (‘twin-sceptred’, 43) helped bind the Atreidae
together into a single unit, just as the andro- words similarly
paired Helen and Clytemnestra and defined them as transgres-
sive. A longer-term, though less perceptible area of concern in
Agamemnon is the offstage Argive populace who, in the third
101
play, will have transformed into Athenian citizens. A sense of

‘the people’ needs to be established here without too much
anachronistic detail. Aeschylus achieves this by using dêmos,
that specific contemporary term for ‘the Athenian people’ not
as a noun in its own right, but as the prefix of various new-
minted compounds that then become part of the lyric
vocabulary. Dêmothrous, ‘uttered by the people’ occurs four
times (883, 938, 1409, 1413); dêmioplêthês ‘of which the people
have a large share’ at 129, and another dêmo- coinage is dêmor-
riphês, ‘hurled by the people’ (1616). This last word occurs in a
startlingly expressive phrase, literally, ‘curses of stoning hurled
by the people’, a vivid compression for a series of events – first
curses and threats, then stoning.
One of the densest accumulations of compound adjectives
and coinages within a single sentence comes at the end of
Calchas’ interpretation of the omen (146f.). This is a key
passage, which exemplifies Lebeck’s principle. The coinages
here are marked in italics, and this time all the compound adjec-
tives are translated literally, rather than being turned into
phrases, so the Greek-less reader can see exactly how they accu-
mulate to create an overwhelming sense of suggestive but
non-specific foreboding:
I call on Apollo, god of cries, that he do not fashion any wind-

adverse lengthy ship-restraining lacks-of-sailing for the
Greeks, (so) speeding on another sacrifice, one which is
without-music (or lawless), without-feast, an innate crafter of
feuds, one which is not man/husband fearful: for there waits a
terrifying, guileful, housekeeper rising-again, mindful child-
avenging Anger.
The dense epithets riddlingly keep at bay the identity of ‘the

house-keeper, Anger’ at the same time as they suggest that
identification is of key importance.12
Aeschylus’ heavy compounds readily lent themselves to
parody, as in Aristophanes Frogs, where they are described as
‘mountainous mouthfuls of words’ and, in the ludicrous
102
manner of comedy, found heavy when weighed in a comic scale

pan. The fictional Aeschylus in the comedy, however, makes a
valid defence: ‘Great thoughts and great imaginings need words
commensurate.’13
In later antiquity, the critic Dionysius of Halicarnassus
described Aeschylus as austêros, ‘strong’, ‘harsh’, or ‘stern’,
saying that his style had an authadês kallos, a ‘self-willed
beauty’. Aeschylus’ ‘weightiness’ is certainly not at all
similar in feel to the fully ‘classical’ technical perfection of
Sophocles and Euripides. He can be stiff and archaic, his
harsh phrasing can lack euphony, his diction (though some-
times oddly colloquial) bombastic, heavy and obscure. On the
other hand, all this creates a uniquely exciting and chal-
lenging world for the imagination of the audience to inhabit.
Furthermore, sentence structure, certainly within the
episodes, is clear enough, and Aeschylus understands the
virtue of contrast. He is capable of a telling directness of
expression where it counts. Cassandra’s final lines (1327f.)
are a good example:
Human concerns! When they go well, they are like a shadow,

when badly, a damp sponge wipes out the picture! For this more
than anything else, I am filled with pity.
Likewise Clytemnestra’s devastating clarity at 1401-6:
You question me as though I were a witless woman. But with

fearless heart I say to you, who know it: it’s all the same whether
you want to praise or blame me. Here is Agamemnon, my
husband, and a corpse, the work if this hand of mine, a worker
of justice. And that’s how it is.
Imagery
Imagery is a general term to cover the figures of simile,
metaphor and metonymy14 and personification. It can create an
enormous range of effects, ranging from subtle and almost
103
imperceptible to startling and frightening. It can have a re-

creative power, forcing the reader or audience to see things
newly, often sensually. On the other hand, much imagery in
language has ceased to be felt as such: in the common allitera-
tive phrase, kludôn kakôn, ‘a sea of troubles’ we hardly feel that
‘sea’ has replaced a more neutral word such as ‘mass’. This is
an example of a dead metaphor, and Aeschylus avoids these,
unless he can startle us by reanimating them.
In Agamemnon certainly, imagery is inextricable from
language: many of the single words already discussed, whether
new-minted or used in a deviant context, formed an image. The
Watchman’s bed, for example, was described as ‘wandering by
night’, but ‘to wander’ is the activity of a human not an inani-
mate object. Persuasion ‘still breathes down on me from god’,
sing the chorus (105-6), momentarily making a daimôn of an
abstract noun by giving it lungs and a mouth.
Some of Aeschylus’ major personified daimones have already
been considered in Chapter 4. It only needs to be added here
that Aeschylus repeatedly creates minor unexpected personifi-
cations, building up a world of multifarious dynamic entities.
Thus grief bites (791), fire and water conspire (650-1), a storm
is an evil shepherd (656-7), disease is a neighbour breaking
through the party wall (1002f.), ears of corn feel pangs of child-
birth (1391-2), dust is mud’s thirsty sister (494-5). In more
extended passages of visionary personification, Ares is a money-
changer trafficking in the bodies of dead men, short-changing
their kin with ashes in an empty urn (437f.). Meanwhile Justice
shines in the houses of the poor but abandons those of the rich
(772f.).15
The major image-systems of the play, can only be briefly
outlined here. One is light and darkness,16 another sacrifice:17
Vidal-Nacquet considered the inter-connectivity of sacrifice and
hunting.18 Knox magnificently explored the multivalency of the
lion-cub story (717ff.), which is perhaps the most overt and
extended expression of this pervasive nexus.19 The simile of the
vultures and its complementing omen of the eagles deserves
careful attention: the images there, densely interwoven with
104
associative repetition, create the springboard for the whole

play.20 Most images connect together through verbal similarity
rather than strict duplication, and we have already described
how the initial presentation comes in a highly condensed form
and is presented almost like a riddle (ainigma): if the audience
could only ‘get it’ early on, they would be able to predict the
future course of events. As it is, they must rely on the gradual
expansion of the image as it develops in time with the unfolding
of the plot.
Leavis rightly warned that ‘it will not do to treat metaphors,
images, and other local effects as if their relations to the poem
were at all like that of plums to a cake’.21 The Agamemnon’s
image-systems interlock and overlap like the circles of a Venn
diagram, accounting for much of the clogged imagery; their
effect is in their totality. Taking into account the fact that the
separation of one element from the whole is an artificial,
inevitably reductive process, two examples of imagery, one
small and one large, are isolated here.
Trampling
In the first stasimon, the chorus, with ostensible reference to
Paris (but some ambiguity), disagree with people who deny the
gods’ interest in men who trample on ‘the beauty of things invi-
olate’ (369f.), and they go on to talk of the ruin of the man who
has ‘kicked the great altar of Justice into invisibility’ (383f.).
Later, in the fatal walk on the carpet, ‘trampling’ becomes
visible and literal. Whoever ‘the trampler’ was at 369f., here he
is Agamemnon, guilty, as the imagery has suggested, of ‘tram-
pling’ the inviolate beauty of his daughter and the population
and the sacred places of Troy. Tempted by his wife, he now liter-
ally tramples on a physical object that should be ‘inviolate’ too
because of its cost. His action is a symbolic enactment of his
crimes and the last example of his hybris. Aeschylus completes
the image of trampling neatly, by giving Agamemnon ‘tram-
pling’ (patôn, 957) as his final word before he goes in to meet
Clytemnestra’s justice.
105
The net
Leavis’ dictum is very relevant here: since nets were widely
used in antiquity for hunting as well as fishing, net imagery
inevitably connects with another image system, ‘the hunt’. ‘The
hunt’ includes the tracking and pursuing of prey, the eagles’
hunt of the hare, Clytemnestra as a hunting dog, as a watchdog
of the house, and as a hateful bitch, and the Fury or Furies as
pursuers of the wicked. Hunting culminates in a kill, and for
this Aeschylus has substituted the pervasive imagery of sacri-
fice (as well as a literal human sacrifice).The connections made
through imagery are not to be severed. However, one aspect of
the net image, which is perhaps one of the most prominent in
the play, is briefly discussed here.
Aeschylus uses at least a dozen different nouns meaning
‘net’, as well as much associated vocabulary conveying the ideas
of entanglement, taming and subduing. Just one function of the
net imagery (as with ‘trampling’) is to make an indirect causal
link between Paris’ crime, the fall of Troy and Agamemnon’s
murder. The image makes a prominent first appearance in
condensed form at the opening of the first stasimon, which
seems to begin as a plain victory ode (355f.):
O Zeus and you Night, possessor of great ornaments! Since

indeed you (Night) have thrown an enveloping net over the
towers of Troy so that no one, full-grown or young, can surmount
disaster’s (atê) all-catching net of slavery; and truly I revere
great Zeus God-of-Host-and-Guest who has achieved these
things, aiming his bow at Paris ….
Here Night and Zeus co-operate: Night throws the net, while
Zeus makes the kill. Much later, in her speech of triumph over
Agamemnon’s body (1372f.), Clytemnestra says:
How else could one build up a net of harm, at a height too great
for him to leap out? … I put around him a boundless net, like
that for fishes, the evil wealth of a robe, and I struck him three
times in gratitude to Zeus-Who-Keeps-the-Dead-Safe.
106
The ‘net’ of Night and the ‘net’ of Clytemnestra echo one another;
both are insurmountable and all-enveloping, blows follow, and
Zeus in different aspects attends. The audience can see the net as
she speaks – an extraordinary effect. Again a mental image has
culminated in physical presence. Net imagery paradoxically
conveys the influence of Zeus first with, then against,
Agamemnon. The net by which the king killed/conquered has now
killed/conquered him.
Viewed like this in isolation, the two net images seem to
uncover a contradiction. But intervening associated images
have made the movement from one to the other comprehen-
sible, by revealing a whole world of entanglement and
subjugation from which no one is exempt. A ‘curb or ‘horse’s
bit’ (132) silences Iphigeneia; a gag is put round her ‘lovely
mouth’ (235-6); Clytemnestra cruelly speculates that
Cassandra, like an untamed horse, ‘does not understand how to
endure the rein before foaming away her spirit in blood’ (1066-
7); Agamemnon captured Cassandra (like Troy) ‘within a
fateful net’ (1048), so she must wear the ‘yoke of slavery’ (1071,
1226); finally this fate spreads to the whole populace, since
Aegisthus is ready to ‘yoke the disobedient with heavy chains’
(1639-40).
After resolving on his daughter’s sacrifice, Agamemnon
put on ‘the yoke-strap of Necessity’ (218). Cassandra foresees
his ultimate entanglement: ‘she (Clytemnestra) stretches out
hand after hand, reaching them out’ (1110), surely the action
of drawing in a huge net; then more clearly (1114f.), ‘Ah,
what is this appearing? Can it be some net of Hades? No, the
net is his wife ….’ The grieving chorus see Clytemnestra as
the spider in the web where Agamemnon lies (1492, repeated
1516); earlier, Clytemnestra had jauntily asserted that if
Agamemnon had been wounded as many times as rumour
reported, ‘he would have been more full of holes than a net’
(868). At the end of the play, Aegisthus describes Agamemnon
as lying ‘amid the woven robes of the Furies’ (1580), and ‘in
the nets of Justice’ (1611). He is himself a ‘ “righteous
stitcher” of the murder’ (1604). The next play will show the
107
current victors entangled and killed in their turn: there is no

escape from the net.
108
The Reception of Agamemnon
Malevole: Egistus, didst ever hear of one Egistus?

Mendoza: Gistus?
Malevole: Ay, Egistus, he was a filthy incontinent fleshmonger,
such a one as thou art.
John Marston, The Malcontent (1604), 1.5.2-13
Introduction
In the light of reception studies, Agamemnon, traditionally
regarded as a ‘fixed’ text, is only the starting point of myriad
fluid re-workings, made in different circumstances in different
European countries over two and a half millennia in all imag-
inable media – painting, etching, opera, ballet, mime, puppetry,
film; burlesque and parody as well as (especially over the past
few decades) serious drama. Sometimes the trilogy has been
reduced to two plays, sometimes an extended sequence has
been created using Euripides’ Iphigeneia in Aulis and
Sophocles’ Electra. The play has inspired re-workings in which
Aegisthus, Electra, Iphigeneia or Cassandra are central
figures; or Thyestes, who is only glancingly referred to by
Aeschylus, may be the centre of interest. Dramatic productions
raise many issues of interest: original language versus vernac-
ular, translation versus adaptation; concepts of
‘appropriation’, ‘refigurisation’ or even ‘foreignisation’ may
arise. Currently, in a new wave of interest, versions of
Agamemnon are being performed in every corner of the globe
(including post-colonial countries), and the play is in the fore-
front of theatrical developments.
109
From among the diverse play versions of Agamemnon down

