The Story of Myth

THE STORY OF MYTH
Sarah Iles Johnston
Cambridge, Massachusetts
London, England 2018
******ebook converter DEMO Watermarks*******

Copyright © 2018 by the President and Fellows of Harvard College
All rights reserved
Cover design by Lisa Roberts

Cover art: John Singer Sargent / Museum of Fine Arts, Boston. Francis Bartlett Donation of 1912 and
Picture Fund
978-0-674-18507-4 (alk. paper)

978-0-674-98955-9 (EPUB)
978-0-674-98956-6 (MOBI)
978-0-674-98957-3 (PDF)
The Library of Congress has cataloged the printed edition as follows:

Names: Johnston, Sarah Iles, 1957– author.
Title: The story of myth / Sarah Iles Johnston.
Description: Cambridge, Massachusetts : Harvard University Press, 2018. | Includes bibliographical
references and index.
Identifiers: LCCN 2018007634
Subjects: LCSH: Mythology, Greek—Comparative studies. | Discourse analysis, Narrative. | Heroes
—Mythology—Greece.
Classification: LCC BL783 .J64 2018 | DDC 292.1 / 3—dc23
LC record available at https://lccn.loc.gov/2018007634

To
Alex and Damian

Wolfgang and Max
Archer and Hugo
Orion and Remy
For what good is a storyteller, without an audience?

Contents
NOTE ON TRANSLITERATIONS AND ABBREVIATIONS
1 The Story of Myth

2 Ritual’s Handmaid
3 Narrating Myths
4 The Greek Mythic Story World
5 Characters
6 Metamorphoses
7 Heroes
Epilogue
NOTES
REFERENCES
ACKNOWLEDGMENTS
INDEX OF NAMES AND TERMS
INDEX LOCORUM

Note on Transliterations and Abbreviations
I HAVE USED A Latinate method of transliterating Greek names in most cases

(thus, “Cronus” instead of “Kronos,” for example), but in a few cases of
names that are more familiar in Greek transliteration, I have stuck with that
(thus, “Knossos” instead of “Cnossus,” for instance). Greek festival names
are usually transliterated in the Greek manner (“Synoikia”). When I
occasionally transliterate short phrases from ancient authors, I use a Greek
method. (In a few cases, I have presented words or phrases in the original
Greek because those who read that language will find it useful to see the
original Greek, but I have provided translations as well.)
In citing Greek sources, I have used the abbreviations found in Henry
George Liddell, Robert T. Scott, and Henry Stuart Jones, A Greek-English
Lexicon, 9th ed. In a few cases, I opted for a longer abbreviation in order to
make my notes more user-friendly to non-Classicists (thus, I use Bacch.
instead of B. for Bacchylides). I refer to the Homeric Hymns by the Greek
names of their honorees, rather than the names of Latin divinities: HHAphr.
instead of h.Ven, for example. When an abbreviation for a Greek source was
not provided by Liddell, Scott, and Jones, I used those in the Oxford
Classical Dictionary, online edition, 2012. Abbreviations of Latin authors
follow those in P. G. W. Glare, The Oxford Latin Dictionary, 2nd ed.
In citing ancient Near Eastern sources, whenever possible I have referred
my readers to a print translation, preferably one with notes, given the
unpredictable longevity of online resources. When no print translation was
available, I referred readers to Oxford’s Electronic Text Corpus of Sumerian
Literature (ETCSL). I cite Hittite texts according to Emmanuel Laroche’s
Catalogue des textes hittites (CTH).
The abbreviations of names of Classics journals can be found in a list in
the Oxford Classical Dictionary.

1
The Story of Myth
THE WORD “myth” is a notoriously slippery beast. Even if we stick to

scholars’ attempts to define it (setting aside the ways in which “myth” is
used informally to mean a misconception or a lie), we find ourselves afloat
upon a sea of capacious, and therefore frustratingly vague, definitions: “a
traditional story with secondary, partial reference to something of collective
importance” (Walter Burkert);1 “a narrative which is considered socially
important, and is told in such a way as to allow the entire social collective
to share in a sense of this importance” (William Bascom, as summarized by
Eric Csapo);2 “ideology in narrative form” (Bruce Lincoln);3 “a basic
cosmic framework … which indicates where cosmic power resides, what it
is called, and so how it can be used” (Mary Mills).4
All of these scholars, and many others, are correct insofar as each of
them has pinpointed some of the salient characteristics of myths as we find
them in many cultures. Yet none of these characteristics can be used in an
absolute way to distinguish myths from other kinds of narratives because
each of them is shared by stories that we typically call by other names. The
story of Sleeping Beauty, for example, is traditional in the sense that
variations of it have been handed down for about 700 years now, and it
refers to something of collective importance in the sense that all versions
have to do with the danger faced by a girl making the transition from
maiden to wife. So far, therefore, the story of Sleeping Beauty meets
Burkert’s criteria for myths. And indeed, as with many myths, the finer
points in this story can be adjusted to make the collectively important
message more specific. In some versions, the prince has sex with Sleeping
Beauty while she is asleep, and she gives birth to a child while still in that
state; it is the child, in fact, who incidentally awakens her by nursing at her
breast. This version emphasizes the subordination of the girl to her
husband’s desires and suggests that it is only in motherhood that she again

regains an independent identity. In other versions, Sleeping Beauty is
claimed by the prince only after he has awakened her with a kiss. The kiss
is a far more chivalric introduction to sexuality than the first version
offered, and begins to normalize what is in other senses a fantastic romance.
The spinning of wool—a standard task of the mature woman—serves in all
versions of the story as the danger that causes Sleeping Beauty to fall into
suspended animation in the first place, suggesting that maturation carries
risks.5
Moreover, like many myths that are periodically reworked so as to
deliver timely ideological messages, as recently as 2014 the story of
Sleeping Beauty was revamped so as to be told from the viewpoint of its
traditional antagonist, the bad fairy who caused Beauty’s problems (Robert
Stromberg’s film Maleficent). The new message, depending upon whom
you believe, was either feminist (an initially bitter victim of violence that
was similar to rape heals herself by bonding with another woman, proceeds
to wreak vengeance upon her violator, and then, newly strengthened, leads
her people into a happier existence) or socioeconomic (“any hierarchical
rise to power inherently happens through the exploitation of others”) or
something else yet again.6 Thus, the story of Sleeping Beauty would seem
to fit Lincoln’s definition of myth, as well.
But in spite of the fact that the story of Sleeping Beauty is traditional, is
of collective importance, and carries ideological messages that can be
changed from setting to setting and time to time, we do not usually call it a
myth; we tend to categorize it, rather, as a fairy tale. One reason is that
usually, the names of the characters in this story are either generic or
straightforwardly descriptive (“The Prince,” “The Sleeping Beauty,” “The
Bad Fairy”) and the place and time in which the story is set are left vague.7
What we most often call myths involve characters with specific names who
are usually members of famous dynasties that include other notable people.
For Greek myths, this includes Theseus, a descendant of the kings of both
Athens and Troezen, for example, and Medea, who is the daughter of
Aeetes, the king of Colchis, and also the niece of Circe, a goddess who
plays an important role in the story of Odysseus as well as the story of
Medea herself. Greek myths usually also take place within specific
geographic settings—Athens, Sparta, Argos, and Troy, for instance, and
somewhat more imaginatively, the Black Sea or the far western
Mediterranean. Greek myths almost always unfold within a larger

chronological scheme, as well. Jason is clearly an older hero than Odysseus
because Odysseus is told about Jason’s voyage while he is in the midst of
his own.8 Heracles, Jason’s shipmate, is clearly younger than Perseus, who
was his great-grandfather. Thus, we can build at least a partial timeline of
mythic heroes that starts with Perseus, continues onward through Heracles
and Jason, and ends in the generation of Odysseus’s children and their age-
mates. We cannot do the same for, say, Sleeping Beauty, Clever Hans, and
Snow White.
Specificity of name, place, and time, and embeddedness within a larger
network of family members and associates don’t define “myth” in any
absolute sense, either, however. If they did, the stories within John
Galsworthy’s three-volume The Forsyte Saga (which quickly became of
“collective social importance” at the time they were published in the early
twentieth century and which have stayed in print—and on the screen—ever
since) could be called myths. Is the final, defining element of myths,
perhaps, the presence of the divine, the supernatural, or at least the fantastic
(which The Forsyte Saga lacks)? Greek myths and the myths of many other
cultures focus closely on gods and other “Invisible Others”9 who have
special powers. What about the Star Wars series, then? It includes elements
of the fantastic and the divine, and it has a cast of characters with specific
names and well-woven connections to one another. Star Wars, moreover,
has been socially important since its inception more than forty years ago.
Characters such as Darth Vader are widely recognized and can be used as
symbolic signifiers in other discourses. Phrases such as “May the Force Be
with You,” “Evil Empire,” and “Luke, I Am Your Father” are part of our
popular lexicon and have been applied figuratively to a variety of other
things. The code of behavior followed by certain Star Wars characters has
even prompted the creation of a new religion with its own particular rituals,
Jediism.
Yet in spite of all of this, and in spite of the fact that George Lucas is on
record as having been inspired by the myths of many cultures when he
created the plots and characters of Star Wars,10 we’d probably hesitate to
call the stories narrated within the Star Wars series myths. Perhaps they still
seem too young, too potentially ephemeral—not yet “traditional” in
Burkert’s sense? Or perhaps they (and The Forsyte Saga, too) are
disqualified by their clearly fictional status—by the fact that we know that
they were created at a specific time by an identifiable individual: must the