the ages there are perhaps two constant threads of interest.1
The first is that the death of Agamemnon, causing political
rupture, repeatedly lends itself to contemporary analogy:
versions have mirrored political crisis in revolutionary France,
America during the Vietnam War and a communist counter-
coup in Russia. Citizen Lemercier produced an Agamemnon in
the Théâtre de la République in Paris in 1797; in America in
1973, David Rabe’s The Orphan (one of his Vietnam Plays
which borrowed from Iphigeneia in Aulis as well as the
Oresteia) portrayed Agamemnon as a caricature of Lyndon
Johnson (the current President), while Aegisthus became a
Nixon figure who cut off Electra’s hands and tongue in prison
to silence her opposition; Stein’s ‘democratic’ German produc-
tion was revived to play in Moscow during anti-communist
activity against Yeltsin in 1994.
The second thread is Clytemnestra: how ruthless and inde-
pendent a murderess is she allowed to be? In line with the
receiving culture’s perception of women,2 Aeschylus’ heroine
has been frequently toned down, so that she only partners
Aegisthus in her husband’s murder. Sometimes she is given
those qualities of maternal and moral anguish so splendidly
absent in Aeschylus’ creation. In these post-feminist times,
Clytemnestra is more likely to be allowed once again to revel in
her solo act and be an archetypal Bad Woman.
An outline history of performances of Agamemnon and
related play-texts is given in the Chronology on pp. 150-2. The
information is taken from the database of the Archive of
Performance of Greek and Roman Drama (APGRD), compiled
by Amanda Wrigley and published in ch. 19 of F. Macintosh
(ed.), Agamemnon Staged: Proceedings of the Agamemnon
Conference 2001. This volume is a superb resource, and an
obligatory starting-point for further research into different
areas of the play’s reception.
Rather than summarising material from Agamemnon
Staged, this chapter concerns itself with the way reception, like
the flow of water along a river, can provoke one and then
110
6. The Reception of Agamemnon
another effect – and continues to do so long after the original

source has been joined by other waters. What can we distin-
guish of Agamemnon once it is diluted in later tragedy, or when
it seems to be present only in homoeopathic quantities? Or
when the resemblance is fortuitous? Here we must draw on
phrases such as ‘deep source, resource, influence, confluence,
tradition, heritage, origin, antecedent, precursor, background,
milieu, subtext, context, intertext, affinity, analogue’:3 there
are, after all, a multitude of possible ways one text may relate
to another.
This chapter explores Seneca’s ‘remake’ of Agamemnon 500
years later in Rome and then, 1,500 years or so after that, some
‘Senecan’ revenge-dramas, including Hamlet, written in
England at the end of the sixteenth century and the beginning
of the seventeenth. Seneca may or may not have been influ-
enced by Aeschylus’ play; Tarrant, who edited Seneca’s text in
1976, talks of ‘an almost complete absence of similarity in
structure and characterisation’.4 Yet Seneca was undoubtedly
aware of it.
By contrast, it is extremely unlikely that any Elizabethan
playwright had access to Aeschylus’ Agamemnon in a recognis-
able form. Greek authors were read in Latin translation, and
six Aeschylean tragedies, rendered into Latin by Jean
Sauravius or Saint-Ravy, had been published in Basel in 1555
and were being read in England and the Continent. However,
the translations were based on Turnèbe’s Greek edition of 1552
(Paris), which derived from the editio princeps, the Aldine
edition of 1518, and this had used a form of the Oresteia
deriving from an incomplete manuscript. In this edition, the
trilogy consists of only two tragedies, Agamemnon and
Eumenides. Lines 311-1066 and 1160-1673 are missing from
Agamemnon and the remaining text has been fused with that of
Libation-Bearers. So a very different Agamemnon results – in
which Agamemnon himself is entirely absent. ‘It means that
only Cassandra alights from the ‘vehiculum’ and that the
reader is suddenly thrust from her foreshortened dialogue with
the chorus to the appearance of Orestes.’5 Yet, despite the
111
supremacy of Seneca for the Elizabethans, Ewbank is not alone

in arguing that Aeschylus (in this form) had ‘surely gone into
the matrix of Shakespeare’s imagination’, and despite being
rooted in utterly different societies and performance conditions,
some connection between Aeschylus’ Agamemnon and Hamlet
is worth arguing for.
Theatrical and cultural change after Aeschylus

From the fourth century BC onwards our evidence for drama
becomes patchy. In Athens we know that new tragedies
continued to be written for centuries, but also that an official
contest for revived tragedy was instituted in 386 BC. No
mention of a revival of Agamemnon happens to survive, but
Aristophanes’ Frogs suggests reasons why, despite the language
possibly seeming old-fashioned, the play could have acquired
classic status: its text was, after all, to survive through to the
Byzantine era along with only six other plays of Aeschylus.
Perhaps scholars and academics, rather than theatre-practi-
tioners, ensured survival.
Meanwhile drama, particularly that of Euripides and
Menander, was spreading to theatres all over the Greek world,
and attaching itself to other public events and festivals other
than those in honour of Dionysus. Tragic performance became
a more fluid concept: individual actors acquired prestige and
travelled extensively with their own troupes; the elements of
song, dance and spectacle in tragedy increased and acquired
separate importance: virtuosi singers, performing lyrics from
dramatic texts (and even passages originally composed for
recitation), were greatly admired for their technique and
emotional expression; in other kinds of production choruses
detached from the action to become a separate interlude. Over
time, ‘tragic performance’ might take one of a variety of forms,
including a single actor reciting, singing (or, in Roman times,
miming to a chorus) celebrated sections from a single tragedy –
or even several together: as Gentili remarks, Hellenistic culture
(as much in the schoolroom and in higher education as in the
112
theatre) tended to be ‘anthological’.6 This is an age of epitomes

and florilegia, collections of speeches on a single topic, antholo-
gies of sayings (Greek gnômai, Latin sententiae or ‘dictes’, as
they were called in Elizabethan English).
Roman society was subject to Greek influence from as early
as 250 BC and at once used texts based on Greek theatre mate-
rial, lightly or heavily adapted for the different cultural context.
Latin writers began both to translate (vertere) and to imitate or
compose (scribere ex novo). While no single tragedy from the
Hellenistic, Republican or Augustan period survives complete,
we can be sure that Roman playwrights continued the well-
established tradition of selecting elements from previous
tragedies and putting them together with their own innovations
to create something new.
Theatrical Rome
When Seneca composed his Agamemnon (between 49 and 65
AD), imperial Rome was in many ways a highly theatricalised
environment. There were three stone theatres, as well as
numerous ad hoc wooden ones equipped with a stage curtain.
High masks and boots were now worn. There was a continuing
taste for spectacle: Cicero tells us that in the revival of Accius’
Clytemnestra, 600 mules accompanied Agamemnon on stage.7
Theatricality was not restricted to the stage: drama sat along-
side mass entertainments, mock sea-battles, wild animal
massacres and gladiatorial combats to the death. Society at
large was theatrical in the sense that legal cases were pleaded
publicly in the forum by lawyers highly trained in the tropes of
rhetoric; higher Roman education largely consisted of declama-
tory role-playing exercises, which sometimes included the
impersonation of mythical figures pleading their cause in high
style with great emotion (suasoria: an exercise largely identical
to writing one’s own tragic role).
In this theatrical city, Nero himself, emperor 54-68 AD, was
avid to display his own Thespian abilities (to the horror of
respectable opinion). It is perhaps not irrelevant that before
113
Seneca took on the role, his earlier tutor had been a

pantomimus (ballet-dancer). On first becoming emperor, he
recited his own poetry in the theatre; during his reign among
much else he enclosed the future site of St Peter’s to display
himself as a charioteer, and introduced a drama festival at
which he both presided in his box and participated on the stage,
acting or singing to the lyre. Pre-paid applause helped to ensure
that he won all possible prizes. Narcissism and megalomania
apart, Nero perhaps grasped that theatre could be manipulated
to enhance his prestige.
Seneca
Ovid had already capitalised on his legal training in suasoria to
create heroines in a dilemma who subtly, wittily and with artful
‘naivety’ reveal the self through declamation. Seneca too is a
master of this art. By training, educated Roman audiences would
have appreciated his characters’ emotive powers as they deliv-
ered intellectually demanding, brilliant and ingenious speeches
full of recognisable rhetorical features such as hyperbole,
paradox, compression, assonance, balance and antithesis,
together with abundant reference to myth. The less well-
educated could respond to the emotional theatricality and
grotesque murders. It remains unclear whether Seneca wrote for
acting on the public stage or for declamation in auditoria. The
text leaves some entries and exits unclear and there are problems
with staging; however, the play is certainly performable.
It is fascinating to read off Seneca’s Agamemnon against that
of Aeschylus. The ordering of the first few scenes seems influ-
enced by the ‘original’, but the play then takes an unexpected
turn by introducing Electra (evidence suggests that she had
appeared in both Livius Andronicus’ and Accius’ versions), and
by giving Cassandra a continuing role: now not only
Agamemnon, but Clytemnestra herself pales in significance as
Seneca attempts to balance out the play by making an ‘equation
of conquerer and conquered’ (Tarrant). From the start Seneca
weaves in plot-lines, motifs and echoes from other tragedies,
114
Greek and Roman, as well as topoi from Augustan poetry. Many

other influences must necessarily be lost to us.
The play is laid out below so that it can be contrasted with
Aeschylus’ version. It is in five acts, the arrangement recom-
mended by Horace, Ars Poetica 189ff., and already visible in
Menander.
Seneca’s Agamemnon
Act One
1-56 Prologue
The avenging ghost of Thyestes appears from Hades. He explains
that after Atreus forced him at a banquet to eat his own sons,
Fortune (= an oracle) told him to beget an avenger through incest
with his surviving daughter – and so Aegisthus was born.
Agamemnon now returns from Troy, so Aegisthus must act. (This
story-line may be influenced by Sophocles’ lost Thyestes plays.)
A family ghost announcing revenge replaces the superstitious,

ignorant Watchman. (Seneca possibly imitates Polydorus’
avenging ghost at the beginning of Euripides Hecuba, merging
him with the avenging gods that begin Hippolytus and
Bacchae.) Through Thyestes, Seneca at once reveals the action
of the play and the identity of one of the murderers – a great
contrast to Aeschylus’ obfuscating opening strategy.
Thyestes makes no mention of Olympian gods, later called
‘fickle’ (606, 930). Contrast theous, ‘gods’, the first word of
Agamemnon, together with its opening description of the
ordered regulation of the heavens, and the dense references to
Zeus throughout. The torments in Hell of the condemned
sinners Ixion, Sisyphus, Titylus and Tantalus, whose company
Thyestes has just left, replace Aeschylus’ Olympians, in whom
so many hopes and fears had resided. In the low horizon of
Seneca’s world, Thyestes loathes both Hades and earth (3).
Seneca’s play-world is virtually godless and deeply
pessimistic. It begins and ends in moral chaos. Characters are
overwhelmed by frustration, despair and uncertainty and their
triumphs are futile and temporary. Not even Clytemnestra
115
cares if she dies (202). Although there is a ‘positive’ counter-

revenge in the play (in the sense that much is made of
Cassandra’s perceived ‘victory’ at Agamemnon’s death), there
is none of the Oresteia’s positive movement away from personal
revenge to the impartial justice of Olympian Zeus.
57-107 First Chorus

Uncharacterised Argive women (note their changed gender and
detachment from any of the following scenes) develop two further
framing themes. In extreme contrast to the complex, imagistic
narrative of Aeschylus’ parodos, these are plain and clear: (1)
kings and courts are corrupt; (2) Fortune always reverses (101-2):
Quidquid in altum Fortuna tulit,

ruitura levat.
(Whatever Fortune has raised on high, she lifts but to bring down.)
Fortune (Greek tychê) as a subject for reflection is not

Aeschylean.8 But here fallax fortuna, deceitful fortune (57-8) –
only in the negative sense of reversal from high state to low – is
presented as a universal law. Where the chorus of Aeschylus’
play had at length, and from different angles, probingly
enquired into the conditions under which a man will fall (e.g.
Ag. 750ff.), leaving the audience to form its own subjective judg-
ment, Seneca presents a closed system.
108-309 Act Two

This divides into two halves: Clytemnestra, with Nurse
dissuading her from action (108-225), and Clytemnestra, with
Aegisthus egging her on (226-309).
The Nutrix perhaps derives ultimately from the Nurses in

Euripides’ Hippolytus and Medea: probably already via
Euripides and Menander the Nurse had become the heroine’s
‘stock’ companion and dramatic foil.
At the point in the play where Aeschylus had positioned
116
Agamemnon’s dilemma, Seneca now sets Clytemnestra’s. She

begins with a soliloquy addressing her animus (self) (109-10,
112, 115):
Quid, segnis anime, tuta consilia expetis?

quid fluctuaris? Clausa iam melior via est.
… periere mores ius decus pietas fides –
… per scelera semper sceleribus tutum est iter
(Why, slothful heart, look for safe counsel? Why waver? The
better path is shut to you. …Good ways, justice, decency, holiness
and faith have gone … the safe way through crime is always by
means of crime.)
Totally contrary to the spirit of Aeschylus’ dramatic strategy, the

Roman rhetorical tradition ensures that each stage figure imme-
diately exposes and explores their animus. Clytemnestra’s
thoughts are at once clear, as are those of Aegisthus, who enters
to address his soul in a similar opening soliloquy: quid terga vertis,
anime? ‘Why turn your back, my heart?’(228). Throughout the
scene, these emotional stances remain unchanged and in fact
much of their emotional expression is identical, differentiated
only by Aegisthus’ awareness of his doomed incestuous birth (non
est poena sic nato mori, 233, ‘for one so born it is no penalty to
die’) and Clytemnestra’s contrasting royal pride (e.g. 162-3, 290-
1). At one point she tells her lover that she still feels amor iugalis,
‘love for her husband’ and suggests they could ‘regain their inno-
cence’: yet all appeals to justice, decency etc. never seem to
operate as more than a decorative frill around the inevitable deci-
sion to murder. This evil inevitability contrasts, again, very much
with Aeschylus’ chorus, which had shown a profound engagement
with the problems of right action.
T.S. Eliot asserted that in Seneca, ‘the drama is all in the
word and the word has no further reality behind it. His charac-
ters all seem to speak with the same voice, and at the top of it;
they recite in turn.’9 This is so, yet the characters’ combination
of emotional chaos and detached intellectual analysis, articu-
lated sometimes with dazzling concision or again with thrilling
117
expansion, the whole conveyed through a ‘wonderful forward

thrust of the verse’10 is stirring and involving. And if there is
little overall unity, each scene is its own rhetorical tour de force.
310-388 Second Chorus