characters in a myth have been understood to be real by at least some of the
audience members? Does the fact that Star Wars’ stories are set in the
furthest reaches of the cosmos make them difficult to understand as myths,
as well? What we call myths typically focus on the earth and its inhabitants,
even if gods, demons, and other supernatural entities pay visits from other
places. But speaking of the cosmos, what about Mill’s definition of myth:
“a basic cosmic framework … which indicates where cosmic power resides,
what it is called, and so how it can be used”? On the one hand, theories
developed by particle physicists describe cosmic power, its origins, and
where it resides, yet we do not call them myths, and on the other hand,
some of the stories that we do call myths have very little, if anything, to do
with cosmic power. The story of Medea’s attempted poisoning of Theseus,
for example, has nothing to do with cosmic power other than (if we stretch
this story very hard) the fact that Medea is a granddaughter of the Sun and
departs in the Sun’s chariot after her plot is revealed. The story has to do,
rather, with Medea’s determination to retain her position at the court of
Theseus’s father, King Aegeus of Athens.
We haven’t gotten very far by empirically testing scholars’ definitions of
myths against other narratives that are usually understood to belong to
categories that are distinct from myths; whatever criteria we use, the lines
between the types are blurry. Sometimes, turning to the origin of a term and
its initial connotations helps to define it—and we might particularly hope
that when we’re dealing with ancient Greek myths, at least, this technique
would work, given that our word “myth” comes from the Greek word
mythos. But mythos originally just meant something that was said; it could
denote a word or a statement or a story of any kind, even if uttered only
within one’s own mind, with no implication as to whether it was fact or
fiction, true or false, or something in between. In the poems of Homer,
which are our earliest examples of Greek literature, a conversation about
how to plan an expedition, for example, could be called a mythos, as could
the report about unfaithful serving maids that Odysseus admonishes
Eurycleia to keep to herself.11 In the fifth century BCE, and especially in
Plato, mythos becomes a trickier word. Plato defines it as a false story that
nonetheless contains some truth—a vital tool in educating children, given
that it is through such stories that the traditions of a society are handed
down. For this reason, the wise founders of a state must closely control the
myths that are narrated, particularly those by mothers and nursemaids but

also those by poets, who are notorious for telling false tales about the gods
and heroes.12 Moreover, even if a myth is true in the sense of saying
something about the gods that is correct, if it is ugly in the sense of
promoting negative morals and ethics, it must not be repeated (so much,
then, for the myth about what Zeus did to Cronus, which might promote
parricide amongst young listeners). Better that the good philosopher / ruler
make up his own myths—as Plato himself does, on occasion.13 Some fifth-
century poets begin to associate the word mythos with falsehood, too,14
although never absolutely; mythos does not slide completely to the “fiction”
side of a fiction / fact divide until the Hellenistic and Imperial rhetoricians
push it there.
It was in the eighteenth century that mythos was revamped to be used
more or less as we now use it—as what is nearly a technical term for a
particular kind of story, especially as it developed amongst the ancient
Greeks. Christian Gottlob Heyne (1729–1812), a classical philologist,
wanted to understand every aspect of the ancient texts he studied, which
meant understanding the remarkable stories that they told. He adopted into
the German language the Neo-Latin word mythus to refer to them. He
meant his new noun mythus to stand in contrast to the Latin word fabula,
which loosely meant a tale told for entertainment. Heyne felt that fabula
implied a certain element of both absurdity and falsehood that he wished to
avoid when speaking of the Greeks’ myths, which he thought contained
essential truths. For Heyne, myths had arisen from early humans’ awe of
the gods and heroes whom they worshipped and from their awe of the
natural environment; there was no frivolity about these things. Heyne’s
younger friend Johann Gottfried Herder (1744–1803) agreed with Heyne’s
views on these matters and made a contribution of his own to the
developing concept of myth that was long to haunt research on the subject.
He proposed that poetry arose in order to express myths, which meant that
myths predated poetry. In other words, a myth could be detached from the
narrative(s) that expressed it. Herder, and to some extent Heyne, was also
convinced that the essential spirit of a people (Volkgeist) could be
uncovered by studying their myths. Believing in the original unity of
humankind, Herder went on to draw what seemed to him to be a logical
conclusion: the comparative study of myths would eventually enable
scholars to reconstruct the earliest stage of the shared human spirit. Myths,

then, were beginning to take on an identity as things of deep significance—
far deeper than their surface meanings might lead one to guess.15
What all of these tangles show us is that scholars have always been sure
that myths are out there somewhere but have never been able to
differentiate completely between them and other sorts of narratives by
applying criteria such as social value, traditionality, cosmic focus,
ideological thrust, or symbolic value. Nor was I able to completely
differentiate between them on the basis of criteria that I myself introduced,
such as specificity of geography, chronology, and characters’ names, the
intertwining of characters with one another, or the inclusion of supernatural
elements. It is impossible to define “myth” in any absolute and final way
from our own perspective—and as my brief review of ancient connotations
of the word mythos showed, the Greeks as a whole had no clearly defined
category of “myth” either; definitions offered by Plato and later thinkers
who engaged with his ideas remained the purview of intellectuals. This left
Heyne’s mythus and all the linguistic variations that soon sprang up in its
wake (Mythus, mythe, myth, mito, and so on) free to be used as each scholar
chose to.
For these reasons, any definition that we assign to the word “myth” must
be understood as heuristic. That is, we should use definitions as tools to
better identify and understand the salient characteristics of particular groups
of narratives that we decide to call myths, rather than to stake out the
evanescent limits of what was an artificially created category to begin with.
However salient the characteristics that we identify within a particular
group of materials may be, it is unlikely that they will all be salient in all
the narratives from various cultures that scholars (or the narrators
themselves) choose to call “myths.”16
This doesn’t mean that myths should not be studied comparatively—
across cultures, across historical periods, and even in comparison with
narratives that we don’t consider to be myths at all. In this book, I’ll use the
myths of other cultures, as well as fairy tales such as “Sleeping Beauty,”
early twentieth-century ghost stories, contemporary fantasy fiction, TV
shows, and novels such as those that comprise The Forsyte Saga to help me
better understand the type of story that I will call a Greek myth. Such an
approach is useful because the salient characteristics of a given body of
narratives never stand out as well as they do when they are placed against
the contrasting backdrop that other narratives provide.

Greek Myths: A Working Definition
For most of my career, I haven’t concerned myself much with thinking
about the categories into which one might divide ancient sources of
information. I’ve worked towards understanding Greek religious practices
and beliefs by combining information drawn from different types of sources
without feeling the need to distinguish them from one another very
precisely. Epics, tragedies, lyric poetry, histories, anecdotes, parodies,
funerary consolations, the remarks of ancient scholars, vase paintings, and
archaeological remains, for example, have helped me put together the most
complete pictures of what I was studying as I could. Thus, my work on the
ancient Greek version of a worldwide phenomenon that I called the child-
killing demon drew together remarks made by characters in ancient
comedies, fragments of lyric poetry and tragedies, Plato’s criticisms of the
ways that women raise children, scholiast’s comments, lexicographers’
remarks, and vase paintings, amongst other things. My work on the
mysteries of Dionysus combined information from inscriptions, epinician
poems, Platonic dialogues, fragments of epics, remarks made by Herodotus,
vase paintings, reliefs on sarcophagi, and so on.17 Although I contextualized
each piece of information according to such things as the genre of the work
from which I drew it or the aims of its author, there was no need for me to
think in the abstract about how those sources might be defined in any
absolute sense. I think that in most cases, this is how research on religions
(ancient or otherwise) should be done: we should gather information from
any place that we can find it.
In this book, however, my task is different. I’ll be focusing almost
exclusively on a cluster of ancient Greek narratives that, ever since Heyne
invented the term, have usually been called myths. These narratives, which
most prominently include epics, tragedies, epinicians, and other types of
poetry, were all composed to be performed aloud; many of them were
composed to be performed to crowds of many people at festivals and on
other occasions when people gathered together. I’ll suggest that, particularly
during the Archaic and Classical periods (roughly speaking, this means
between approximately 800 and 300 BCE), hearing myths narrated and
watching myths performed in this way contributed substantially to the
Greeks’ belief that gods and heroes existed and could wield significant
power over humans.18

Before I can go any further with my plan, I need to offer my own
heuristic definition of Greek myths. I’ve already mentioned four of its
characteristics, each of which will be explored in more depth in the chapters
to come. (1) Greek myths have to do with the gods and heroes—their
exploits, their interactions, and the enduring implications that these had for
humans. (2) Greek myths draw upon a large but limited cast of specific
(rather than generic) characters, whose names are typically unique to the
single character and who are always related to at least several other
characters from the same or other myths. That is, Greek mythic characters
belong to a network—Hesiod’s Catalogue of Women being an early, and
splendidly intentional, example of this. (3) Greek myths are set in the
distant past, although the Greeks understood that past to be continuous with
the time in which they themselves lived. (4) Greek myths are set in
geographically specific places that are usually located within those parts of
the world that at least some Greeks had visited themselves. Some Greek
myths are set instead on Olympus or in Hades, and others are set in places
such as Colchis, Phaeacia, or the island of the Cyclopes that we might
describe as fantasy lands but that the Greeks themselves located within the
real world.
So far, my heuristic definition does not differ much from those of some
other recent scholars; characteristics 1 and 3 are almost always included or
assumed, and characteristic 2 was mentioned by Burkert in his 1979 study
of Greek myths.19 Characteristic 4 is seldom mentioned but is usually
implicit. The fifth and final characteristic of my definition is also often
implicit, but I’ll be emphasizing it in this book and exploring its
ramifications in more detail than earlier scholars did. It is that myths are
stories. What I mean by this is that the most influential mode by which
myths were shared amongst the Greeks was through narratives that were
meant to entertain and engage their listeners, rather than simply to convey
information. I’ve already mentioned the various genres of public
performance through which myths were conveyed, but there were other,
less public modes of conveying much of that information, too, some of
which were surely engaging in their own ways, as well.20 We can follow
Plato, for example, in assuming that mothers and nursemaids told their
children tales about the gods and heroes, some of which shared plots and
characters with the more polished stories told by the poets,21 and common
sense suggests that the Greeks, like most peoples, told stories informally on