The chorus (barely celebrating Troy’s capture and certainly not
moving on to deeper, thematic reflection) sing a cult hymn in
praise of Apollo, Juno, Athena, Diana and Jupiter.
392a-588 Act Three

Clytemnestra with the Herald, Eurybates. He delivers a single
bravura messenger speech of 158 verses describing the return of
the Greek fleet (omitting Menelaus’ disappearance).
The positioning and opening section of this scene is very like

Aeschylus’ play. Seneca’s Herald greets his native gods while
Clytemnestra feigns delight at his headline news. His narrative
is generally cast as a reversal of Fortune from good to bad (and
ends by interplaying with many literary antecedents in its
description of the fate of Locrian Ajax and Nauplius’ revenge).
Coffey describes the speech as ‘an enormous rhetorical cadenza,
intrinsically brilliant, but too long to be accommodated to a
dramatical structure’:11 it may have been written with an eye to
independent performance.
589-658 Third Chorus

A subsidiary chorus of captive Trojan women, entering with
Cassandra, long for the peace of death and give a brief account
of Troy’s fateful last night.
The entry of captives and Cassandra ahead of Agamemnon may

be indebted to the entry of the captive women of Oechalia with
Iole among them at Sophocles Trachiniae 229ff., also ahead of a
returning husband. The Wooden Horse account has echoes of
Vergil Aeneid 2.239ff.; Vergil was himself probably influenced
by Euripides Trojan Women 522ff. and other lost sources,
including Cyclic Epic.
118
Eurybates had given no description of the fall of Troy

(contrast the many references by this stage in Aes. Ag., e.g. at
320-50, 355-67, 524-37, 555-66). The lack is now partly
supplied by this entry, which initiates Seneca’s different
design for the second half of his play: he plans that the
revenge on Agamemnon will be enacted literally by his wife
and his cousin, and virtually by the conquered Trojans. The
germ of this idea perhaps derives from Aes. Ag. 1279-80 when
Cassandra, envisioning Orestes, says, ‘Yet I swear by the gods
I shall not have died unavenged, for in turn another will
come ….’ The same theme is handled more allusively in
Euripides’ Trojan Women, where the women’s failure to exact
revenge in Troy is ironically offset by the audience’s fore-
knowledge of the Greeks’ shipwreck, which is a divine
punishment on them. Now Seneca makes Agamemnon’s
death ‘recompense’ the Trojans.
The chorus open on the appropriate theme of libera mors,
‘generous death’, a popular topos derived from Stoic and Cynic
thought (589-92):
Heu quam dulce malum mortalibus additum

Vitae dirus amor, cum pateat malis
Effugium et miseros libera mors vocet
Portus aeterna placidus quiete.
(Alas for our terrible love for life – what a sweetened evil given
mortals! – when escape from misfortune is available and
generous death, a haven peaceful in its stillness, calls to those in
misery. )
It seems typical that Seneca has replaced with a comparatively

impersonal choral topos the concrete, personal agony of
Aeschylus’ Cassandra (e.g. at Ag. 1136-9, 1146-9, 1258f., 1275f.,
1136-9, 1146-9) as she faces and finally walks in to her death.
659-807 Act Four

Cassandra, first with chorus in prophecy, then from 782 with
Agamemnon.
119
Cassandra now comes to the fore as she tears off her infulae
(fillets, 693): the same action that she makes at Aes. Ag. 1264ff.,
and already imitated by Euripides’ Cassandra at Trojan Women
451ff.12 At first calm, she declares that the fall of Troy is such
that Fortuna vires ipsa consumpsit suas, ‘Fortune herself has
exhausted all her powers’. The chorus then describe her
prophetic transfiguration (710ff.), an account that owes much
more to Vergil’s Sybil (Aen. 6.46ff.) than to Aeschylus. Cassandra
foretells Agamemnon’s murder, in which Clytemnestra acts as
sole murderess (Seneca repeats Aeschylus’ lioness/lion imagery
738-9). Finally, in a novel twist, she prays that the veil between
earth and Hades might be temporarily removed so that the dead
Trojans can witness Agamemnon’s murder and exult.
The chorus now makes an extraordinary entry-announce-
ment for Agamemnon (775f.). Where Aeschylus’ chorus had
chanted Agamemnon in over 27 verses, Seneca’s chorus says:
En deos tandem suos

victrice lauru cinctus Agamemnon adit,
et festa coniunx obvios illi tulit
gressus reditque iuncta concordi gradu.
(Lo! Clad in the victor’s laurel Agamemnon comes at last to his

own gods and his joyful wife has met him and returns, linked
with him in harmonious step.)
The chariot-entry, and the whole central episode of Aes. Ag.

(810-974, the confrontation of husband and wife including the
‘carpet scene’) are thus clumsily referred to and deleted. The
stagecraft here seems inept: ‘the apparent wish to avoid a
spoken confrontation between Agamemnon and Clytemnestra
has been most awkwardly combined with the need to account
for her movements at this solemn point’.13
The thrilling stichomythia that Aeschylus’ Agamemnon had
shared with Clytemnestra (Aes. Ag. 931-43) takes place instead
with Cassandra. Many identical words (marked in italics) are
batted to and fro between the speakers in Seneca’s brilliant
verbal display (791-9):
120
Ag. Festus dies est. Cass. Festus et Troiae fuit.

Ag. Veneremur aras. Cass. Cecidit ante aras pater.
Ag. Iovem precemur pariter. Cass. Herceum Iovem?
Ag. Credis videre te Ilium? Cass. Et Priamum simul.
Ag. Hic Troia non est. Cass. Ubi Helena est Troiam puto.
Ag. Ne metue dominam famula. Cass. Libertas adest.
Ag. Secura vive. Cass. Mihi mori est securitas.
Ag. Nullum est periclum tibimet. Cass. At magnum tibi.
Ag. Victor timere quid potest? Cass. Quod non timet.
(Ag. This is a feast day. Cass. It was a feast day in Troy too.
Ag. Let us worship at the altars. Cass. It was before altars my
father fell.
Ag. Let us pray alike to Jupiter. Cass. Jupiter of the Household?
(Priam fell before this altar)
Ag. You think you see Troy? Cass. Yes and Priam too.
Ag. There’s no Troy here. Cass. Where there’s a Helen there’s a
Troy.
(referring to Clytemnestra)
Ag. Don’t fear your mistress, though a slave. Cass. My freedom
approaches.
Ag. Live in safety. Cass. Death is my safety.
Ag. You’re not in any danger. Cass. You are though – great danger.
Ag. What can a victor fear? Cass. The fact that he is not afraid.)
808-866 Fourth Chorus

The main chorus sing of Argive Hercules, his twelve labours
(perhaps Seneca looked at Eur. Heracles 348ff.) and his previous
capture of Troy in the time of Laomedon.
Aeschylus had used Cassandra’s scene of prophecy to cover the

interval between Agamemnon’s exit and his murder. Since the
prophecy has already taken place, the choral song placed here
conveys a particularly strong sense of being an interlude.
867-1012 Act Five

The third episode divides into three sections: 867-909,
Cassandra’s soliloquy, which is a description of the murder;
910-952, the escape of Orestes engineered by Electra, due to the
timely arrival of Strophius, together with his non-speaking son
Pylades, returning victorious from the Olympian games in a
121
chariot (artful new use of the prop traditionally associated with

Agamemnon); 953-end, a confrontation between first Electra
and Clytemnestra, then these two plus Cassandra and
Aegisthus.
Cassandra describes the murder in the prophetic present tense:

significantly her ‘headline’ is that it is a revenge for Troy
(vicimus victi Phryges! … resurgit Troia. ‘We have conquered,
we conquered Trojans! Troy rises again’, 869-70). Her account
blends Homer’s banquet and Aeschylus’ robe: while feasting,
Clytemnestra makes Agamemnon don clothing that restricts his
head and hands (but is not described as a net); the ‘half-man’
(semivir, 890) Aegisthus ineffectually stabs him; Clytemnestra
almost entirely decapitates him with an axe. Amid much gore,
the two keep striking long after he is dead. Cassandra neatly
spells out the separate motives of each (986-7):
Uterque tanto scelere respondet suis –

Est hic Thyeste natus, haec Helenae soror.
(Each of the two makes answer to his kin in this mighty crime:
he is Thyestes’ son, she Helen’s sister.)
After Cassandra’s speech, apart from her own death, all the
material of Aeschylus’ play has been used up. Instead, in the
last 100 lines, Seneca introduces one major character (Electra),
one minor (Strophius) and two kôpha prosôpa (Orestes and
Pylades), and at the end, oddly, the murderers grapple with no
fewer than two defiant virgins.
The little ‘flight’ scene (first word Fuge, ‘Flee’) with its
supplication and rescue mimics motifs from many Euripidean
tragedies. Seneca’s strategy with Orestes deviates from that of
Aeschylus, who had made Clytemnestra herself report that she
had removed Orestes, 877f. Seneca instead creates a scene of
some immediacy with Electra as a new ‘heroine’ shown in the
act of saving her brother. He also vividly points to Orestes’
future revenge (merely mentioned by Aeschylus’ Cassandra,
122
1280ff.) by having the boy symbolically crowned with Strophius’

olive wreath and hidden behind his palm frond. Nonetheless,
Strophius’ so-timely arrival is a little gratuitous, as is the
appearance of no less than four unanticipated characters at this
late stage.
Between 953 and 980, Clytemnestra and Electra argue
(rather in the manner of their namesakes at Soph. Electra
516ff.), with some trenchant stichomythia. Electra refuses to
yield up Orestes, then voluntarily leaves the altar and offers
her body to her mother for death.14 Aegisthus appears and,
against Clytemnestra’s wish that he should decapitate Electra,
counter-proposes her perpetual immurement (cf. the punish-
ment of Sophocles’ Antigone, another defiant virgin). Exit
Electra under guard. Finally, Cassandra too spontaneously
leaves her refuge at the altar. She will gladly die at
Clytemnestra’s hands. The closing lines (1010-12) merit
quoting in full: the powerful antilabê of the final line is partic-
ularly fine (and the ‘madness’ picked up to be Cassandra’s final
word indicates the madness of Orestes, who will grow up to kill
both Clytemnestra and Aegisthus).
Cass. Nihil moramur, rapite, quin grates ago:

iam, iam iuvat vixisse post Troiam, iuvat.
Cl: Furiosa, morere. Cass. Veniet et vobis furor.
(Cass. No more delay! Seize me – really I thank you.

Now I am happy to have outlived Troy – yes, happy!
Cl. Die, mad creature! Cass. On you too will come a madness.)
Seneca’s Agamemnon is a palimpsest of classical and

Hellenistic elements, in which Aeschylus’ masterpiece appears
only fitfully. Its declamatory style and lack of unity reflect
contemporary Roman aesthetics, just as its view that death is
the only true freedom is derived from popular philosophy. The
play’s hellish world of violence, incest, cruelty and suffering is
perhaps not alien to the world of the late Julio-Claudian
emperors. Until the nineteenth century this version eclipsed
that of Aeschylus.
123
English Renaissance revenge dramas

Elizabethan London was a city in many ways not unlike Seneca’s
Rome. Barbarous spectacles included public hanging, drawing and
quartering, beheading and burning of religious dissidents.
Culturally too it was similar: grammar schooling was modelled on
Roman educational practice, and declamation was a large element
of the university curriculum and of public life. Oratory was prac-
tised in the pulpit; the Queen delivered orations in English and
Latin, and made a public pageant of herself and her court much
as Nero had done. Theatrical performances were popular both at
court and among the populace. From the late fourteenth century,
Seneca’s plays were regularly performed in Latin at university,
schools and Inns of Court. The first vernacular translation of his
plays appeared in 1581 (Newton’s Tenne Tragedies) and was
quickly followed by others. Seneca was the model for the English
Renaissance theatre between at least 1586 and the 1620s (though
this discussion does not include plays written after 1607). Seneca
informs Kyd’s A Spanish Tragedy (1586-7, very much the
‘founding father’ of this genre), Marlowe’s Jew of Malta (1594),
Shakespeare’s Titus Andronicus (by 1594) and Hamlet (1601-3),
Marston’s The Malcontent (1604), Tourneur and Middleton’s The
Revenger’s Tragedy (1607), Tourneur’s The Atheist’s Tragedy
(1611), Webster’s White Devil (1612) and Duchess of Malfi (1614),
Middleton and Rowley’s The Changeling (1621/2) and
Middleton’s Women Beware Women (?1623).
These plays are ‘neo-Latin’ in more ways than in being
frequently larded with tags from Ovid, Horace and Lucan as
well as Seneca: (for example the line per scelera semper
sceleribus tutum est iter at Ag. 115 appears both in The Spanish
Tragedy (misquoted, 3.12) and in The Malcontent (correctly,
5.3). Their subject is revenge (what Bacon called ‘wild justice’
in his essay Of Revenge), often multiple and interlocking
revenge (as attempted in Seneca’s Agamemnon, where the
secondary revenge of the Trojans is stressed). In these plays
revenge is frequently personified and becomes a stage figure:
Revenge sits on the stage with the ghost of Andrea watching the
124
action of A Spanish Tragedy; the hero of The Revenger’s