other occasions as well. We should remember, for instance, that Odysseus
and Penelope, after making love on the first night that they had spent
together in twenty years, “took delight in telling the tales [mythoi]” of all
that had happened to them—which must have been fascinating, given the
narrative skills that each of them possessed, even if the occasion was about
as informal as ever one could be.22 We cannot recover most informally told
tales, however, because no one bothered to write them down. By and large,
therefore, when we study Greek myths, we are studying finely wrought
professional compositions, purposefully devised to enthrall as well as
inform their audiences (and of course, it is within just such a composition
that we hear about Odysseus’s and Penelope’s informal storytelling). This
point, to which I’ll return in the next section of this chapter, is central to
everything else that I do in this book.
When I choose to make Greek myths’ nature as stories one of the
defining elements of my heuristic definition, I also mean to emphasize
certain things that stories do that simpler statements cannot. Recent work in
cognitive studies suggests that creating and listening to stories—a habit that
grew out of the almost exclusively human ability to play and to imagine23—
has offered us important evolutionary advantages. More than other species,
humans constantly refine their skills by adjusting them to new
circumstances. Stories give us the opportunity to do this vicariously in
response to a wide array of conditions and in a mode that presents no actual
risk to our welfare. They lure us into doing so by engaging us: they present
characters and situations that arouse strong emotions within us and they
keep new information flowing at a brisker, more exciting pace than we
experience in normal life. And yet, because we encounter these new
characters and situations at one remove, stories allow us to evaluate those
characters and situations more objectively than we could if we encountered
them in real life, weighing their choices and considering whether we would
do the same under similar circumstances.24
Stories, moreover, can be infinitely tailored to suit particular tasks. They
are especially good at describing events that have not been experienced by
either the narrators or the listeners themselves so persuasively that those
events become credible, thus enlarging the audience’s sense of what might
be possible. Stories can coax us to look beyond the witnesses of our five
senses and imagine that another reality exists, in addition to the reality that
we experience every day. Stories—indeed, only stories—can produce the

effect of reality whether what the stories say is real or not.25 Stories, in sum,
are ideal laboratories for the refinement not only of the talents, opinions,
strategies, and judgments that we need when we interact with the people
whom we meet in our day-to-day lives but also of the talents, opinions,
strategies, and judgments that we need when we meet Invisible Others.
What counts as a story, however? For the purposes of this book, stories
are narratives that are meant to entertain and engage as well as inform us.
This means that stories are not reports about such narratives, summaries of
such narratives, or references to them. Consider, for example, what the
historian Diodorus Siculus, in describing a festival called the
Thesmophoria, says: “And it is [women’s] custom during these days to
indulge in coarse language as they associate with one another, the reason
being that by such coarseness the goddess, grieved though she was at the
rape of Kore, burst into laughter.”26 Behind the phrase “the goddess, grieved
though she was at the rape of Kore, burst into laughter” lies one of the best
known of all Greek myths: the story of how Kore (also called Persephone)
was kidnapped by Hades and how Kore’s mother, Demeter, searched for her
daughter all over the earth, grieving inconsolably until she was welcomed
into a group of human women who were able to make her laugh. These
events were narrated vividly and evocatively by many ancient authors, our
earliest extant example being the Homeric Hymn to Demeter. But neither
Diodorus’s stark reference to the story or my own, slightly fuller exposition
of what Diodorus said is meant to entertain and engage you, and neither of
us is likely to help you imagine what Demeter and her daughter endured
and felt. Thus, by my reckoning, neither of these summaries is a story and
neither of them, therefore, is a myth. Similarly, although mythographers
such as Pherecydes and Apollodorus have preserved for us the basic plots
and characters of some myths that we no longer have, mythographers have
not preserved the myths themselves; the information they provide is
typically dry in tone and scant in details. They do not tell stories; they
summarize stories and leave us yearning to hear more. Although I’ll use
ancient reports, summaries, and references to myths from Diodorus Siculus,
Pherecydes, Apollodorus, and others like them in this book to fill out our
information on the stories that remain to us, it is on the stories themselves
that I will concentrate, because my primary interest is in the ways that these
stories that we call myths are able, through their charm and their power, to

make us imagine another reality that lies beyond our daily reality, inhabited
by the gods and the heroes about whom the stories talk.
The Venues of Greek Myths

That concludes my heuristic definition of Greek myths, but I need to say
something more about the venues in which the myths were narrated.
To do this, I’ll adapt a model developed by the folklorist Dorothy Noyes,
who is interested in the varying ways in which cultural forms (under which
term she includes everything from dance to oratory to parody to crafts to
storytelling, and so on) enable the messages that they carry to resist
entropy.27
Noyes chooses to categorize cultural forms not according to any literary
genre or artistic medium to which they might belong, or the degree to which
they are socially relevant or convey ideological messages, or their proclivity
to express ideas of religious or cosmic importance, for instance, but rather
according to how people consume the forms. Her categories emphasize, in
other words, not the qualities that might be internal to a given form (which
often can be found in other forms as well, as we saw in the first part of this
chapter, when we looked at myths side by side with fairy tales, novels, and
other narratives) but rather the external circumstances through which a
form is experienced.
Noyes applies two principles of organization to her categories. The first
concerns the sort of attention that a cultural form receives. By attention she
means “the extent to which incoming information is processed.”28 She
divides attention into two types: focused and flexible—that is, the recipients
may concentrate closely on the information that the cultural form is
delivering (focused), or they may allow their minds to wander back and
forth between that and other incoming information (flexible). If we set aside
other cultural forms for the moment and think about only narratives, we can
exemplify the difference as that between consuming a story under
circumstances that allow you to engage with it completely (an example
would be watching a film in a darkened movie theater) and consuming a
story in an environment filled with distractions (an example would be
watching the same film on TV while you are at home preparing dinner).
Noyes’s second principle of organization differentiates between
experiences in which information is sought by recipients and those in which
it is unsought. When combined with her first principle, this produces four

possible ways of consuming information: (1) A sought experience to which
one applies focused attention. As an illustration, we can return to the
example of watching a film in a darkened movie theater. The viewer has
chosen to enter the theater and thereby to immerse herself in the film as
completely as possible, rather than waiting until she can stream the film at
home on TV, where her environment may distract her. (2) A sought
experience to which one applies flexible attention. As an illustration, we can
return to the example of watching the same film on TV while at home
preparing dinner. The viewer has chosen to turn on the television and
stream the particular film, but she moves between paying attention to what
is unfolding on the screen and paying attention to what must be done to
produce a well-cooked meal. (3) An unsought experience to which one
applies unfocused attention. An example would be listening to a
meandering anecdote told by a loquacious person at a party. The listener’s
attention is likely to switch back and forth between the seemingly
innumerable details of the anecdote and the barely audible (but juicy)
gossip being exchanged by the couple standing a few feet away, or the
music the host has selected. (4) An unsought experience to which one
applies focused attention. An example would be a news bulletin that
suddenly interrupts a program that one is watching on television. The
viewer did not choose to watch the bulletin, but if it is sufficiently
interesting (if, say, it describes a daring prison break that has just taken
place near the viewer’s home), then he will focus attention upon it.
Information that is important to a group of people is likely to be
conveyed redundantly, manifesting itself in a variety of forms that are
spread across Noyes’s four categories. In ancient Greece, these would
include jokes, anecdotes, figures of speech, children’s games, songs sung by
women as they weave, allusions made during political orations and funeral
speeches, dances and songs performed at festivals, works of visual art
displayed both publicly and privately, recitations of epic, theatrical
performances, epinician odes, and many other things. Such coverage keeps
the information, and the messages that it delivers, constantly on the
collective horizon; the forms converge towards a common goal.
The forms impact their recipients in different ways, however. Forms that
fall under Noyes’s second and third categories (which share the quality of
being “unfocused”) tend to fade into the “surround”—the daily environment
in which people live. This includes, for example, artistic representations of

gods and heroes on buildings that people pass by every day as they move
through the outside world; artistic representations of gods and heroes that
they see every day as they move through their own homes; the admonitory
myths that they heard as children and perhaps continue, as adults, to hear
their mothers and grandmothers telling other children (or that they repeat
themselves, if they are the mothers and grandmothers); the anecdotes about
an ancestor’s participation in a famous mythic battle that are dusted off each
year as the time for a certain festival rolls around; and so on. Repetition and
familiarity may dull recipients’ awareness of such forms, but the forms
nonetheless contribute to a background that continuously hums at a “low
frequency” with the messages that they carry.29
Other elements of the surround are more noticeable, at least at first: a
new joke about a king’s mistress, who shares her name with a mythical
man-eating monster, for example;30 or an innovative statue group erected in
the marketplace, heroizing two youths who killed a tyrant; or the Parthenon
rising upon the Acropolis to replace the temple of Athena that was
destroyed by the Persians, with new metopes and new pediments that host
new depictions of myths. Some of these things shift from one of Noyes’s
categories to another as an individual’s experience with them increases.
Initially, the Parthenon must have belonged in her first category: people
surely sought to see it and they surely gave it their focused attention the
first few times that they did. But as they continued to see it, many people’s
experiences of the Parthenon probably moved into Noyes’s second or third
category: they gave it flexible attention during the annual festivals that
brought them into its presence but they did not particularly seek to see it.
For many Athenians during most of the year, it must have been just one of
the buildings that loomed over them from atop the Acropolis as they went
about their business in the marketplace below. The art that adorned it, and
the stories that it told, initially stunned each viewer, but eventually faded
into the background hum.
The products of other cultural forms, however, particularly those that
belong in Noyes’s first category, never fade into the hum. We might call
these products “punctual” in the sense that they occur only at certain times
—for example, dramatic performances presented onstage or first-run
movies showing in Imax theaters, which audience members must seek out if
they wish to experience them because they won’t be available at other times
or in other places. We might also call them “punctual” in a sense that

reflects the word’s etymology: these products emphatically stand out from
the surrounding cultural landscape as singular “points” of interest that
engross our focused attention. Not surprisingly, those who create them
typically put a good deal of effort into those acts of creation and they
thereby wield a good deal of influence over their audiences. As Noyes says,
[These forms are] consciously evaluated for their beauty and skillful
execution. They are produced to be so evaluated.… All of this
conduces to the making of concentrated “affecting presences,” objects
with maximal internal coherence, maximal salience in relation to their
surround, and maximal indexical connections with their audience,
cultivated over generations of mutual attention. Explicitly marked as
art, these forms compel us both socially and sensorily, both
consciously and unconsciously.31
In ancient Greece, the cultural forms that most pronouncedly belonged in