Tragedy is called Vindice; in Titus Andronicus 5.2, Tamora
visits Titus with her two sons; she is disguised as Revenge, they
as Rape and Murder. In the same general way, the hero of The
Malcontent is a deposed Duke who disguises himself as
‘Malevole’ to achieve revenge. Very often, Revenge appears as a
ghost from Hell: Hamlet’s father’s ghost comes from the under-
world enjoining him to ‘Revenge his foul and most unnatural
murder’ (1.5.25) very much like Seneca’s Thyestes: his respice
ad patrem (Ag. 52), ‘Think of your father!’ seems echoed by the
ghost’s ‘Remember me!’ (Hamlet 1.5.91).
Elizabethan revenge involves complicated trickery, elaborate
poisoning, rape, mutilation, cannibalism and bizarre jests with
corpses and body parts as well as plain murder: Seneca’s
Cassandra had described the repeated hacking of Agamemnon’s
body after he had been duped into putting on the restraining
robe (897-905). Now in Titus Andronicus, Lavinia is raped and
her hands and tongue cut out; later on Aaron the Moor promises
Titus that if he cuts off one of his hands, the lives of his two sons
will be spared: Titus cuts it off, but Aaron wittily returns it to
him together with his sons’ heads. Barabas and Ithamore in The
Jew of Malta strangle Friar Barnabas and prop him up against
the wall so that Friar Jacomo, innocently entering, is made to
believe he has killed him; in The Revenger’s Tragedy 3.5, Vindice
appears ‘with the skull of his love drest up in Tires’. He smears
poison on its lips, and in the darkness persuades the Duke to kiss
them. Even Hamlet has its skull and poison.
In this world of Senecan furor, violence may be turned on
itself, as when Hieronymo triumphantly bites out his own
tongue (Spanish Tragedy 4.4.191). Defiant passion prevails
over stoic acceptance and characters adopt madness or melan-
choly as a ruse (e.g. Hieronymo, Vindice, Titus, Hamlet) or they
become genuinely mad through suffering (e.g. Hieronymo’s
wife Isabella, Ophelia). When the boundary between real and
assumed madness disappears, Senecan declamation finds a
splendid new home in the transforming power of intense, unbri-
dled emotion.
125
Banquets and plays-within-plays

The ‘perverted banquet’ motif had appeared three times in
Seneca’s Agamemnon: Tantalus had cooked his son Pelops for a
banquet with the gods (19-21), Atreus had feasted Thyestes on
his murdered children (treated more fully Sen. Thyestes 641-
788, 920-1068) and, narrated during the play itself, is the
banquet at which Agamemnon dies. Grotesque and multiple
revenge murder at a banquet is an almost standard culmination
of these Elizabethan plays: at the end of Titus Andronicus,
Titus gets his revenge on Tamora by making her eat two pies
containing the heads of her sons; Barabas, in The Jew of Malta,
invites Calymath and his Bassoes to a banquet, having
contrived a ‘dainty gallery’ that will collapse and plunge them
into a deep pit; however, he falls into it himself, into a boiling
cauldron. Hamlet ends with a banquet at which there is a
poisoned rapier and a poisoned cup.
In A Spanish Tragedy, final revenge is achieved by a play
rather than a banquet. Hieronymo puts on a play, Solimon and
Perseda, whose plot imitates the ‘real’ background of the play.
Disguised as an actor, he is able ‘genuinely’ to stab the King’s
son. Plays and masques set within plays are not a Senecan
strategy but may nonetheless derive from the Senecan banquet:
both are court entertainments. With their theatrically potent
effects of double narrative, double role-play, concealed violence
and final unmasking they are deeply engrossing: the tragi-
comic Malcontent ends with a masque of four dancing Dukes
(including Malevole), with ‘pistolets and short swords under
their robes’: when the music stops, they have surrounded the
wicked Mendoza, unmasked themselves, and turned their
pistols on him. The joint authors of The Revenger’s Tragedy
employ an extremely complex variant of the same motif at 5.3.
Hamlet
In Hamlet the play-within-a-play idea gets a very different
treatment: the ‘performance’ is intended as a test, not as a
126
means to achieve revenge; the ‘actors’ in the embedded play

are not disguised stage figures; rather, the visiting players are
new and distinct personages, introduced as old acquaintances
of Hamlet’s (2.2.385f.). There are two phases to Shakespeare’s
use of the ‘embedded play’ motif. First, he heightens the crisis
of Hamlet’s delay by making Hamlet ask the First Player to
recite a speech (2.2.410-78) from a play which, as we learn,
mirrors his own situation. Pyrrhus’ murder of Priam is
another story of a son avenging a father’s death, and within
the speech, Hamlet’s hesitation is mirrored by Pyrrhus’: the
Player declaims, ‘… as a painted tyrant Pyrrhus stood,/ And
like a neutral to his will and matter/ Did nothing’ (437-9; the
unfinished line 439 lengthens the pause here). Then, ‘a roused
vengeance sets him new a-work’ (446).
Shakespeare uses this scene to elicit Hamlet’s complex self-
criticisms and doubts (soliloquy 2.2.501f., ‘O what a rogue and
peasant slave am I! Is it not monstrous that this player here/
But in a fiction, in a dream of passion/ Could force his soul so to
his own conceit …’). It is only now that he decides on a ‘trial by
play’ since, after all, ‘The spirit that I have seen (his father’s
ghost)/ May be a devil – and the devil hath power/ T’assume a
pleasing shape. Yea, and perhaps,/ … Abuses me to damn me’
(2.2.551ff.).
The Mousetrap, which Hamlet asks the players to perform,
is, he says, ‘a very choice Italian play’, ‘the image of a murder
done in Vienna’ – but also (he is using Hieronymo’s tactic) the
image of his father’s murder. Hamlet wants his own lines
inserted into the text, and it is an open question whether
Shakespeare temporarily provoked his audience to imagine that
Hamlet might, if he felt the case against Claudius sufficiently
proved at that point, rise up to challenge and kill him. Of course
we never know, because the performance is cut short. But we
know from the dumb show that the action begins with a loving
queen laying her husband to sleep; he is poisoned and the queen
is then successfully wooed by his murderer: surely in the
fashion of the plays of the day, the audience would then expect
a counter-murder?
127
While it is playing, Hamlet, Ophelia, Gertrude and Claudius

make running comments. This forces the embedded stage
action into ‘ “real” clock-time’, and there is an excitingly
heightened reality effect – created by this redoubled fiction of a
play-within-a-play. When Claudius interrupts it, there is a real
sense of shock as single ‘play-time’ re-establishes itself.
Hamlet now believes he has proof of his uncle’s guilt: ‘O good
Horatio, I’ll take the ghost’s word for a thousand pound’
(3.2.260). But since he misses Claudius’ confession before the
altar (3.3.36f.), he is excluded from the audience’s knowledge of
it. All this feeds into the dilemma of revenge presented in the
play through Hamlet’s indecision.
Hamlet and Aeschylus’ Agamemnon

Hamlet is much the most complex and fascinating of all these
Senecan dramas, with their Hadean revenge plots, exhilarating
rhetoric and grotesque spectacle. As superb theatrical devices,
the broken-off performance of The Mousetrap can be compared
with the carpet scene: both are richly theatrical and powerfully
ambiguous ‘staged’ tests of guilt, deceitfully enacted by one
character seeking revenge upon another, with the audience’s
fearful awareness that they cannot fully understand what is
happening, nor predict what will happen next.
As is often noted, Hamlet’s family story closely mirrors both
Agamemnons: a warrior king (Agamemnon/ Hamlet’s father) is
killed by a scheming cousin (Aegisthus/ Claudius) and an adul-
terous queen (Clytemnestra/ Gertrude), leaving the son
(Orestes/ Hamlet) to exact revenge. With Seneca there are some
interesting textual parallels too: in particular the mother-
daughter quarrel between Clytemnestra and Electra (Sen. Ag.
953-65) has been called a ‘striking anticipation’ of Hamlet 3.4.8-
18, a mother-son quarrel between Gertrude and Hamlet. Both
children challenge their mother with adultery, express loyalty
to their murdered father and contrast him with a degenerate
husband. Both passages are characterised by competitive
echoing.15
128
But despite points of similarity with Seneca, Hamlet

nonetheless transforms revenge drama. This transformation is
largely achieved through the particular nature of its central
figure. Miola sees Hamlet as a figure conflicted between Stoic
reason and the revenger’s passion. (A phenomenon possibly
arising from the rhetorical practice, common to Romans and
Elizabethans, of arguing in utramque partem, ‘both sides of a
case’.) Hamlet propounds paradoxes, quips and retorts in a way
that resembles the dictes of the philosophers; the Greek maxim,
‘know thyself’, adopted by Stoics (Sen. Consolatio ad Marciam
11.3) gives him the mandate for self-examination and medita-
tion which is the correct Stoic response to contumelia, ‘the
proud man’s contumely’ (3.1.71f.), the outrages of Fortune. He
praises Horatio for not being ‘passion’s slave’ (3.2.61-4). Yet
passion is exactly the theme he explores with the players, and
he is all passion at the loss of Ophelia. In short, he vacillates
between two Senecan ideals: the apatheia or detachment of
philosophy, and the passionate action of tragedy: between the
two, a new kind of hero is generated.16
Although this insight into Hamlet is based on two different
aspects of Seneca, it nonetheless provides the means of
bypassing Seneca and other Elizabethan revenge-dramas and of
tracing some direct confluence between Aeschylus’ Agamemnon
and Shakespeare’s Hamlet. Given that Seneca dramatised the
same myth as Aeschylus, and that it was Seneca that
Shakespeare knew, arguments for direct links based on resem-
blance of family or plot seem futile (even though mythical,
psychological and structural arguments have all been adduced
in the attempt).17 Even if Agamemnon and Hamlet can both be
called revenge dramas, they do not, after all, much resemble
one another.
However, there is one shared feature that it is well worth
arguing for: the nature of the central figures, the chorus and
Hamlet. Both, despite suffering acutely from apprehension and
uncertainty, are continually engaged in the struggle to make
sense of the world around them, to act rightly and to link
human life to their god or gods. Both, against all the odds, make
129
a serious appeal to the possibility of an ordered and uncor-

rupted society. Although Hamlet can actively seek order
through the creation of fictions like the Mousetrap, while the
chorus of old men can only try to understand by using tradi-
tional wisdom and their own reflections, yet the deep and
anguished reflection of each casts a brilliant moral and intellec-
tual light onto the deeds of revenge. It is here that Agamemnon
and Hamlet connect – in the stirring dramatisation of their
passionate mental effort.
130
Notes

1. Cf. Odyssey 4.351ff.
2. Herington, Poetry into Drama is the fullest treatment;
Easterling, ‘A Show for Dionysus’ discusses the extent to which
tragedy was ‘Dionysiac’.
3. Aristotle Poetics 4.1449a.
4. Garvie, ‘Aeschylus’ Simple Plots’, p. 105.
5. See Carey, Democracy 18ff. for a brief and lucid account of this.
Primary sources and secondary literature are to be found in Robinson
(ed.), Ancient Greek Democracy, pp. 76-151.
6. A theory discussed by Cartledge, ‘ “Deep Plays” ’, p. 22ff.
7. This was a form of glamorous public taxation, whereby wealthy
citizens financed something to the city’s benefit. Some of these leitour-
giai were military, others civic. Payment for the chorus, the most
expensive part of a drama production, was called chorêgia. On
chorêgia, see P. Wilson, The Athenian Institution of the Khorêgia: the
chorus, the city and the stage (Cambridge, 2000).
8. See Goldhill, ‘The Language of Tragedy’, p. 59ff. on this point. On
the arrangements for the City or Great Dionysia drama festival in
general, see Pickard-Cambridge, Dramatic Festivals (primary sources
and discussion, ch. 2, p. 57ff.) and Goldhill, ‘The Great Dionysia’.
9. This did not always continue to be the case in tragedy. Hall’s
Inventing the Barbarian (Oxford, 1989) describes the gradual appear-
ance of xenophobia and more blatant patriotism in the later years of
the fifth century. Mills’ book, Theseus, Tragedy and the Athenian
Empire (Oxford, 1997), explores the figure of Theseus in three
tragedies and concludes that the hero reflects ‘an unambiguously
favourable self-image’ which reaffirms for the audience ‘their heroic
ancestry and the Athenian way of life’ (p. 264).
10. Nussbaum, Fragility, Preface, p. xxvii.
11. For contrasting views on this see Henderson, ‘Women and the
Athenian Dramatic Festivals’, Transactions of the American
Philological Association 121 (1991), pp. 133-47 and Goldhill,
‘Language of Tragedy’.
12. Denniston and Page, Introduction, p. xxxix.
131
Notes to pages 17-38
13. For a discussion of the evidence for reconstructing the Danaid

trilogy, of which Suppliants was the first play, see A.F. Garvie,
Aeschylus’ Supplices: Play and Trilogy (Cambridge, 1969), p. 163ff.
14. See M. Griffith, The Authenticity of Prometheus Bound
(Cambridge, 1977). Evidence for the trilogy is discussed in M. Griffith,
Aeschylus Prometheus Bound (Cambridge, 1983), pp. 281-305.
15. A.N. Michelini, Tradition and Dramatic Form in the Persians of
Aeschylus (Leiden, 1982), well exposes the fatuity of assuming any
clear linear development in respect of the formal elements of the new
genre of tragedy.
16. Buxton, Complete World, p. 18.
17. This long-standing view receives recent endorsement from B.
Graziosi in Inventing Homer: The Early Reception of the Epic
(Cambridge, 2002), p. 246.
18. Athenaeus (fl. c. 200 AD), Deipnosophistae (= ‘The Learned
Banquet’) VIII 347e.
19. See Fraenkel’s valuable and sensitive analysis of this scene:
Fraenkel, Aeschylus Agamemnon, pp. 487-8.
20. Pickard-Cambridge, Dramatic Festivals, p. 139.
21. Given that Agamemnon contains some of the greatest surviving
poetry in ancient Greek, it is most regrettable that there is no space for
a proper discussion of metre in this book. However, there are several
good introductory handbooks on the subject, such as M.L. West’s
Introduction to Greek Metre (Oxford, 1986), as well as the excellent
metrical appendix in The Cambridge History of Classical Literature
(CHCL), vol. 1. Though metre appears a dry technical subject, the
ability to ‘hear’ the rhythmical patterns against the sound of the words
makes the effort highly rewarding. Ancient poetry, which was written
to move the ear not the eye, cannot be fully appreciated without it.
There is a tape of extracts from Agamemnon, performed by
Bradfield College. It is available for hire from Resources for Classics,
0845 456 0992, http://resources-for-classics.co.uk, info@resources-for-
classics.co.uk.
2. Theatrical Space
1. Taplin, Stagecraft and Greek Tragedy in Action.
2. Rosenmeyer, The Art of Aeschylus, pp. 116-17.
3. ‘Lure-murder’ is a term used to describe these deaths-by-deceit
in Goward, Telling Tragedy, pp. 32-6.
4. Knox, ‘Aeschylus and the Third Actor’, p. 42.
5. A glance at the lexicon entry makes clear the many senses of
oikos; there is also a good entry under household in The Oxford
Classical Dictionary (3rd edn 1996, ed. Hornblower and Spawforth).
132
See also the brief account in P.V. Jones (ed.), World of Athens
(Cambridge, 1984), p. 157ff.