this category were the polished public performances of professionally
composed poetry, underwritten by men who could afford to guarantee that
skilled poets would be hired and that the events themselves would be well
produced, with all the sophisticated techniques that were then available.
Tragedies, recitations of epic, poems presented at victory celebrations, and
hymns and dithyrambs performed in honor of the gods all belong in this
group, for example. Some of these works were subsequently re-presented in
smaller, less formal venues, where attention might have been more flexible
and the decision to perform the pieces more spontaneous (such as during
symposia, parties that brought together groups of men who sometimes
spontaneously recited poems that had originally been presented formally, or
portions of those poems), but the aura surrounding any work that had once
merited a specially marked public performance and then subsequently
earned memorization and re-presentation guaranteed that the work would
carry weight even under less focused circumstances.
The genres of public performance that I’ve just characterized were
amongst the most prominent cultural forms through which myths were
presented in ancient Greece, and myths constituted the primary subject
matter for such genres (of the more than 200 tragedies that we know about,
for instance, only a handful focused on recent events in human history, in

contrast to the affairs of the gods and heroes).32 Thus, in ancient Greece,
myths were frequently, publicly and elaborately conveyed in a manner that
guaranteed that they would become strong “affecting presences,” to borrow
Noyes’s phrase. They constituted a marked discourse, and the messages that
they conveyed, therefore, were highly likely to influence their audiences’
opinions and beliefs. (I will talk more about some of the characteristics of
mythic narration that contributed to these effects in Chapters 3–5.)
As we go through this book, however, we shouldn’t forget that the
environment in which these forms operated was itself replete with myths
that were less formally narrated, through both words and visual images.
This meant that by the time an individual encountered a myth in a polished
narrative form—as a tragedy or in an epinician ode or a hymn, for example
—he or she had probably already heard its story, or seen representations of
it, or at least heard stories that were similar to it or that involved some of
the same characters (for instance, the individual may not have previously
heard the story of how Odysseus, disguised as a beggar, tested Penelope’s
resolve, but perhaps he or she had heard a story about Odysseus and the
Cyclops or Odysseus and the Trojan Horse, and thus was familiar with
Odysseus as a character who exhibited resourcefulness). Such repetition
helped to inculcate at least the simpler messages of a story (to stick with
Odysseus as an example, “if you want to be a hero, be clever as well as
strong”), but the backdrop of familiarity that repetition provided allowed
the poets to do more complicated things as well, which made myths even
more effective communicators. Poets could present a well-known story with
breathtakingly new dialogue and details, for instance. What woman present
at a recitation of what we now call Odyssey 19 wouldn’t have thrilled to
hear Penelope tease the beggar with the declaration that there would be an
archery contest, and that she herself would be the prize? How many male
hearts in Aeschylus’s audience curdled when they heard Clytemnestra
swear that she knew no more about adultery than she knew about how to
temper bronze?33 Or, poets could confound expectations by altering a well-
known myth—how could Pindar’s striking revision of the story of Tantalus
and Pelops have affected his audience, if they hadn’t already known
another, very different myth about the pair? Or yet again, poets could place
a well-known figure in an unexpected role: Pindar’s Chiron as an adviser to
lovelorn Apollo must have come as a bit of a surprise.34 Or, finally, poets
could invent wholly new episodes or details and drop them into well-known

myths: Bacchylides tells us that, in the middle of Theseus’s voyage to Crete
to defeat the Minotaur, he had to leap into the sea to retrieve a ring that
Minos had tossed overboard.35 In these and many other situations, new or
newly revised myths manage to preserve some of what I’m arguing were
the defining traits of Greek myths (for example, these stories involve well-
known characters who are firmly woven into a network of other characters)
and yet they also engage listeners with their novelty. By focusing on such
myths in this book, I’ll be looking mostly at material that belongs in
Noyes’s first category, but we must not forget that the poets were able to do
what they did in part because the environment in which they existed was
already full of the gods and heroes.
A Few Preparatory Remarks

Two planks of my methodology can be inferred from what I’ve already
said: I engage in comparativism both across cultures and across genres of
narrative, and I think that close readings of ancient myths, with special
attention to the fact that they were works of literary sophistication, will
enable us to understand how they helped to create and sustain belief in the
gods and heroes—whether their composers purposefully aimed at that
specific goal or not.
But there are a few more preparatory remarks that I should make. One
thing that motivated me to write this book was noticing, over the years, the
powerful effect that narratives had on my friends, my relatives, my students,
and myself. The novels and comic books that we read and the television
shows and movies that we watched changed our political and ethical
outlooks, our ambitions, our expectations of what our family and
professional lives should be, our religious and spiritual beliefs, and,
cumulatively, our worldviews. The years during which I finished the book,
moreover, were a time when the power of narratives to make major
differences in the world—as well as in the individual—became increasingly
acute to anyone observing the American and European political scenes. I
became interested in how scholars of contemporary narrative tried to
understand at least some of what I was observing and I eventually came to
conclude that some of their approaches and methods could be applied to my
ancient materials. For, although it is true that a particular culture determines
the specific meanings that a narrative carries—and any comparativist must
therefore take care not to export meanings across cultural and temporal

borders irresponsibly—it is nonetheless also the case that, because the basic
structures and mechanisms of human cognition have not changed much
over the millennia, at least some of the techniques by which meanings are
conveyed through narratives—the techniques that work to engage our
attention and win our trust—tend to endure.36 I will move forward in this
book on that assumption and will use the observations and methods
developed by scholars who have studied modern and contemporary
narratives when they are helpful.
At the start of this chapter, I said that “myth” is a notoriously slippery
beast, but “belief,” which I will use a lot in the pages that follow, is at least
as slippery. I define “belief in something” as the conviction that the thing
really exists outside of the imagination, but behind that deceptively simple
phrase lurks another problem: How do we know whether people really
believe in something? What kind of proof can we adduce for a phenomenon
that goes on inside their heads and hearts? The problem becomes
particularly fraught when we study ancient beliefs because of the
differences between pre- and post-Enlightenment assumptions about the
universe, its inhabitants and how we might gain knowledge of them. Even a
vigilant scholar has to work hard to avoid slipping into the assumption that
ancient Greeks of average intelligence couldn’t “really” have believed that
Heracles slew a nine-headed water snake or that Zeus might see—and
punish—bad behavior towards a guest. The folklorist David Hufford has
astutely observed that there are modern “traditions of disbelief,” that is,
dismissive ascriptions of any belief that cannot be proven by modern
scientific methods to various kinds of errors on the part of the believers—
hallucinations, illusions, or gullibility, for example.37 In other words, we
tend to reject not only the possibility that the anomalous phenomena that
others believe in could actually exist but also the idea that healthy,
reasonable people would really believe in them in the first place. We are
schooled to assume that such beliefs are either marks of insanity, transient
reactions to psychotropic pharmaceuticals, tainted food, extreme physical
stress, or third-party hoaxes.
In the end, I think, the best we can do to determine what people “really
believe” is simply to accept what they say about the topic themselves. There
are relatively few formal statements about ancient Greek belief in the gods
and heroes, outside of philosophic discussions, but we do have numerous
testimonies of entreaties to the gods and heroes for their help or thanks for

their having given it, such as those inscribed and displayed at the great
healing sanctuary of Asclepius at Epidaurus; the small representations of
body parts left as votives at Epidaurus and other healing shrines; the many
inscriptions that credit gods and heroes for success in conceiving and
bearing a child, for support in battles, for turning away plagues, and for
other things; the small gold tablets found in graves, expressing the
expectation that the gods of the afterlife will aid the individuals who were
buried with the tablets; and sometimes statements that simply ask for help
or express thanks without any indication of what the specific situations
involved. We also have copious evidence for the Greeks’ participation in a
huge number of rituals that were directed towards honoring the gods and
heroes and motivating them to help humans (prayers, sacrifices, libations,
votive offerings, and so on). Whatever else scholars have argued that these
rituals accomplished (social cohesion or ideological formation or reduction
of anxiety or displacement of authority onto abstractions, and so on), we
must assume that a majority of those who participated in them were doing
so because they believed that the recipients of the rituals existed and were
paying attention. This book attempts to understand one of the factors that
created and sustained such beliefs, namely, the public performance of
myths.
Having raised the topic of belief, I should pause on Claude Calame’s
recent discussion of feintise ludique, which might be translated as “playful
pretense” or “making believe.”38 Calame derives the concept from the
philosopher Jean-Marie Schaeffer, who developed it in connection with a
broad range of fictional forms—video games, traditional children’s games,
opera, and films, as well as the forms of fiction that we typically think of
when we use the word: novels, for instance.39 Schaeffer stresses that the
boundary between the real world and fiction, even in adulthood, is far more
porous than has previously been recognized. The habit of playful pretense
endures far beyond childhood games of playing house or cops and robbers,
becoming subtler and more sophisticated as the individual grows up and
engages with role-playing games such as Dungeons and Dragons and with
literary creations that range from openly fictional novels to works that
purport to present truths, such as The Protocols of the Elders of Zion.
Schaeffer stresses how easy it is for the creators of fiction (much like
Hesiod’s Muses) to mislead consumers with lies that are so well narrated
that they seem like truths. Under Schaeffer’s model, then, any person’s span

of experiences with fiction is likely to include both situations in which that
person participates in a “playful pretense” that a given book, game, or
online experience is true, knowing full well that it has actually been created
out of whole cloth by its author(s); and situations in which the person is
convinced that a work that has been constructed out of whole cloth is
actually truthful—situations in which, we would say, the person has been
deceived by what the author(s) created. We could exemplify the latter
situation by someone who is convinced that The Protocols of the Elders of
Zion is genuinely ancient and must be acted upon, and we could exemplify
playful pretense by someone engaging with the online role-playing game
“Hogwarts Extreme: The Interactive Harry Potter Experience,” well
knowing that Harry Potter and his friends were invented by J. K. Rowling
and that the game itself was developed by a group of individuals
represented by a company called HEXRPG, LLC.40
Yet there is a third way of engaging with works that look like fiction (at
least to us), which I take up in depth in Chapters 3–5 of this book. A work
that presents situations unlike those experienced in everyday life might be
understood by some or all of its audience members as truthful without its
author having intended any deceit. Perhaps even the author him- or herself
understands it to be truthful. After all, the Greeks thought that the Muses
might whisper stories to a poet and then leave it up to the poet to shape
those whispers into entrancing poetry.41 A myth might be understood to be
true, in other words, even if it were acknowledged that the way that the
characters in the myth were represented and the details of the plot had been
devised by a particular human at a particular time for a particular purpose.
Audiences might engage with the myth, therefore, neither in a spirit of
“playful pretense” nor in a state of delusion, but rather in a state of sincere
belief.
Calame remarks upon this possibility as well; he is especially interested,
here as well as in some of his earlier work, in the fact that when a
performed myth was received by its audience as true, its pragmatic effect—
its ability to persuade people to apply the ideas that they encountered in the
myth to situations that they encountered in their lives—was strengthened.
But in Calame’s formulation, the pragmatic effect of a performed myth
concentrates on the specific, and typically immediate, aftermath of that
performance. The royal audience of Pindar’s fourth Pythian ode, for
instance, might be moved to pardon the exiled man who had hired Pindar to