1. See L.M. Slatkin, ‘Metis and Composition by Theme’, pp. 227f. in
S.L. Schein (ed.), Reading the Odyssey (Princeton, 1996) and C. Segal,
‘Kleos and its Ironies’, p. 208f. in the same volume.
2. The surviving fragments of the Epic Cycle, translated with
discussion, may be found in Davies, The Epic Cycle.
3. Pythian XI was probably first performed sixteen years before
Agamemnon in 474, although the scholiast does also offer a later date
of 454.
4. March, ‘The Creative Poet’, pp. 79-118.
5. Prince, Dictionary of Narratology, pp. 94-5.
6. Rabinowitz, ‘What’s Hecuba?’ and ‘Truth in Fiction’.
7. Goldhill, Reading Greek Tragedy, p. 6 terms the effect ‘the
progressive undercutting of the secure exchange of language’.
8. Goward, Telling Tragedy, p. 60f.
9. Rosenmeyer, Art, p. 189.
10. Goldhill, Language, p. 15.
11. Nussbaum, Fragility, p. 47.
4. Gods and Humans

1. See Lloyd-Jones, Justice.
2. ibid., 162.
3. Contrast the ‘realistic’ treatment of Euripides in the parodos of
Electra, where the chorus of Argive girls offer to lend Electra a dress
to wear at Hera’s festival.
4. Hesiod, Works and Days, 101.
5. ibid., 252-5.
6. The similarity is a modern view: the fact that olbos and ploutos
are contrasted within the same sentence at Persians 161-4 seems
rather to indicate a tradition of difference. As Harrison notes (corre-
spondence), there is ‘a principle of unknowability’ at work.
7. For thoughts on the possible course of the Danaid trilogy of which
Supplices is the first play, see A.F. Garvie, Aeschylus’ Supplices: Play
and Trilogy (Cambridge, 1969), p. 163ff.; for more on the Prometheus
plays, see M. Griffith, The Authenticity of the Prometheus Bound
(Cambridge, 1977), and pp. 281-305 in M. Griffith, Aeschylus,
Prometheus Bound (Cambridge, 1983).
8. Herington, ‘Aeschylus: The Last Phase’.
133
9. Winnington-Ingram, Studies; Podlecki and Stanford also adopt

their own versions of the traditional positive view.
10. Winnington-Ingram, Studies, p. 155.
11. Cohen, ‘Theodicy’.
12. Goldhill, Reading Greek Tragedy.
13. Shapiro and Burian, Aeschylus, The Oresteia.
14. ibid., pp. 6-7.
15. See Pindar Nemean IV 31-2 for a similarly worded sentiment.
16. Dawe, ‘Inconsistency’, p. 50.
17. Goldhill, Reading Greek Tragedy, p. 180.
18. Easterling, ‘Presentation of Character’.
19. See now also Easterling ‘Constructing Character’, pp. 83-99.
20. Easterling, ‘Presentation of Character’ p. 6.
21. Gould, ‘Dramatic Character’, p. 44.
22. ibid., pp. 59-60.
23. Rehm, The Play of Space.
24. Foley’s classic exposition of 1981, ‘The Conception of Women’,
has been followed by many others but is still worth reading.
25. Thucydides II.46.
26. See S. Blundell, Women in Ancient Greece (British Museum
Press, 1995), ch. 11.
27. ibid., p. 117.
28. See Carey, Democracy, p. 37.
29. P.E. Slater, The Glory of Hera (Boston, 1968).
30. See Lloyd, The agon in Euripides.
31. Though this is not invariably so, but will always depend on the
symbolic configuration of the theatre space. For an excellent view on
this see Easterling, ‘Women in Tragic Space’.
32. Hall, ‘Eating Children’.

1. Stanford, Aeschylus in his Style, p. 86.
2. Lebeck, The Oresteia.
3. Zeitlin, ‘The Motif’, p. 463.
4. Earp, Style, p. 98.
5. Goldhill, Reading Greek Tragedy.
6. Roberts, Apollo, p. 21.
7. Lebeck, The Oresteia, p. 31.
8. See Ferrari, ‘Figures in the Text’.
9. This receives a superb analysis from Knox, ‘The Lion in the
House’; see also Goldhill, Oresteia.
10. The simile is well analysed by Lebeck.
11. See Earp, Style, whose figures are quoted here.
134
12. See Goldhill, Oresteia, p. 74f., for a fuller analysis of the ambi-
guity and thematic significance of this passage.
13. Aristophanes, Frogs 1059.
14. Metonymy (literally, ‘change of name’) may be simply defined as
a practice of calling something not by its direct name but by something
with which it is associated. A poem by James Shirley (Song from The
Contention of Ajax and Ulysses) includes the lines ‘Sceptre and Crown/
Must tumble down/ And in the dust be equal made/ With the poor
crooked Scythe and Spade’. Here monarchs are designated by their
attributes of sceptre and crown, peasants by the scythe and spade.
15. See Stanford, Aeschylus in his Style.
16. Peradotto, Gantz.
17. Zeitlin, ‘The Motif of the Corrupted Sacrifice’ and ‘Postcript’.
18. Vidal-Nacquet, ‘Hunting and Sacrifice’.
19. Knox, ‘The Lion’.
20. Lebeck, Oresteia.
21. F.R. Leavis, ‘Imagery and Movement, Scrutiny, 1945.

1. Michelakis, ‘Agamemnon(s) in Performance’.
2. Hall, ‘Clytemnestra versus her Senecan Tradition’.
3. Miola, Shakespeare, p. 7.
4. Tarrant, Seneca Agamemnon, p. 215.
5. Ewbank, ‘Striking too Short’.
6. Gentili, Theatrical Performances, p. 21.
7. Cicero, ad Familiares 7.1.2, but very likely an exaggeration.
8. It seems to begin as a minor topos in Euripides’ ‘recognition-
thrillers’ (e.g. Ion 1512-15, Helen 711-12, Iphigeneia in Tauris 721-2),
where the vital identification of kin is made to hang excitingly on the
operation of chance.
9. Seneca, his Tenne Tragedies, edited by Thomas Newton, with an
Introduction by T.S. Eliot (London, 1927), Introduction, p. ix.
10. Herington, ‘Senecan Tragedy’.
11. Coffey, ‘Seneca and His Tragedies’.
12. See Easterling, ‘Agamemnon for the Ancients’.
13. See Tarrant ad loc., who also suggests that a dumb-show might
have taken place while these lines were spoken.
14. This seems a brief echo of Polyxena, Euripides Hecuba 557f. – in
both texts two alternate entry points for the weapon are suggested.
15. Miola, Shakespeare, p. 49.
16. ibid., p.54ff.
17. ibid., p. 49ff.
135
The Plays of Aeschylus
Extant plays are shown in bold, known satyr plays are marked by
an asterisk. Alternative titles result from the concern of
Alexandrian scholars to avoid confusion between two plays of the
same name. For many of these plays only the bare title or perhaps
a single line or two survives; sometimes lines survive which
cannot be ascribed to any particular play (known as adespota,
literally ‘without an owner’). This subject can be pursued in the
Loeb volume Aeschylus II by H.W. Smyth, with appendix and
addendum by H. Lloyd-Jones.
Extant plays
472: Persians
468-7: Seven Against Thebes
467-56: Suppliants
458: Oresteia trilogy: Agamemnon, Libation-Bearers, Eumenides
(late): Prometheus Bound (authorship disputed)
Attested tetralogies
472: Phineus, Persians, Glaucus of Potniae, *Prometheus Fire-Kindler
468-7: Laius, Oedipus, Seven Against Thebes, *Sphinx: about
Oedipus and Thebes
458: Agamemnon, Libation-Bearers, Eumenides, *Proteus: about
Orestes
(undated): Edonian Women, Bassarids, Neaniskoi (= Young Men),
Lycourgus: about Lycurgus
Probable groups of plays (chronology uncertain)

Suppliants, Egyptians, Danaids, *Amymone: about the daughters of
Danaus
136
The Plays of Aeschylus
Psychagogoi (Spirit-Raisers), Ostologoi (Bone-Gatherers), Penelope,

*Circe: about Odysseus
Prometheus Bound, Prometheus Set Free, Prometheus Fire-Bearer:
about Prometheus
Award of Arms, Thracian Women, Salaminians: about Ajax
Myrmidons, Nereids, Phrygians or The Ransom of Hector: about Achilles
Argo, The Lemnians, Hypsipyle, *Cabeiri: about Jason and the
Argonauts
Eleusinians, Argives, Epigoni (= Sons of the Seven): about mythical
Argive war with Thebes
Dictyoulkoi (Net-draggers), Polydectes, Daughters of Phorcis: about
Danae and Perseus
Memnon, Psychostasia (Weighing of Souls)
Women of Perrhaebia, Ixion
Mysians, Telephus
Remaining titles
Five plays dealing with the Theban myth of Dionysus: Semele or
Water-Carriers, Nurses of Dionysus, Bacchae, Xantriae, Pentheus.
The rest: Aetnaeae (= Women of Etna), Alcmene, Atalanta,
Athamas, Callisto, Carians or Europa, *Cercyon, *Cerukes (=
Heralds or Messengers), Cretan Women (in which the seer
Polyidus restores Glaucus to life), Cycnus, *Glaucus Pontius
(= Glaucus of the Sea), Heliades (about Phaethon),
Heraclidae (= Children of Heracles), Hiereiai (= Priestesses),
Iphigeneia, Leon, Nemea, Niobe, *Oreithuia, Palamedes,
Philoctetes, Propompoi, (= Processional Escorts), *Sisyphus
Drapetes (= Sisyphus the Runaway), *Sisyphus Petrokulistes
(= Sisyphus the Stone-Roller) (perhaps two titles for the same
play), Thalamopoioi (= Makers of the Bridal Chambers ?in
which the Danaids murdered their husbands: if so, probably
an alternative title for Egyptians), Theoroi (= Spectators)
probably the same play as Isthmiastae (= Spectators at the
Isthmian Games), Toxotides (= The Archer-Maidens) (about
Actaeon).
137
Outline of Agamemnon
1-39 Prologue
The Watchman on the roof of the palace at Argos sets the scene.
He has been waiting for news of Troy for nine years: now he
sees the longed-for beacon-fire.
40-103 Entry anapaests

The chorus of Theban elders says it is ten years since Menelaus
and Agamemnon set out for Troy; they themselves were already
past fighting age then. The outcome of the expedition is
unknown. Does Clytemnestra have information? She is silent.
104-257 Parodos
The chorus assume narrative authority. Ten years ago, as the
expedition readied for departure from Aulis, a portent appeared
of eagles devouring a hare, which the prophet Calchas inter-
preted as success for the army at Troy on the one hand, but the
anger of Artemis on the other. In the following so-called ‘Hymn
to Zeus’ (160-91) they ponder Zeus’ harsh but educative inten-
tions towards humans. The narrative continues. Artemis in her
anger kept the winds contrary, preventing the expeditionary
force from setting sail. Agamemnon made the decision to
appease her by the sacrifice of his daughter Iphigeneia.
258-354 First episode

(Clytemnestra and chorus)
Clytemnestra informs the chorus of Troy’s fall: in two speeches
she describes first the movement of fire across a series of
beacons linking Troy to Argos (the ‘Beacon Speech’), then the
contrasting fortunes of the winners and losers at Troy. She adds
138
that the Greeks must not offend the gods by desecrating reli-
gious spaces during their sack of Troy. At first sceptical, the
chorus say they are now convinced by her narrative.
355-488 First stasimon

The chorus celebrates the victory. Zeus, god of hospitality, has
now punished Paris’ transgression when, as Menelaus’ guest,
he abducted Helen. Those who sin against what is sacred will be
punished, and so will their kin. Wealth does not protect one
from Justice. Paris’ story exemplifies the pattern of crime and
punishment.
When Helen left for Troy, prophets were inspired and Menelaus
grieved. Because of Helen, many now grieve: War, like a money-
changer, has taken living men and given them back as ashes in an
urn. Relatives of the dead feel anger and the people are restless.
Gods and Furies are fully aware of those who kill many, and they
reverse the fortunes of those who prosper without justice. May
they themselves never be involved in a sacking. But can the news
– coming from a woman – really be true?
489-680 Second episode

(Herald and chorus, Clytemnestra)
Fresh from Troy, the Herald confirms the earlier news, unknow-
ingly also confirming the fear that Agamemnon incurred impiety
by destroying religious sites; in stichomythia the chorus
attempts to convey the idea of danger in Argos, but the Herald
blithely continues his narrative. Clytemnestra asserts that the
Herald’s news is confirmation of what she had said; soon
Agamemnon himself will be back. The Herald tells them that
Menelaus, however, has disappeared after a storm at sea.
681-781 Second stasimon