compose the ode. I agree with Calame on this point, but my interest in this
book goes further. I will be arguing that engagingly performed myths that
were regarded as true cumulatively helped to create and sustain belief in the
gods and heroes more generally.
This is not to say that every person left every performance of a myth with
newly confirmed beliefs. I take it as a given that each performance of a
myth affected different people in different ways. Some people remained
unchanged; the narrative entertained them, and perhaps even pragmatically
moved them to change their opinions or behavior, but it did not affect their
level of belief in the gods and heroes. Other people present at the
performance became newly confirmed in what they already believed. And
some people were so affected by the performance that they came to believe
—or at least started to believe—things that they hadn’t believed before.
This breadth of effect is another quality that characterizes stories: stories,
including myths, offer a latitude of ambiguity that opens doors but does not
compel people to walk through them.
The Chapters of This Book

So far, I have offered a general description of what I intend to do and how I
intend to do it; a sketch of the remaining chapters may help to orient readers
further.
Chapter 2, “Ritual’s Handmaid,” looks at the primary ways in which
Greek myths have been studied and lays out some of the standing
assumptions that I challenge in the rest of the book. Most importantly, this
chapter tells the story of how scholars followed Herder’s lead in
essentializing Greek myths—that is, studying them only after they had
severed what they considered to be the “real myths” from the poetic
compositions that they viewed as merely the myths’ narrative vehicles. In
telling this story, the chapter also explores the far-reaching, and deleterious,
effects that essentialization has had on our understanding of Greek myths.
In particular, I focus in this chapter on the ritualist approach, which was
pioneered by the Semiticist W. Robertson Smith in 1888 but made famous
by the slightly younger Cambridge classicists James Frazer and Jane Ellen
Harrison—and then was energetically revived, decades later, by Walter
Burkert and his students. At the heart of the ritualist approach lay the
assumption that ancient myths had emerged in order to support rituals. A
myth may have done so by verbally articulating a ritual’s goals, by

justifying the ritual’s existence through telling the story of its creation, or (if
the myth is understood as a performative speech act) by helping, literally, to
bring the ritual to fruition. Particularly central to the ritualist approach have
been aitia (myths that tell about the origin of a ritual or some other aspect
of a cult) because these have often been interpreted to perform at least two
of these functions—or even all three—at once. I show that the assumption
that aitia were regularly narrated in connection with Greek rituals is wrong.
Although myths in general were very frequently narrated in connection with
rituals, aitia themselves seldom were.
I also note that the essentialization of Greek myths was further
encouraged by two other approaches to the study of myths that developed
later in the twentieth century: the psychological and the structuralist. I end
the chapter by emphasizing that because, for the Greeks, the “vehicle” was
the myth, the only way in which we will fully understand what Greek myths
did and how they did it is to look more closely at the effect that these
polished narratives had on their audiences.
Chapter 3, “Narrating Myths,” begins by showing that, in contrast to
myths found in many other ancient Mediterranean cultures that served as
performative utterances (that is, as recitations that ensured that rituals they
accompanied would succeed), Greek myths found their power in
metaphorical, figurative language, which the Greek poets used to subtly
bring the world in which the myths took place closer to the world in which
the myths were being performed.
Building upon this idea, I show in more detail how Greek myths,
precisely in the poetic versions in which they have come down to us, could
help to create and sustain beliefs concerning the gods, the heroes, and the
ways in which the divine world operated. I start by looking at two recent
explorations of how certain kinds of fiction, when consumed under
particular conditions, can alter people’s beliefs. The first is Joshua Landy’s
study of “formative” fiction—fiction that inculcates its audience not with
facts but with new modes of thinking and experiencing the world.
Formative fiction, Landy demonstrates, has the potential to change audience
members’ understanding of the nature of reality and their place within it.
His case studies range from Plato’s dialogues and the Gospel of Mark to
Jean Eugène Robert-Houdin’s nineteenth-century stage magic and Samuel
Beckett’s novels.42 The second exploration is Tanya Luhrmann’s work on an
Evangelical church called the Vineyard. Vineyard congregants condition

themselves to feel the presence of God through techniques that hone their
ability to perceive what cannot be perceived by the corporeal senses alone
—techniques that include the reading of fiction.43 Although Luhrmann’s
work is firmly anchored in twenty-first-century America, her insights as to
how fiction affects human emotions and cognition are broadly applicable to
experiences across time and place.
I finish this part of Chapter 3 by looking at another phenomenon that
attests to the power of well-wrought fiction: the parasocial relationship
(PSR). PSRs are established when an individual forms an attachment to
another person who is not even aware that the individual exists (a fan
sending letters to a movie star, which only the star’s staff members ever
read, is an example). Recently, psychologists have shown that virtually
everyone forms at least a few PSRs in the course of life—and also that
PSRs with vividly drawn fictional characters not only are common
(Sherlock Holmes is an enduring case) but also elicit the same cognitive
and emotional responses as do PSRs with real people.44 Psychologists
suggest that the only significant experiential difference between PSRs
directed towards real people and those directed towards fictional characters
is that an individual who has formed a PSR with a real person is told by his
or her society that the person really exists (even if he or she is out of reach)
but that an individual who has formed a PSR with a fictional person is told
by his or her society that the person does not exist. I draw on this idea to
propose not only that the vividly drawn characters of Greek myths elicited
the same cognitive and emotional responses as engaging fictional characters
do nowadays but also that, in a society that promoted the idea that those
characters did exist and could affect the lives of humans, what we might
view as PSRs (imaginary unilateral relationships) would be experienced by
worshippers as vibrant, fully reciprocal social relationships with gods and
heroes to whom they paid cult.
The effective narration of well-composed fiction, in short, disposes its
audience members to embrace entities and even whole worlds for which
everyday logic and experience provide no support. In the remaining
sections of this chapter, I look at some of the specific features of Greek
mythic narration that made it particularly good at doing this. First, I review
some recent work showing that narrating a story episodically (that is,
narrating only a portion of a larger story at one time, as was the practice
with Greek myths) encourages audience members to think about and talk

about the story’s characters in between episodes, which gives the characters
additional depth and reality and strengthens the audience members’
attachment to them. Then, I review the work of folklorists who have studied
the ways in which ordinary people narrate supernatural experiences in
contemporary Western communities. These scholars have isolated the types
of speech and speech patterns that best convince listeners to believe such
stories. I note that these real-life techniques have been mimicked by authors
of fictional ghost stories since the nineteenth century and then look at the
ways in which the techniques are used in Greek myths about the gods and
heroes, as well. I suggest that, consciously or not, Greek poets adapted
techniques of narration that they had observed to be effective in convincing
listeners that a remarkable story was true.
Chapter 4, “The Greek Mythic Story World,” shows how the nature of
Greek myth-ology (i.e., a structure that was above and beyond individual
Greek myths) meant that each contribution that the narration of a myth
made to belief in a given hero or god simultaneously contributed to belief in
a larger divine world more generally. In particular, I suggest that an
essential element that enabled this to happen was the story world that was
cumulatively being created, on a continual basis, by the myths themselves.
Because, for example, a skillfully narrated myth about Heracles was
embedded within this story world, it had the power to sustain and enhance
belief not only in Heracles himself but also in the entire cadre of gods and
heroes to which the story world linked him.
I begin with a discussion of what makes story worlds coherent and
credible in general, with particular attention to recent theoretical work by
scholars who have studied fantasy worlds such as Middle-Earth and Oz—
worlds that are often assumed to be similar to those of myths.45 I go on to
show that the story world constructed by Greek myths possesses relatively
few of the characteristics that the story worlds of fantasy do. Rather, I
suggest, the credibility and coherence of the Greek mythic story world was
anchored in the fact that the myths drew on a pool of characters who were
each closely entwined with the others, creating a dense network of
relationships.
I then discuss some of the other benefits that accrue from embedding
myths in such a story world, drawing on novels and television shows for my
comparative material. For example, the subtlest of allusions, when made
within such a system, potentially evokes the history of not only a single

character but also an entire dynasty or a complex of companions who are
associated with that character. And, a well-known character may be
“crossed over” into a new myth that a poet is creating and used to endorse
the new myth by his or her mere presence. Thus, for instance, the fact that
Persephone became Dionysus’s mother in the myth that poets developed to
anchor the Bacchic mysteries in the late Archaic Age secured the new myth
within longer-standing myths about Persephone and her mother, and
thereby lent weight and credibility to the new mystery cult that the myth of
Dionysus and Persephone underpinned.
The final part of this chapter takes up the question of whether the story
world of Greek myths is truly a “Secondary” world (to use J. R. R.
Tolkien’s influential term).46 That is, was it significantly different from the
“Primary” (that is, the everyday) world in which the Greeks really existed?
I conclude that it was not, and suggest that this situation served Greek
religion well, given that the gods and heroes of the myths were presumed to
still exist and intervene in the lives of mortals at the time that the myths
were being narrated. The world of the myths was understood to be distant in
time from the real world but not overwhelmingly distant in nature.
Chapter 5, “Characters,” falls into four parts. The first draws on the work
of analytical philosophers and narratologists to explore the question of what
makes a fictional character “real”—and, in some cases, develop a life that is
independent of the work(s) in which the character appears.
I begin by proposing that there are four types of characters that are
familiar to twenty-first-century people: (1) The fictional character who is
the work of only one author or artist. (2) The fictional character who has
been presented by more than one—perhaps many—authors and artists.
These characters develop composite identities and in some cases take on
existences that transcend the works that described them (Sherlock Holmes
is a well-known example). (3) Real people who are familiar to us through
news reports or history books but who have also been represented in
fictionalizing works (Queen Elizabeth II is an example). As with the second
type of character, these characters develop a composite identity. (4) Entities
who are the subject of belief and who are described by both canonical and
noncanonical narratives (for example, the Christian God as he is known in
the modern Western world). The composite identities of these entities is
constructed along the same lines as those of the third type.

I then propose that the gods and heroes of Greek myth were different
from all four of these familiar types. The gods and heroes were not strictly
bound by any canonical narratives. Therefore, there was no “Real Hermes”
who could be clearly distinguished from “fictionalizing” portrayals of
Hermes; different portraits of Hermes simply sat alongside each other, none
of them claiming to be true in an absolute sense. For centuries, therefore,
Greek authors, artists, and their audiences implicitly participated in an
ongoing conversation about exactly who the gods were and what they had
done during the early epochs of the cosmos’ existence. The material that fed
this conversation came almost completely from what we in the twenty-first
century understand to be fictional sources—sources that the Greeks
themselves understood to be, if not fully fictional in the sense of having no
basis in fact, then at least fictional in the sense of being ficta, that is, as
having been presented in forms that had been generated by human
individuals who adapted their material to suit the occasions on which the
stories were performed, their own ideas, or their patrons’ desires.
Characters such as these—characters for whom there is no clear original
and whose existences are anchored instead within the drifting overlaps of
traits shared by different portrayals—can be adapted in bolder ways than
can other types of characters. Indeed, the disagreement amongst poets about
exactly who they were and what they had done gave them an air of
verisimilitude like that of famous “real” people.
In the second part of the chapter, I look at two related concepts,
plurimediality and accretive characters. When characters are presented
through more than one narrative and perhaps more than one medium, each
narrative offers a different instantiation of that character—sometimes a
strikingly different instantiation. Such characters are plurimedial. A single
person’s experience of a plurimedial character is accretive insofar as when
he or she encounters the character in different instantiations, his or her
concept of the character gradually accrues traits from some or all of those
instantiations. This process requires an investment of cognitive and
emotional energy that encourages each person to form an unusually close
bond with the character.
I go on to look at how plurimediality and accretive characters worked in
antiquity, focusing on Theseus as an example. I suggest that the many
different types of works—verbal, visual, and cultic—through which a god
or hero was presented offered rich resources from which an accretive

character could be built over time. I also suggest that in a society where
people were told that these characters really existed, the close emotional
and cognitive relationship that an individual might have forged with
Theseus (for instance) would be even stronger. I also look at the way in
which richly plurimedial characters elude the confines of any particular
narrative that portrays them and come to live in the “margins” between
works.
The third part of the chapter looks at the ways in which gods are
presented in Greek myths, drawing on the work of the anthropologist Pascal
Boyer and the psychologist Justin Barrett, who note that Invisible Others
are described in different ways on different occasions. In some cases, the
“theologically correct” view of an Invisible Other is emphasized: a god, for
example, may be said to be omniscient, omnipotent, and omnipresent. In
other cases, the same god is portrayed as being closer to humans in nature:
the god is prone to being deceived by other gods or even by humans, for
example.
A “human” god is easier to believe in, but a “theologically correct” god
can better be used to explain certain aspects of the way the world works and
to elicit thought about how, if at all, humans can hope to improve the
circumstances of their existence. I look at the ways that the Greek gods
were portrayed in myths, demonstrating that each of these modes was
adopted by certain types of narratives: epics and the Homeric Hymns tended
towards “human” gods, and tragedy and epinician tended towards
theological correctness. I suggest that together, these modes provided gods
who were usefully flexible; either the “human” or the “theologically
correct” god could be pondered as a situation demanded.
Chapter 5 finishes with a look at names. The many different
instantiations of plurimedial, accretive characters are held together by the
stability of a name—however each of us might conceive of the king of the
gods, he is always called “Zeus.” Correspondingly, this means that behind
each name, each time it is used, there shimmers a rich reserve of
significance and associations with other characters within the mythic
network that I examine in Chapter 4. Poets could play with these
significances and associations simply by mentioning a name. “Medea” was
likely to evoke infanticide, however subtly, even when that part of her story
was not being narrated. Evocative names are one of the salient features that
set myths apart from other types of stories, in fact. Novels, for instance,

must spend a great deal of time developing their characters because those
characters are new to the readers; their names mean little or nothing when
the novel begins. Fairy tales are populated by characters whose names are
typically generic and who carry no individualized personalities or histories.
The closest analogy for what myth accomplishes through its specifically
named characters probably comes from contemporary fan fiction, whose
characters already possess well-developed personalities and histories that
they bring forward into the new narratives that fans write and share with
one another.
An important backdrop to my project as a whole, as I present it in
Chapters 3–5, is the fact that the places and times in which Greek myths
were publicly performed were in themselves conducive to creating and
sustaining belief—that is, the myths were frequently performed in
sanctuaries dedicated to the gods and heroes, during festivals dedicated to
the gods and heroes. The audiences were primed by these conditions to
open their minds to the ideas that the myths conveyed. Festivals and myths,
thus, mutually supported each other.
Chapters 6 and 7 focus on the two subjects that I suggest are most
distinctively characteristic of Greek myths and of their particular ways of
engaging audiences: the frequency with which myths tell of metamorphosis
(a person being changed into an animal, plant, or mineral) and the
prominence of heroes (humans with extraordinary traits or abilities). Each
of these chapters is underpinned by the methodology and conclusions of
Chapters 3–5 (that is, they look for the particular ways in which the
narrative vividness inherent to each type of myth enabled it to help create
and sustain belief in certain aspects of the divine world and its inhabitants),
but in order to do this, each delves into its topic in other ways as well.
These two chapters also share an increased emphasis on comparison.
Because I want to argue that these two types of myths are characteristically
Greek, I discuss how they do, and do not, share certain aspects with myths
taken from other cultures and sometimes with stories from other cultures
that are typically called legends or fairy tales. Comparison brings the
qualities of the Greek myths into sharper focus and thus enables us better to
understand why Greek myths are different from, or similar to, other myths.
Chapter 6, “Metamorphoses,” begins by noting just how remarkable the
Greek penchant for stories about metamorphosis is; other cultures fall far
behind in this respect. (I distinguish between metamorphosis, in which an

outside power permanently changes a human into something else, and
shape-shifting, in which a human has the ability to turn him- or herself into
something else temporarily. Stories of shape-shifting are fairly common in
other cultures but very rare in Greece.)
In the next section, I look at hybrids (e.g., the Chimaera, Pegasus, the
Hydra), a category of creature that is often discussed alongside myths of
metamorphosis but which, I argue, should be kept distinct. I show that
mythmakers put considerable effort into anchoring hybrids within the
network of Greek mythic characters that is discussed in Chapter 4. The
hybrids have mothers and fathers, children, and friends amongst the gods
and heroes; they are, in that sense, a normal part of the divine world. Story
details flesh out those relationships: Callirhoe and Pasiphae suckle their
monstrous babies (Geryon and the Minotaur) and Circe tells Odysseus that
Crataeis, the mother of Scylla, will be able to restrain her daughter from
making a second attack on Odysseus’s ship, if Odysseus asks Crataeis to do
so. For hybrids, family relations determine the experiences of a character
just as much as they do for anyone else in the divine world, in other words.
The chapter then shifts to its main focus, metamorphosis. I argue that the
long history of scholarly attempts to find a system of specific meanings
hidden behind metamorphic myths has been misguided. Although a myth
about a woman turning into a dog might secondarily say something about
the male view of women, for example, or a myth about a cannibalistic king
turning into a wolf might secondarily say something about human dietary
prohibitions, no single interpretative key (or even set of keys) has ever been
found that convincingly decodes metamorphic myths as a whole. The
individual stories themselves changed quite a bit in antiquity, moreover; the
details behind Hecabe’s transformation into a dog are staggeringly varied,
for instance.
After having presented case studies that test several modes of
interpretation, I suggest two things. The first is that the most successful
approach to understanding metamorphic myths will combine aspects of
work pioneered by Maurizio Bettini47 that centers on affordances
(characteristics of a person, animal, or object that are open to multiple
interpretations) and aspects of work by John Scheid and Jesper Svenbro48
that emphasizes the fluidity of myths, each of which is a concatenation of
various ideas that uniquely suits the circumstances of its narration. I use a
myth about Arachne as a case study to show this. The second is that the

primary characteristic of metamorphic myths, collectively, is their ability
induce wonder—to grab and hold attention.
In this sense, the metamorphic myths fall into line alongside sightings of
the Dioscuri atop the masts of ships, reports of the gods appearing on
battlefields to shore up their favorite side, and Asclepius appearing to cure
patients in their dreams—all of which are alleged to have happened in the
“real life” of historical Greece. The metamorphic myths included—indeed,
they fervently embraced—the negative side of the gods’ power and how it
affected humans, however. Metamorphic myths, then, were a place where
the Greeks could think about an aspect of the gods’ nature, and its potential
to explode at any moment into mortal life that they did not care to confront
in other venues, such as public inscriptions and prayers.
Chapter 7, “Heroes,” begins by emphasizing the enormous popularity of
myths about heroes in ancient Greece. A survey of Archaic and Classical
literature demonstrates that, although gods are found to some degree in
almost every kind of mythic narrative, heroes are far more often the main
characters, around whom the narratives revolve. For comparison, I look at
myths from ancient Near Eastern cultures. Although scholars agree that
Greek cosmogonic and theogonic myths were strongly influenced by
ancient Near Eastern myths, the situation is different when it comes to
myths about heroes. In ancient Near Eastern myths, gods most often took
center stage. The hero, then, as the Greeks developed him, was a distinctive
feature of their myths and cults.
The chapter goes on to look at the hero as a narrative character,
developing some of the ideas on episodic narration that were presented in
Chapter 3. There are two distinct types of episodic narration: serials, in
which narratives told across episodes have an overall arc; and series, in
which there is no overall narrative arc. In serials, characters undergo
changes and develop; in series, they change very little, remaining stable in
episode after episode, existing eternally in a sort of “golden age.” Whereas
gods in Greek myths are series-type characters, heroes partake of both sorts
of episodic narration. Because every hero is born and dies, there is
necessarily a narrative arc to his overall story (which audience members
knew very well and which poets sometimes emphasized), but because poets
could almost infinitely invent new adventures for the heroes, or insert new
side episodes into existing adventures, heroes also went on forever,