The chorus meditates on the appropriateness of Helen’s name
(the hel- prefix means something like ‘deadly’). They compare
her to a hand-reared lion cub – first charming, then murderous;
her arrival in Troy as a bride was the arrival of a Fury. Too
much prosperity (koros) leads to an act of arrogance (hybris)
139
and this in turn creates disaster (atê); the sequence is expressed

genealogically, and it works genealogically down human gener-
ations. Justice steers everything to its appointed end.
782-809 Entry anapaests

The chorus heralds Agamemnon’s triumphal entry on a chariot.
In veiled terms they urge him to test the loyalty of his subjects.
810-974 Third episode

(Agamemnon and Clytemnestra: silent chorus and silent
Cassandra)
Paying due respect to the gods, Agamemnon announces his
success at Troy and, in response to the chorus, his readiness to
deal with any dissidence in Argos. Clytemnestra describes her
extreme nervous anxiety during his absence, and explains that
their son Orestes has been taken for safe-keeping to Phocis. She
is ecstatic to have her husband safely home. Will he step into
the house on luxurious cloths? Agamemnon refuses on the
grounds that it is arrogant and impious to trample on the
wealth of the house, but is finally persuaded during a brief
section of stichomythia in which his wife verbally outwits him.
As he reluctantly walks up on the cloths to the central door of
the stage building, Clytemnestra outrageously compares the
riches of the palace to the inexhaustible wealth of the sea. In a
series of extravagant phrases she compares the king to a vine-
stock giving shade in summer, he is the sun’s warmth after
winter, the cool weather when grapes are pressed. He exits and
she follows. Both are inside the stage building, Cassandra still
on stage in the chariot (the second part of this episode is known
as the carpet scene).
975-1034 Third stasimon

The chorus describe the pounding of their hearts. Why should
they feel so disturbed since they have, after all, just witnessed
Agamemnon’s safe return? But only a hair’s breadth separates
a thing from its opposite, as is the case with health and sick-
ness. Some dangers can be averted or modified, as cargo is
140
jettisoned from a ship to ensure its safety. But Death cannot be

modified: the dead do not return to life. Inopportunely, their
heart races ahead of anything their tongue can express.
1035-1330 Fourth episode

1035-71 (Clytemnestra, chorus, silent Cassandra)
Clytemnestra re-enters to persuade Cassandra inside too for a
‘victory sacrifice’. The chorus tries to help, but Cassandra
remains entirely mute and unmoving.
1072-1177 Amoebean (Cassandra and chorus).

Cassandra at last breaks her silence, singing of her ruin at the
hand of Apollo and describing visionary fragments of the
house’s history: skewered babies eaten by a father, and now a
hideous new crime: the bull gored in the bath. She foresees her
own death. The chorus becomes increasingly agitated, but is
unable to comprehend.
1178-1330 Cassandra now tries again in iambic trimeters.

(The rest of this section consists of three rhêseis from
Cassandra followed by generous sections of dialogue.) She
describes a dancing troupe of Furies under the roof. She tells
them that after she rejected Apollo’s advances, he inflicted on
her the punishment that she should prophesy but not be
believed. Again she associates the past history of the house
and its murdered children with the coming revenge-murder:
no matter if the chorus cannot understand or believe her: it
is going to happen anyway. The chorus remain unenlightened
even when, in the second rhêsis (1214-41), Cassandra
describes further visions. In despair she tears off her
prophetic insignia. Composed again, she asserts that an
avenger will come after her death: Orestes, whose father’s
murdered body will exert a force drawing him home to kill his
mother. Three times Cassandra goes up to the entrance to the
palace only to turn aside with more words: the chorus express
profound sympathy for her coming death. Finally she enters
the stage building.
141
1331-42 Choral anapaests

Left alone on stage, the chorus’ apprehensions are growing:
they speculate that real danger awaits their king.
1343-5 Agamemnon (offstage) cries out as he is struck down.
1346-71 In paired trimeters, individual members of the chorus

react to the death cries with contradictory suggestions. No
consensus is achieved before:
1372-1406 Clytemnestra enters from the palace door together

with the corpses of Agamemnon and Cassandra. At last able to
speak the truth freely, she declares that she exults in the
murder of her husband.
1407-1576 Epirrhematic composition

(Clytemnestra, chorus)
The appalled chorus, in song, claims that the people will
sentence Clytemnestra to banishment. In calmer iambic trime-
ters she counter-claims that Agamemnon, by sacrificing
Iphigeneia, deserved to die. She has brought to fulfilment the
Dikê, Atê and Erinys claimed by her daughter. And as long as
Aegisthus ‘lights the fire on her hearth’ she is safe from punish-
ment. She exults again over both bodies.
The chorus longs for death now their king is dead – killed by
a woman, just as another woman, Helen, caused the death of so
many. Clytemnestra, now chanting in anapaests, rejects this
‘femme fatale’ rationalisation. The chorus’ next suggestion is
that the death was brought about by the daimôn of the house
which attacks every generation. Clytemnestra assents to this
line of approach, but this makes the chorus reflect on the role
of Zeus in all this: he is the cause of all, and their lives are all
pain. They grieve for their king.
Clytemnestra asserts her personal innocence of the murder:
she was the mere embodiment of the alastôr (avenging spirit)
set in motion by Agamemnon’s father Atreus when he forced
his brother Thyestes unwittingly to eat his own children and
142
was cursed by him. The chorus rejects this: she is certainly

guilty, and an alastôr may arise against her to avenge
Agamemnon’s murder (they are hinting at Orestes): but
Clytemnestra again reminds them of Iphigeneia’s claim. The
chorus feels the house is collapsing under a rain of blood: Dikê
is still sharpening her sword. And how will Agamemnon receive
proper burial ritual from loving kin? Clytemnestra replies that
his contact with kin will be with Iphigeneia in Hades.
The chorus sum up the debate. It is hard to decide between
the claims and counter-claims, but Zeus’ fixed law is that the
perpetrator must suffer. The house is wedded to its ruin (atê).
Clytemnestra, however, claims she will now be able to draw a
line under the past and make the daimôn of the house depart.
1577-1648
(Clytemnestra, Aegisthus, chorus)
Aegisthus exults over Agamemnon’s murder. He describes the
quarrel between their two fathers, and the moment when
Atreus falsely lured Thyestes back from exile and feasted him
on the flesh of his own children. Himself the sole survivor,
brought up in exile, he has returned and exacted Dikê. When
the chorus claims the populace will punish him he turns on
them with cruelty, reminding them of their physical powerless-
ness. He intends to ‘break’ the people to his will. The chorus
taunts him with cowardice in leaving Clytemnestra to do his
work for him; one day Orestes will kill them both.
1649-end Exodos
In trochaic tetrameters Aegisthus and the chorus draw their
swords but Clytemnestra intervenes. Still uttering threats and
insults, the chorus departs and Clytemnestra and Aegisthus
enter the house to assume control.
143
Select Bibliography
Abbreviations
AJPh = American Journal of Philology
AS = Agamemnon Staged: Proceedings of the Agamemnon Conference
2001 (Oxford, 2005)
BICS = Bulletin of the Institute of Classical Studies
CCGT = P.E. Easterling (ed.), The Cambridge Companion to Greek
Tragedy (Cambridge, 1997)
CPh = Classical Philology
CQ = Classical Quarterly
G&R = Greece and Rome
HSCPh = Harvard Studies in Classical Philology
ICS = Institute of Classical Studies
JHS = Journal of Hellenic Studies
PCPhS = Proceedings of the Cambridge Philological Society
TAPhA = Transactions of the American Philological Association
Greek texts with English commentaries

Denniston, J.D. and Page, D., Aeschylus Agamemnon, edited with
commentary (Oxford, 1957)
Fraenkel, E., Aeschylus Agamemnon, edited with translation and
commentary, 3 vols (Oxford, 1950)
Translations
Collard, C., Aeschylus, Oresteia (Oxford, 2002), accurate with helpful
introduction
Ewans, M., Aischylos, The Oresteia (London, 1995), written for perfor-
mance
De May, P., Aeschylus, Agamemnon (Cambridge, 2003), accurate, with
simple running critique
Fagles, R., Aeschylus, The Oresteia (New York, 1966), current Black
Penguin Classic; free
144
Select Bibliography
Harrison, T., The Oresteia (London, 1982), free

Hughes, T., The Oresteia (London, 1999), free
Lloyd-Jones, H., Agamemnon by Aeschylus, trans. with commentary,
Prentice-Hall Greek Drama series (New Jersey, 1970; reprinted in
Aeschylus Oresteia, London and Dallas, 1979)
Meineck, P., Aeschylus, Oresteia with introduction by Helene Foley
(Indianapolis, 1998), written for performance
Shapiro, A. and Burian, P., Aeschylus, The Oresteia (Oxford, 2003),
thoughtful translation with excellent introduction by Burian, very
useful full notes and glossary
Smyth, H.W., Aeschylus, Agamemnon, Libation-Bearers, Eumenides,
Fragments, with appendix and addendum by Hugh Lloyd-Jones, 1957
(Cambridge, Mass. and London 1929), close Loeb parallel text edition
Books and articles

Beacham, R.C., The Roman Theatre and its Audience (London, 1991)
Boyle, A.J., Tragic Seneca: An Essay in the Theatrical Tradition
(London and New York, 1997)
Burian, P., ‘Myth into Muthos: The Shaping of Tragic Plot’, CCGT, pp.
178-208
Buxton, R.G.A., The Complete World of Greek Mythology (London, 2004)
——— Persuasion in Greek Tragedy: A Study of Peitho (Cambridge,
1982)
Carey, C., Democracy in Classical Athens (London, 2000)
Cartledge, P., ‘ “Deep Plays”: Theatre as Process in Greek Civic Life’,
CCGT, pp. 3-35
Coffey, M., ‘Seneca and His Tragedies’, Proceedings of the African
Classical Association vol. 3 (1960), p. 16
Cohen, D., ‘The Theodicy of Aeschylus: Justice and Tyranny in the
Oresteia’, G&R (1986), repr. in I. McAuslan and P. Walcot (eds),
Greek Tragedy (Oxford, 1993)
Davies, M., The Epic Cycle (Bristol, 1989)
Dawe, R.D., ‘Inconsistency of Plot and Character in Aeschylus’,
PCPhS 189 n.s. 9 (1963), pp. 21-62
Earp, F.R., The Style of Aeschylus (Cambridge, 1948)
Easterling, P.E., ‘Agamemnon for the Ancients’, AS, ch. 1
——— (ed.), The Cambridge Companion to Greek Tragedy (Cambridge,
1997)
——— ‘Form and Performance’, CCGT, pp. 151-77
——— ‘A Show for Dionysus’, CCGT, pp. 36-68
——— ‘Constructing Character in Greek Tragedy’, pp. 83-99 in C.
Pelling (ed.), Characterization and Individuality in Greek Literature
(Oxford, 1990)
145
Select Bibliography
——— ‘Women in Tragic Space’, BICS 34 (1987), pp. 15-26

——— ‘Presentation of Character in Aeschylus’, G&R 20 (1973), pp. 3-
19
Ewbank, I-S., ‘Striking too Short at Greeks’, AS, ch. 3
Ferrari, G., ‘Figures in the Text: Metaphors and Riddles in the
Agamemnon’, CPh 92 (1997), pp. 1-45
Foley, H.P., Female Acts in Greek Tragedy (New Jersey and Oxford),
2001
——— ‘The Conception of Women in Greek Drama’, in H.P. Foley
(ed.), Reflections of Women in Antiquity (New York, 1981), pp.
127-68
Gagarin, M., Aeschylean Drama (Berkeley, Los Angeles and London,
1976)
Gantz, T.N., ‘The Fires of the Oresteia’, JHS (1977), pp. 28-38
Garvie, A.F., ‘Aeschylus’ Simple Plots’, in R.D. Dawe, J. Diggle and P.E.
Easterling (eds), Dionysiaca: Nine Studies by Former Pupils
Presented to Sir Denys Page on his Seventieth Birthday (Cambridge,
1978), pp. 63-86
Gentili, B., Theatrical Performances in the Ancient World: Hellenistic
and Early Roman Theatre (Amsterdam, 1979)
Goldhill, S., The Oresteia, Landmarks of World Literature series
(Cambridge, 1992; 2nd edn 2004)
——— ‘The Language of Tragedy: Rhetoric and Communication’,
CCGT, pp. 127-50
——— ‘The Great Dionysia and Civic Ideology’ in J.J. Winkler and F.
Zeitlin (eds), Nothing To Do With Dionysus: Athenian Drama in its
Social Context (Princeton, 1990)
——— Reading Greek Tragedy (Cambridge, 1986)
——— Language, Sexuality, Narrative: The Oresteia (Cambridge
1984)
Gould, J., ‘Dramatic Character and “Human Intelligibility” in Greek
tragedy’, PCPhS 14, pp. 43-67, repr. in Myth, Ritual Memory and
Exchange (Oxford, 2001)
Goward, B., Telling Tragedy: Narrative Technique in Aeschylus,
Sophocles and Euripides (London, 1999)
Hall, E., ‘Clytemnestra versus her Senecan Tradition’, AS, ch. 4
——— ‘Eating Children is Bad for You: The Offspring of the Past in
Aeschylus’ Agamemnon’, in D. Stuttard and T. Shasha (eds), Essays
on Agamemnon (Brighton, 2002), pp. 11-26
——— ‘The Sociology of Athenian Tragedy’, CCGT, pp. 93-126
Hardwick, L., Reception Studies, G&R New Surveys in the Classics no.
33 (Oxford, 2003)
Helm, J.J., ‘Aeschylus and the Genealogy of Morals’, TAPhA 134
(2004), pp. 23-54
146
Select Bibliography
Henry, D. and Walker, B., ‘Seneca and the Agamemnon: Some