eternally unchanged. This dual mode of presenting heroes kept them in a
balanced tension between the human and divine.
I next discuss the hero as monster killer. I begin again from ancient Near
Eastern myths, in which the norm is for gods to kill monsters and in which
monster killing usually has cosmogonic associations. In contrast, Greek
gods seldom kill monsters. Instead, there are many Greek myths about
heroes killing monsters; this is, indeed, one of the defining characteristics
of the Greek hero. A close reading of some passages from Hesiod and the
Homeric Hymn to Apollo suggests that originally, Greek gods, like their
Near Eastern counterparts, did kill monsters but that, as the Greeks
developed the figure of the hero during the Archaic Age, the heroes
gradually took over this task. The heroes—figures poised halfway between
gods and humans—thus assumed a significant part of the responsibility for
creating a world in which normal humans could live and thrive.
The chapter finishes with a look at the end of heroes. Once again
developing a new twist on some Near Eastern stories, the epic poets sang of
a moment in cosmic history when the gods had come to feel that the
flourishing human race was a burden. Zeus decided that war was the best
means of trimming the population and instigated first the Theban War and
then the Trojan War. Many heroes died on these battlefields or while
returning home from them.
These wars gave the last of the heroes ample opportunities to display
their skills in combat, but after the carnage had cleared, the humans who
were left behind could only pray for their heroic ancestors’ help from
beyond the grave and reiterate what had made those ancestors great by
telling myths about their exploits. Those stories, and the stories about the
gods that were intertwined with the heroes’ stories, are what the pages that
follow will discuss.

2
Ritual’s Handmaid
AT THE TURN of the last century—a time when classicists still ruled the
academic roost and every educated man and woman knew their Greek and
Roman myths—there arose an idea that circled out into the wider world,
changing the way that the myths and literatures of many cultures were
understood, and then circled back again into classics like an aboriginal
boomerang. I want to begin the work that I’ll do in the rest of this book with
a history of that idea—for, until we examine its ancestry and the effects that
it has had, we won’t understand how the study of Greek myths ended up
where it did and why we now need to take a different path.
By “idea,” I mean what is commonly known as the “ritualist approach”—
that is, the idea that myths are somehow connected to rituals. In its strongest
version (the one offered by its founding father), this meant that myths were
created in order to explain or justify the existence of rituals. As time went
on, some scholars argued the opposite—that rituals enacted myths—or
conceded that myths and rituals might arise hand in hand, yet even in these
versions of the theory, myths were implicitly the weaker partners: rituals
were what people did and myths were only what people said.1 Whichever
version of the theory one embraced, myths had to be adjusted in some way
before they neatly matched up with the rituals in question. Typically, this
meant stripping away or altering what was taken to be the extraneous
narrative material through which myths were conveyed—the characters’
names or details of the plot, for example—in order to get at what was
understood to be a myth’s true core and its real, essential meaning. Having
been thus extracted from any real, lived experience, myths ended up
seeming more like equations to be solved by clever scholars than tales that
had once engaged and entertained ordinary people—a situation that set the
study of myth on a misguided track for more than a hundred years.

The King That Wouldn’t Die
In a series of lectures delivered at the University of Aberdeen between 1888
and 1891, William Robertson Smith, a biblical critic and Semiticist,2
proposed that whereas the heart of any modern religion lay in its system of
beliefs, “primitive” religions focused instead on rituals.3 Rather than beliefs,
these religions had myths, and the myths were intended to explain not the
nature of the universe and all it contained (as E. B. Tylor had proposed two
decades earlier in his magnum opus, Primitive Religion) but the existence
of the rituals themselves, whose real origins had been long forgotten. In
other words, as props for religious systems, myths were doubly derivative
and doubly defective: they existed only because rituals did, and those who
performed the rituals could not even remember why they did so in the first
place. “Primitive,” a word that Smith used more or less interchangeably
with “ancient,” included the religions of Greece, where
certain things were done at a temple, and people were agreed that it
would be impious not to do them. But if you had asked why they were
done, you would probably have had several mutually exclusive
explanations from different persons, and no one would have thought it
a matter of the least religious importance which of these you chose to
adopt.4
Smith’s book drew considerable attention within the scholarly

community, but it was his younger (and much longer-lived) friend James
Frazer who carried the link between myth and ritual out into the wider
world with The Golden Bough, a work that, in its third edition, filled twelve
volumes with myths and rituals that Frazer had patiently gathered from
“primitive” cultures throughout the world and across historical periods.5 He
used this material to promote two ideas that interest us here. The first was
that cultures pass through three evolutionary stages. During the earliest and
most primitive stage, people trust in magic and therefore focus their
energies on the performance of rituals. During the second, people trust in
religion and focus upon dogma or creed, which is expressed through myths.
During the last and most highly developed stage, into which the
industrialized Western world had already moved, according to Frazer,
people trust in science and need neither myth nor ritual. Cultures that are

stalled between the first and second stages—cultures hanging halfway
between magic and religion—tend to bind together rituals and myths,
Frazer proposed. More specifically, in these stalled cultures, myths are used
to lend meaning to rituals, and rituals reenact myths in such a way as to
ensure that the action narrated by the myth will recur in the real world.
Frazer’s second idea was that many myths and rituals center on a figure
who sometimes appears as a god and sometimes as a king but who always
represents the vegetation, and thus the vitality of the world. Like the
vegetation, this figure has to periodically die and be “reborn” in order to
renew the world’s vitality; many rituals enact this death and rebirth either
metaphorically or in reality. Frazer proposed that Adonis, Attis, Osiris,
Balder, and Dionysus were instantiations of this dying-and-reviving god,
which doesn’t stretch credibility too far, but he also proposed that many, far
less obvious candidates belonged in the group as well: Jupiter and the entire
mythical dynasties of Boeotia and Thessaly, for example.
The paradigm of the dying-and-reviving god caught the imaginations not
only of other scholars, who used it to interpret the myths and literatures of
the cultures they studied but also of literary authors, who helped to bring it
to wider public attention—T. S. Eliot evokes it in The Waste Land and
William Butler Yeats in Sailing to Byzantium, for example. This, in turn,
gave the accompanying idea that myths and rituals were mutually
dependent a remarkable tenaciousness. The paradigm also helped to revive
an older and more general paradigm: the idea that, even in the absence of a
central god or king, many myths and rituals were at heart about the annual
growth, death, and regeneration of plants, however deeply buried this
agenda might be beneath what were dismissed as superficial features.
Versions of this paradigm had been around in various forms since the time
of the ancient allegorists, but by focusing it upon the figure of a god or king
who dies and is resurrected, Frazer gave it new élan and, within the largely
Christian West, a new relevance.
This gave new life to an idea that had been around since the early
nineteenth century, when comparative mythology had become popular:
namely, that there were a limited number of “Ur-myths”—myths that had
been shared by all or most cultures since the beginning of time, however
varied the forms of expression they took. The identification of these Ur-
myths and their underlying meanings was, in fact, amongst the most
pressing tasks of a comparativist such as Frazer. The search for Ur-myths

(and the Ur-figures who populated them, such as Max Müller’s solar deity
or Frazer’s dying-and-reviving god) had led to the scholarly habit of
reducing any given myth to what one had decided was its true essence—
which in turn meant disengaging the myth from its narratives, which were
understood to be merely its vehicles. Thus, for Frazer, who provided fifty-
two pages of parallels for the story of Odysseus and Polyphemus in his
commentary to Apollodorus’s Library of Mythology, what seems to have
mattered most was not how Apollodorus or Homer or anyone else had told
the tale but rather the fact that certain episodes within it were echoed by
episodes in similar stories from around the world. For example, Frazer
proposed that Odysseus’s taunting of the blinded Polyphemus, which led
Polyphemus to throw rocks at Odysseus’s ship, echoes stories in which a
hero has a ring or some other magical object stuck to his body that cries out
to a blinded foe as the hero tries to escape. Frazer concluded that behind all
of these stories there must be “a common original, whether that original was
the narrative in the Odyssey, or, more probably, a still older folktale which
Homer incorporated in his epic.”6
Frazer was by no means consistent in his presentation of this idea (how
could he be, over the span of a continuously evolving, twelve-volume work,
to say nothing of his other publications on the topic?), but generally
speaking, at least for the cultures on which he focused his attention—the
cultures stranded between the first and second evolutionary stages,
including ancient Greece—he suggested, in contrast to Smith, that myths
were chronologically prior to the rituals with which they were paired.
Rituals enacted what myths had already described; without the preexisting
tale of a god of vegetation who dies and is revived there could be no ritual
in which a human king, standing in for the god, was killed and revived in
order to ensure the return of that god and the vegetation that accompanies
him.
The next important figure in this story—and probably the most important
figure in this story, period—returned to Smith’s prioritization of ritual. Jane
Ellen Harrison, who like Frazer was associated with the University of
Cambridge off and on throughout her career, was at the center of a group of
scholars who later came to be called the Cambridge Ritualists. The label is
not entirely correct: they were never a formal group, as the capitalization of
“Ritualists” might imply, and “Cambridge” is too limited a geographic tag,
given that one of them, Gilbert Murray, taught at Oxford and that the