Thoughts on Tragic Doom’, CPh 58 (1963), pp. 1-10
Herington, C.J., Aeschylus (New Haven and London, 1986)
——— Poetry into Drama: Early Tragedy and the Greek Poetic
Tradition, Sather Classical Lectures 49 (California and London,
1985)
——— ‘Senecan Tragedy’, Arion 5 (1966), pp. 422-71
——— ‘Aeschylus: The Last Phase’, Arion IV.3 (1965), pp. 367-403,
repr. in E. Segal (ed.), Oxford Readings in Greek Tragedy (Oxford,
1983)
Jenkins, I., ‘The Ambiguity of Greek Textiles’, Arethusa 18.2 (Fall
1985)
Knox, B., ‘Aeschylus and the Third Actor’, AJPh 93.1 (1972), repr. in
B. Knox, Word in Action (Baltimore, 1979)
——— ‘The Lion in the House’, CPh 47 (1952), repr. in B. Knox, Word
in Action (Baltimore, 1979)
Lebeck, A., The Oresteia: A Study in Language and Structure
(Cambridge Mass., 1971)
Lesky, A., ‘Decision and Responsibility in the Tragedy of Aeschylus’,
JHS 86 (1966), pp. 78-86, repr. in E. Segal (ed.), Oxford Readings in
Greek Tragedy (Oxford 1982)
Lloyd, M., The Agon in Euripides (Oxford, 1992)
Lloyd-Jones, H., The Justice of Zeus, Sather Classical Lectures 41 (Los
Angeles and London, 1971; 2nd edition 1983)
——— ‘The Guilt of Agamemnon’, CQ 12 (1962), pp. 187-99, repr. with
minor revisions in E. Segal (ed.), Oxford Readings in Greek Tragedy
(Oxford, 1982)
——— Agamemnon by Aeschylus, trans. with commentary, Prentice-
Hall Greek Drama series (New Jersey, 1970), repr. in Aeschylus
Oresteia (London and Dallas, 1979)
Lowe, N.J., The Classical Plot and the Invention of Classical Narrative
(Cambridge, 2000)
Macintosh, F. (ed.), Agamemnon Staged: Proceedings of the
Agamemnon Conference 2001 (Oxford, 2005)
March, J.R., ‘The Creative Poet’, ICS suppl. 49 (1987), pp. 79-118
Michelakis, P., ‘Agamemnons in Performance’, AS, Introduction
Miller, F.J., Seneca’s Tragedies I, II, Loeb Classical Library (London
and New York, 1917)
Miola, R.S., Shakespeare and Classical Tragedy: The Influence of
Seneca (Oxford, 1992)
Nussbaum, M.C., The Fragility of Goodness: Luck and Ethics in Greek
Tragedy and Philosophy (Cambridge, 1986; rev. edn 2001)
Peradotto, J.J., ‘Some Patterns of Nature Imagery in the Oresteia’,
AJPh 85 (1964), pp. 379-93
147
Select Bibliography
Peradotto, J.J., ‘The Omen of the Eagles’, Phoenix 23 (1969), pp. 237-
43
Pickard-Cambridge, A., The Dramatic Festivals of Athens, 2nd edn
revised by J. Gould and Lewis (Oxford, 1988)
Podlecki, A.J., The Political Background of Aeschylean Tragedy (Ann
Arbor, 1966; 2nd edn London, 1999), pp. 63-100
Prince, G., Dictionary of Narratology (Nebraska, 1987)
Rabinowitz, P., ‘What’s Hecuba to Us? The Audience’s Experience of
Literary Borrowings’, in S.R. Suleiman and I. Crosman (eds), The
Reader in the Text (Princeton, 1980), pp. 241-63
——— ‘Truth in Fiction: A Re-examination of Audiences’, Critical
Inquiry 4 (1977), pp. 121-42
Rehm R., The Play of Space (New Jersey, 2002), p. 76ff.
——— Greek Tragic Theatre (London and New York, 1992)
Rimmon-Kenan, R., Narrative Fiction: Contemporary Poetics (London
and New York, 1983)
Roberts, D.H., Apollo and his Oracle in the Oresteia (Gottingen, 1984)
Robinson, E.W., Ancient Greek Democracy: Readings and Sources
(Oxford, 2004)
Rosenmeyer, T.G., The Art of Aeschylus (Los Angeles and London,
1982)
Silk, M.S. (ed.), Tragedy and the Tragic: Greek Theatre and Beyond
(Oxford, 1996)
Solmsen, F., Hesiod and Aeschylus, Cornell Studies in Classical
Philology vol. 30 (New York, 1949)
Sommerstein, A.H., Aeschylean Tragedy (Bari, 1996)
Stanford, W.B., Aeschylus in his Style (Dublin, 1942)
Taplin, O., Greek Tragedy in Action (Oxford, 1978)
——— The Stagecraft of Aeschylus: The Dramatic Use of Exits and
Entrances in Greek Tragedy (Oxford, 1977)
Tarrant, R.J., ‘Senecan Drama and its Antecedents’, HSCPh 82 (1978),
pp. 213-64
——— Seneca Agamemnon, edited with a commentary (Cambridge,
1976)
Vidal-Nacquet, P., ‘Hunting and Sacrifice in Aeschylus’ Oresteia’, in
J-P. Vernant and P. Vidal-Nacquet, Tragedy and Myth in Ancient
Greece (Brighton and New Jersey, 1981), ch. 7; first published in
France as Mythe et Tragedie en Grece Ancienne (Paris, 1972)
Webster, T.B.L., ‘Some Psychological Terms in Greek Tragedy’, JHS 77
(1957), 149-54
Wiles, D., Tragedy in Athens: Performance Space and Theatrical
Meaning (Cambridge, 1997)
Winnington-Ingram, R.P., Studies in Aeschylus (Cambridge, 1983)
Wrigley, A., ‘Agamemnons on the Database’, AS, ch. 19
148
Select Bibliography
Zeitlin, F.I., ‘The Motif of the Corrupted Sacrifice in Aeschylus’

Oresteia’, TAPhA 96 (1965), 463-508, repr. in Zeitlin, Playing the
Other (Chicago, 1996)
——— ‘Postscript to Sacrificial Imagery in the Oresteia’, TAPhA 96
(1965), pp. 463-508
149
Chronology
Listed here are some key productions, adaptations and re-work-

ings of Agamemnon (sometimes Senecan rather than
Aeschylean) in original and vernacular translation. No school or
student production has been included, nor any of the large
number of operatic treatments; the list is biased towards English
language versions. These entries derive from the APGRD
(Archive of Performances of Greek and Roman Drama) database,
compiled by Amanda Wrigley.
BC
458: Oresteia by Aeschylus, Athens
452-451: Agamemnon by Ion of Chios, Athens (lost)
240-207: Aegisthus by Livius Andronicus, Rome (lost)
140-86: Clytemnestra by Lucius Accius, Rome (lost)
55: Accius’ Clytemnestra revived for gala opening of Pompey’s Theatre,
Rome
AD
49-65: Agamemnon by Seneca, Rome
1584: Agamemnon and Ulysses, Greenwich, London (lost)
1599: Agamemnon adapted by Henry Chettle and Thomas Dekker,
Elizabeth Rose Theatre, London
c. 1609-13: The Iron Age, including an adaptation of Agamemnon as
part of a series of four Ages, by Thomas Heywood; Red Bull,
London
c. 1609-13: The Tragedie of Orestes, original play in English by Thomas
Goffe based on Eur. Orestes, Sen. Agamemnon and Thyestes,
Shakespeare Hamlet and Soph. Electra; Christ Church, Oxford
c. 1619-22: The True Tragedie of Herod and Antipater by Gervase
Markham and William Sampson; this includes a ‘Dumbe Shewe’ of
the murder in Agamemnon; Red Bull, London
1738: Agamemnon, adapted from Seneca and Aeschylus by James
Thomson, Drury Lane, London
150
Chronology
1794: Clytemnestre, cantata for solo voice by Cherubini, France

1797: Agamemnon adapted by ‘Citizen’ Lemercier from Seneca and
Agamemnon, Théâtre de la République, Paris
1816: Agamemnon read by Napoleon to companions on Saint Helena
1842: Agamennone adapted by Alfieri, Teatro Re, Milan
1868: Agamemnon and Cassandra; or, The Prophet and Loss of Troy!,
burlesque by Robert Reece, Liverpool, Dublin and Portsmouth
1873: Les Erinnyes, adaptation of Agamemnon and Libation-Bearers
by Leconte de Lisle; revivals had music by Massenet. France
1880: May: Agamemnon, translated by L. Campbell, produced by
Fleeming Jenkin in his private theatre, Edinburgh
1880: June: Agamemnon in original Greek, produced by F. Benson at
Balliol, Oxford and then on tour
1885: The Orestean Trilogy of Aeschylus in translation, produced in
Cambridge by Benson and then on tour
1886: The Story of Orestes (abridged version of Oresteia), translated
and directed by G.C.W. Warr, Piccadilly, London
1904: The Orestean Trilogy of Aeschylus performed in London and on
tour
1931: Premiere of Eugene O’Neill’s Mourning Becomes Electra, an
adaptation of the Oresteia, Guild Theatre, New York
1936: Louis MacNeice’s translation of Agamemnon, directed by
Rupert Doone with music by Benjamin Britten, performed
Westminster Theatre, London
1937: Mourning Becomes Electra performed in London
1946: Louis MacNeice’s Agamemnon performed on BBC radio,
directed by Val Gielgud
1947: Film of Mourning Becomes Electra, directed by Dudley Nichols,
USA
1955: Mourning Becomes Electra, directed by Peter Hall, Arts Theatre,
London
1957: Agamemnon, translated by Richmond Lattimore, directed by
Wayne Richardson, Theater Marquee, New York
1958: Clytemnestra, three-act dance treatment of Oresteia, choreo-
graphed by Martha Graham, Adelphi Theatre, New York;
frequently revived and on tour in Europe
1962: Song of a Goat, version of Agamemnon by J.P. Clark, directed by
Wole Soyinka, Mbari Centre, Ibadan, Nigeria. Later performed in
London. New production, Lagos, 1991
1973: Agamemnon Part 1, adapted and directed by Steven Berkoff,
Round House, London and again in 1976
1973: The Orphan by D. Rabe, drawing on Oresteia and Eur.
Iphigeneia in Aulis, directed by Jeff Bleckner, New York; twice
revived in 1974
151
Chronology
1973: Appunti per un’Orestiade Africana; film directed by P.P. Pasolini,

shown first in Venice and then USA
1979: The Serpent Son, adapted by F. Raphael and K. McLeish, broad-
cast in three parts on BBC 2 television; new production in Sheffield,
1991
1980: Agamemnon, part of The Greeks, ten Greek tragedies adapted by
John Barton and Kenneth Cavander, directed by Barton; Aldwych
Theatre, London. Tours in USA in 1981, 1982, 1985 and 1986; many
revivals
1980: Die Orestie, translated and directed by Peter Stein; Berlin, then
Paris and Ostia
1981: Oresteia, translated by Tony Harrison, directed by Peter Hall
with music by Harrison Birtwhistle, Olivier Theatre, London and
on tour in Epidauros. Broadcast on Channel 4 TV, 1983
1987: Oresteia, translated by P.P. Pasolini, directed by L. Salveti,
Vicenza
1990: Les Atrides, adaptation of the Oresteia preceded by Eur.
Iphigeneia in Aulis. Translated by Jean Bollack, Ariane
Mnouchkine, Helene Cixous; directed by Mnouchkine. The Theatre
du Soleil performed in Paris and then toured internationally until
1993
1992: The Gift of the Gorgon, loose adaptation of Agamemnon by Peter
Shaffer, directed by Peter Hall, Pit Theatre, London, then
Wyndham’s
1992: The Clytemnestra Project: translations of Iphigeneia in Aulis,
Agamemnon and Electra, directed by G. Wright, Guthrie Theater,
Minneapolis; new productions in South Dakota and Los Angeles
1991, Massachusetts and Los Angeles 2001 and 2003
1993: Klytemnestra’s Bairns, adaptation of Oresteia by Bill Dunlop,
directed by Toby Gough, Edinburgh
1994: Peter Stein’s 1980 production of Oresteia revived with changes
and translated by Boris Shekassiouk for Moscow, then on tour
1996: L’Orestie, adapted and directed by Silviu Purcarete, Limoges,
France. Revised production given at National Theatre of Craiova,
Romania, and on tour in Britain, 1998
1999: The Oresteia: Part One The Home Guard, Part Two Daughters
of Darkness, translated by Ted Hughes, directed by Katie Mitchell,
Cottesloe Theatre, London
2002: Ariel, adapted from Iphigeneia in Aulis and the Oresteia by
Marina Carr, directed by Conall Morrison, Abbey Theatre, London
2003: Mourning Becomes Electra, directed by Howard Davies,
Lyttleton Theatre, London
152
Glossary
In the theatre
aulêtês: musician with a double pipe accompanying the sung parts of
the play
deuteragonist: the second actor
ekkyklêma: flat cart, wheeled out from the skênê to show a tableau
from inside
eisodos: wide entrance into the orchêstra, one on each side of the skênê
kôphon prosôpon: lit. ‘empty mask’, a non-speaking actor
koryphaios: the chorus-leader
mêchanê: a crane positioned behind the skênê, originally used for
divine epiphanies
orchêstra: round central area of the theatre
parodos: alternative name for eisodos
protagonist: the first actor
skênê: the stage building at the back of the orchêstra
tritagonist: the third actor
In the text
agôn: a scene of dispute between two characters
amoebaean: matching groups of sung verse assigned alternately to
two characters
anapaest: simple, heavily rhythmic metre suitable for walking to, and
so often used while characters process into or out of the orchêstra
antilabê: division of a line of verse between two speakers
distichomythia: dialogue between two characters in double alter-
nating verses (i.e. two each)
epeisodion or episode: the portions of dialogue between two choral odes
epirrhematic composition: a dialogue in which only one participant
sings
exodos: the closing section of the play leading to the exit of the chorus
kommos: lament sung alternately by one or more characters and the
chorus
153
Glossary
parodos: the chorus’ first entry-song

prologue: the material of the play preceding the chorus’ first entry
rhêsis: a continuous speech by one character
stichomythia: dialogue between two characters in single alternating
lines of verse (i.e. one each)
stasimon: choral song between episodes
154
Index
Aegisthus, 7, 23, 35, 42, 44, 54, alastôr (the driver-astray or

57, 58, 80, 88, 107, 109, 115, avenging spirit), 42, 57, 73, 76
117, 122f. APGRD, see Archive of
Aeschylus: life, 10f.; epitaph, 12; Performances of Greek and
output, 12, 16f., 136-7 Roman Drama
Agamemnon, king of Argos, 7, Apollo, 37, 38, 53, 56, 59, 70, 85,
9, 10, 23; chariot entry and 95, 102
barefoot exit, 32f.; death Archive of Performances of
cries, 38, 39; earlier treat- Greek and Roman Drama, 110
ments, 43-7; 52, 55, 58; his Aristophanes, 26, 27, 37, 102-3,
dilemma, 62f.; 66, 67, 70, 79, 112
85; reception of murder, 107, Aristotle, 19, 81, 93
115 Artemis, 46, 54, 55, 62, 70, 76,
Agamemnon: passim; ethical 85, 98, 99
concerns, 14, 62-3; legal atê (disaster, ruin), 32, 53, 73-5, 95
aspects, 14, 92-3; story of, 15; Athens, 10, 11, 13-14, 85-7, 92
formal structure and outline Atridae (Agamemnon and
of, 22f., 138f.; assignation of Menelaus, the sons of Atreus),
roles, 23; symbolic dimensions 7, 84-5, 101-2
of, 29f.; Beacon Speech, 31, audience, 15; authorial and
64, 98; chariot entry, 27, 32, narrative, 48f.
120; lure-murder, 33; carpet-
scene, 33f., 56, 89, 120, 128; banquets and plays-within-plays,
exodos, 42; scene-shape, 50; 7, 44, 126f.
causality, 54f.; entry- Burian, P., 79
anapaests and parodos, 60f.,
98f.; stichomythia (931-43), Calchas, 43, 55, 57, 61, 62, 64,
66, 95; characters in, 80f.; 67, 70, 98-100, 101
paired characters, 84; gender Cassandra, 10, 23, 28, 34, 36; exit,
in, 85; third stasimon (975f.), 38-9; 41, 43, 46, 49, 52f., 56-8,
95-6; reception of, 109f.; 67, 70, 80, 85, 87-9; denied
Seneca’s remake, 115f.; peithô, 97; final lines, 103; 107,
compared with Hamlet, 128f. 109, 111, 114, 116, 118f.
155
Index
causality, problems of, 54f. Eliot, T.S., 117

Choephoroe, see Libation-Bearers English Renaissance revenge
choral lyric, 10; as origin of dramas, 124f.
tragedy, 18f.; typical nature Ephialtes, 12
and capacities of, 19, 22 Epic Cycle, 45, 118
chorus of Agamemnon: passim, epic poetry, 10, 20f.; Aeschylus’
14, 20; reaction to carpet, 35; transformation of, 98f.
to death cries, 39-40; narrative Erinyes, see Furies
voice in opening anapaests and Eumenides, 9,12, 27, 41, 88, 92,
parodos, 59f.; as repositories of 111
traditional wisdom, 75f.; Euripides, 27, 35, 40, 46, 47, 51,
inability to grasp the truth, 54, 63, 76, 81, 109, 112, 115,
95f., 116 116, 118-22
chronology of productions, 150 Ewbank, I.-S., 112
Clytemnestra: passim, 7, 9, 23,
26; problematic entries and fallax fortuna, 116, 120, 129
exits, 29-30; control of focalisation, 57-8
threshold, 32, 33, 35; in Fraenkel, E., 23 n.19, 64, 80
carpet-scene, 33f.; with Furies, 9, 10, 27, 30, 39, 41, 50,
Cassandra, 36, 38-9, 52-3; in 53, 56, 67, 78, 96, 106, 107
tableau, 41; earlier
Clytemnestras, 43-7; 49, 52, gender, 65, 85f.
55-8; narrative voice, 63f.; 70, gods, 10, 30-2, 69f.
80; in Gould’s reading, 83f.; Goldhill, S., 60, 79, 82, 88, 92
paired with Helen, 84; as Gould, J., 82
murderess, 88f.; 94-5, 100-1, Great Dionysia, 11, 13f.
103, 105-7; in reception, 110f.
Cypria, 45, 46 Hamlet, 111, 126f.
hapax legomena, 100f.
daimôn, 42, 55, 57, 73, 76, 85; Helen, 7, 10, 31, 32, 45, 52, 57, 67,
daimones, 73f., 93, 104 76, 79, 84, 85, 89, 94, 100-2
defamiliarisation, 49 Herald, 23, 27, 31, 32, 51, 52, 56.
dêmos and compounds, 102 58, 64, 65, 70, 94, 118f.
Denniston, J.D. and Page, D., 30, Hesiod, 72f., 76, 84
65, 80, 95-6, 99 Homer, 15, 21f., 57, 72, 78; Iliad,
Dikê or Dika, see Justice 21, 22, 43, 74, 81; Odyssey,
discrepant awareness, 49f., 128 44f., 51, 98
dithyramb, 14 House of Atreus, 7, 30, 38f., 41,
42, 45, 47, 53, 56, 66-7, 72, 76,
Easterling, P.E., 81, 82 80
eisodos, 26f. human characters, 80f.
ekkyklêma, 26f., 41 hybris, 56, 72, 73, 74-5
156
Index
Hymn to Zeus, 54, 61, 77, 78 narrative voice, 57f.; of

chorus, 59f.; of Clytemnestra,
imagery, 91, 103f.; ‘trampling’, 63f.; embedded voices, 67f.
105; ‘the net’, 106f. Nero, 113-14, 124
Iphigeneia, 34, 41, 46, 54-5, 62f., net, 46, 106f.
65, 67, 70, 80, 85, 99, 100-1, nostos (return home), 9, 51, 52-3,
107, 109 59
Nussbaum, M., 14, 63
Jacobson, R., 97, 100
Justice (Dikê or Dika), 14, 50, offstage cries, 39f.
71f., 75, 77f., 105, 107 oikos (house, family), 10, 26, 38f.,
86
Knox, B., 37, 98 n.9, 104 orchêstra, 24f.
kôphon prosôpon, 27, 36f., 122 Oresteia, outline of, 9-10,12, 50,
koryphaios, 19, 20, 23, 59 72, 78, 87, 91, 116
Orestes, 10, 35, 37, 41, 42, 43, 44,
language, 82f., 91f.; choral and 50, 57, 66, 76, 111, 121-2
civic, 92f.; powers and dangers
of, 93f.; proper name puns, pathei mathos (learning through
93f.; euphêmia, dysphêmia suffering), 61-2, 69-70, 71, 77,
and silence, 30, 94; persuasion 80
and lack of it, 95-7 Panatheneia, 10, 11
Leavis, F.R., 105, 106 Pandora, 84
Lebeck, A., 91, 99, 102 Paris, 7, 31, 52, 76, 105, 106
Libation-Bearers, 9, 28, 35, 37, Peisistratos, 10, 11, 85
40, 41, 51, 53, 111 Peithô (Persuasion), 73, 75, 95,
libera mors, 119 97, 104
liturgy, 13 n.7 Pericles, 12, 13, 85
Lloyd-Jones, H., 71 nn.1&2, 78 Persians, 12, 17, 25, 33, 51, 100
lure-murders, 33f. Pickard-Cambridge, A., 23
Pindar, 12, 18, 46
mêchanê, 27 polis (community, society, city),
Menelaus, 10, 31, 52, 56, 64, 85, 10, 14, 79
101 portent of eagles, 55, 98f.
metre, 23 n.21 Prometheus Bound, 12, 16, 17, 77
Miola, R.S., 129f.
myth, 19, 20f., earlier myths of Rabinowitz, P., 48
Agamemnon’s homecoming, reception of Agamemnon, 109f.
43f. religion, 70f.
revenge, 124f.; its transformation
narrative technique, 43f.; ‘story’ in Hamlet, 129f.
and ‘plot’ contrasted, 49f.; Russian Formalists, 49
157
Index
sacrilege, 55f. Suppliants, 12, 17, 25, 77

satyr plays, 10, 14 suspense, 15, 47f., 59
Sauravius, 111
Seneca, 111, 114-15, 124, 129f. tableaux and mirror tableaux, 41
Seneca’s Agamemnon, 115-23, Taplin, O., 29, 30, 32
125-7 Tarrant, R.J., 111, 114
Seven Against Thebes, 12, 25 tetralogies, 16-17, 136-7
Shakespeare, 16, 124, 126f. text, problems of, 16, 28
silence, 30, 36-7, 37, 93-4 theatre: diagram of, 25; symbolic
skênê: development of, 24f.; use space, 24-7, 33f., 89-90
in Ag., 29f., 38, 39 Thyestes, 7, 42, 44, 53, 54, 109,
Slater, P.E., 87 115, 122, 126
Solon, 17, 74 time, 51f.
Sophocles,, 12, 19, 20, 26, 28, 35, tragedy: as poetry, 17-18; origins,
37, 40, 41, 46, 51, 81, 109, 18f.; structure of, 22-3
118, 123 trampling, 34, 105f.
Stesichorus, 18, 46
style, 97f.; transformation of Vergil, 118, 120
traditional features, 98-9;
violence of, 97f.; compound Watchman, 23, 29, 38, 50, 52, 69,
words and coinages, 98f.; 71, 91, 93, 101
metaphor and ambiguity, 98f., Winnington-Ingram, R.P., 78
102; catachresis, 100; personi- women, role of, 15, 85f.
fication, 104; see also imagery;
language. Zeus, 31, 45, 59-62, 69ff., 94, 99,
suasoria, 113, 114 106-7, 115-16
158

Aeschylus Agamemnon (Barbara Goward)

Uploaded by

Copyright:

Available Formats

Aeschylus Agamemnon (Barbara Goward)

Uploaded by

Document Information

Original Description:

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

Aeschylus Agamemnon (Barbara Goward)

Uploaded by

Copyright:

Available Formats

Aeschylus.

qxd 15/08/2005 15:50 Page 7

The ancient and appalling history of the House of Atreus, told

I would like to thank the following people: Ruth and Steven

Orientation: Aeschylus, Athens

There is the sea. And who will drain its resources?

Agamemnon is the first play of the Oresteia, our only surviving

these plays the story continues as follows: ‘in revenge,

Life and times

1. Orientation: Aeschylus, Athens and Dramatic Poetry

would had have the opportunity to attend performances by

We now enter the period of the seven surviving plays (see

This tomb covers the body of the Athenian Aeschylus, son of

1. Orientation: Aeschylus, Athens and Dramatic Poetry

Aeschylus had experienced a good deal in his lifetime: kings

The political and social context

The first contest was for choral song in honour of Dionysus

1. Orientation: Aeschylus, Athens and Dramatic Poetry

parts. Recent experience is mirrored in the play’s vivid descrip-

were to become almost standard in the hands of later theatre

Loss and survival

1. Orientation: Aeschylus, Athens and Dramatic Poetry

tragedies and a satyr play. Certainly Suppliants13 and less

of poetry were heard and seen in performance in settings such

1. Orientation: Aeschylus, Athens and Dramatic Poetry

flattering analogy between the voyage of the colony’s founder

moment in the text when a non-speaking actor playing

Myth and epic poetry

1. Orientation: Aeschylus, Athens and Dramatic Poetry

myth). Most of Aeschylus’ surviving plays use myths which

Despite the hierarchical world the heroes inhabit, there is a

The structure of Agamemnon

1. Orientation: Aeschylus, Athens and Dramatic Poetry

When, within an episode, a stage figure sings and the chorus

And now, my dear, step out of that chariot …

Imagine you are a director. Holding the script of Agamemnon in

The physical theatre at the time of Aeschylus

The ancient theatre in Athens. A = orchêstra; B = skênê; C = eisodos.

We need to pause here. The inability to see ‘inside’ (contrast

fully for their likely ‘shocking’ effect on the audience, and in

The use of theatrical space in Agamemnon

directions, so that each exit and entrance has to be inferred

problem is not a textual one but one of stage meaning:

The most striking stage action of this scene (if Taplin is

triumph over Troy. He reappears after ten years – but only to

they one continuous piece, consisting of several looms-worth

to make his walk correspond exactly to the fifteen allotted lines.

Silence and the kôphon prosôpon

begins a significant alteration in the prevailing character and

In Agamemnon, the significance of Troy, so important earlier,

flesh, which their father ate, and in the bathhouse a new

effect, of course, is to make the murder palpably immediate to

respond, Aeschylus introduces his next shock: Clytemnestra

the daimôn or alastôr (spirit or avenging spirit) of the House

The Story: Myth and Narrative

This chapter turns away from Aeschylus’ wizardry in the theatre

Literary precursors: earlier treatments of the myth

first to see Priam’s return from the Greek camp (24.697-701).

3. The Story: Myth and Narrative Technique