theories of (the German) Friedrich Nietzsche and (the French) Émile
Durkheim were at least as important to their version of the ritualist
paradigm as were Frazer’s. The fact that the label has nonetheless stuck
tells us something about how deeply later generations were invested in the
theory that the group was understood to promote; situating its birth within a
collective of scholars working at one of the fonts of European learning gave
it additional legitimacy.
The word “collective” also stretches a point. Murray and Francis
Cornford, who are always included amongst the “Cambridge Ritualists,” as
well as A. B. Cook, who is sometimes included, certainly had important and
influential careers as classicists and made contributions to the theory and its
popularization. Murray’s widely read Four Stages of Greek Religion,
revised as Five Stages of Greek Religion, was particularly significant in
underscoring the primitive stage of Greek religion (a world ruled by “Giants
and Gorgons”) that had preceded the time of “quiet splendour” overseen by
the Olympians.7 It was Harrison, however, who put a distinctive stamp, as
well as a dark glamour, upon it. One of the first female scholars to make an
international name for herself, she knew how to play to her audience: her
flamboyant mode of dress and her use of the magic lantern (a proto-version
of PowerPoint and at the time the latest thing in technology), in
combination with the intrinsic excitement of her material and
interpretations, gave her lectures a drama seldom found in the rhetorical
desert of academia and helped to carry her ideas well beyond Cambridge.8
Harrison took up Frazer’s dying-and-reviving god and gave him a
different name, Eniautos-Daimon, a Greek term of her own invention that
meant “Year-Spirit.”9 Her introduction of this figure rested on an ancient
hymn that had been recently discovered in a sanctuary of Zeus on Crete.
She proposed that the hymn was the libretto of a ritual performed annually
to welcome home Zeus in the guise of a young man (kouros). The advent of
this god reawakened the vitality not only of the fields and flocks but also of
the social order itself. And with the involvement of the social order, things
took a new turn. In a development that cast a very long shadow over the
study of Greek religion, Harrison proposed that the ritual celebrating the
return of the divine kouros simultaneously functioned as an “initiation”
ritual like those that anthropologists were discovering amongst primitive
tribes of her own day. That is, the hymn was composed to accompany a
ritual in which adult status was conferred on local youths, and Zeus, as the

kouros of the hymn, represented those youths.10 Durkheim’s sociological
approach to religion is clearly evident here, but Frazer himself provided the
most valuable example for Harrison’s initiation paradigm, which he found
amongst the Wiradthuri, aborigines of New South Wales.11 Here again, we
see comparativism at work and, in its wake, the idea that, if primitive
cultures elsewhere had initiation rituals and myths, then ancient Greece
must have had them as well, and in approximately the same forms, which
we would be able to discover if only we could prevent ourselves from being
distracted by extraneous particulars that disguised them.
In this case, those extraneous particulars included the fact that whereas
initiation myths and rituals typically narrate or dramatize the advent, death,
and resurrection of their central characters, the Cretan hymn tells about only
the advent of the kouros—not his death and resurrection. To help fill the
gap, Harrison drew on myths in which Zeus, Dionysus, and an enigmatic
god known as Zagreus appeared as babies or young children who were
guarded by divinities known as Kouretes; in some of these myths, the
young god was killed in spite of the Kouretes’ care and then was
resurrected. With what Harrison assumed was the full myth behind the
Cretan hymn now in hand, she reconstructed what must have had happened
in the Cretan ceremony, closely following along the lines of Frazer’s
example from New South Wales and concluding that “with the Cretan ritual
in our minds it is clear that the Wiradthuri rites present more than an
analogy; mutato nomine, the account could have been written of Zagreus.”12
In spite of Harrison’s precarious mode of argumentation (the myth behind
the hymn is reconstructed from other myths, and the reconstructed myth is
paired with selected aboriginal comparanda to reconstruct the ritual, which
then matches up nicely with the reconstructed myth), she was more or less
right as far as subsequent scholarship has concluded: at some point, what
we can call an initiation-type ritual probably was performed for young men
on Crete, and the myths of Zeus, Dionysus, and Zagreus probably reflect
that ritual (although as we see them in our existing sources, these myths are
being put to other uses).13 Not content to stop with the Cretan material,
however, Harrison strove to prove that her own version of the initiation
paradigm, centering on the Eniautos-Daimon, was more widely established
in ancient Greece. She claimed that it lurked behind the myths of Heracles,
Achilles, and Neoptolemus and, more distantly, the fates of mythic sinners

condemned to suffer punishment in the Underworld, for example. Francis
Cornford contributed a chapter to her book in which he argued that it
underlay the myth of Pelops, as well; Gilbert Murray contributed a chapter
in which he suggested that aspects of Theseus, Hippolytus, and other heroes
could also be explained by this paradigm.14 The fact that some of these
characters were killed but never resurrected didn’t seem to pose a problem.
Nor did the fact that some of them underwent their deaths and
“resurrections” when they were fully grown. Myths, in the opinion of the
Cambridge Ritualists, were notoriously unreliable, forever drifting away
from representing rituals accurately until a scholar yanked them back into
line. A statement about this issue that Harrison made early in her career
became emblematic not only of her own work but also of the ritualist
approach more generally: “ritual practice misunderstood explains the
elaboration of myth.”15
This is not to say that for Harrison, strictly speaking, rituals had to exist
before myths, as they had for Smith. At one point, she conjectured that the
two could have arisen pari passu, with the first myths (muthoi) being
nothing other than the emotionally charged, barely articulate outcries
(“mu!”) made by participants when they performed the rituals. Slowly, she
suggested, these cries evolved into stories that narrated what the ritual
enacted and then, much later, became aitia that justified a ritual’s existence
in much the way that Smith had imagined.16 But her conjecture that myths
and rituals had arisen simultaneously at the dawn of human existence in no
way changed her opinion that the myths we inherited from classical authors
were secondary growths, aggravatingly distorted by the niceties of the
narrative vehicles that carried them forward in time. These myths would
always be only handmaids to rituals, in the sense of justifying their actions,
and handmaids to scholars in the sense of helping them reconstruct those
rituals, which was the real prize after which the scholars quested. Like
handmaids, in other words, the myths that scholars had inherited needed to
be properly disciplined if they were to be of any use. By their nature, they
tended to dissemble, hiding their true meanings behind the loveliness of
their expression. They were, for Harrison, “shifting,” “manifold,”
“unsatisfactory,” and even “absurd.”17
This is where Nietzsche’s influence on Harrison becomes especially
interesting.18 Harrison turned the schematic opposition between what

Nietzsche called the Apollonian and Dionysian forces in Greek culture into
her own evolutionary (or rather devolutionary) theory of Greek religion. In
the earliest stages of Greek culture, she suggests, the Greeks were
Dionysian; they practiced genuine rituals that were full of excitement,
enacting raw, primal instincts. Their “gods” were not really gods at all but
rather daimonic projections of Durkheimian collective effervescence. As
these daimones began to be anthropomorphized, they became “ghosts,
sprites and bogeys,”19 to whom one dedicated rituals that had been
previously performed without any thought of a recipient. These bogeys then
developed into more clearly differentiated entities such as the Erinyes, and
from them, in turn, there developed goddesses concerned with ethics and
morals, such as Athena. Eventually, the bogeys were altogether displaced
by the family of Olympian gods, and Greek religion entered its fully
rational “Apollonian” period—that is, it moved to the other end of the
schematic spectrum. For Harrison, this period was characterized by the
subordination of genuinely religious rituals to a religion dominated by
myths that were intended primarily to showcase Olympian splendor.20
This was not the sort of religion for which Harrison felt any affinity:
“Great things in literature, Greek plays for example, I most enjoy when
behind their splendours I see moving darker and older shapes,”21 she wrote
in her autobiography of 1925. The sentiment echoes the broader mood of
the time, a period when an irrational, and even savage, side of human nature
seemed to lurk just beneath its civilized shell, ready to burst through at any
moment. The fifteen years preceding Harrison’s Prolegomena saw the
publication of Robert Louis Stevenson’s Dr. Jekyll and Mr. Hyde (1896),
Oscar Wilde’s The Picture of Dorian Gray (1890), H. G. Wells’s The Island
of Dr. Moreau (1896), and Bram Stoker’s Dracula (1897), to mention just
the high points of a literary trend towards tales of “the savage within.”22
Hand in hand with a desire of the time to peer into the dark recesses of the
soul went a desire to peer at the soul beyond the curtain of death, as well.
Interest in spiritualism soared, and classicists were not exempt: Gilbert
Murray was a “sensitive” (that is, a medium); A. B. Cook had a strong
interest in uncanny phenomena; Margaret Verrall—the old friend of
Harrison’s with whom she coauthored Mythology and Monuments of Athens
and who, along with her husband, the classicist Arthur Verrall, sat on the
fringes of the Cambridge Ritualist group—was a psychic investigator.23

Somewhat earlier (1882), the Society for Psychical Research had been
cofounded by Frederic Myers, a Cambridge-trained classicist and former
fellow and lecturer at Trinity College; Murray served as its president
twice.24 Yet another marker of the mood of the time was the emergence in
England of modern pagan witchcraft—a movement that sought to recover
earlier, more genuine religious impulses (read: “rituals”), which was
nourished by the widely popular writings of Frazer and Harrison.25
Late in life, Harrison turned away from classics and embarked on the
study of Russian literature. Nonetheless, during this period she produced
two books that explicitly focused on ancient myth: Mythology (1924) and
The Myths of Greece and Rome (1928). Of the latter there isn’t much one
can say. It is tiny in size, which suits the series in which it appeared
—“Little Books of Modern Knowledge”—and it concentrates on “The
Gods of Homer’s Olympus,” as a section heading proclaims. It was
published in the year of Harrison’s death and was probably written at some
point during the three preceding years, while she and her friend Hope
Mirrlees shared a flat in London, working together on translations of
Russian novels.26 One wonders whether Harrison wrote it because money
was tight.
The other book, Mythology, is more interesting. It must have been written
while Harrison and Mirrlees were living in Paris, shortly after Harrison
departed from Cambridge and the life of a classicist. Like The Myths of
Greece and Rome, it was written for the average reader: it was the twenty-
sixth volume in a series published by the Marshall Jones Company under
the rubric “Our Debt to Greece and Rome” (in the postwar era, the public
was eager to be reminded of the more admirable aspects of European
civilization). Nevertheless, even in this venue, Harrison manages to make it
clear that “real” Greek religion had to be excavated from beneath the bright
surfaces laid on by myths. She begins by reasserting her familiar distinction
between ritual (“what a man does in relation to his religion”) and myth
(“what he thinks and imagines”) and then throughout the book continually
reminds her readers that myths as we know them were created by poets and
artists, who were at heart shapers and makers—that is to say (as anyone
who has read Plato knows quite well) dissemblers: “The Greeks were not
priest-ridden, they were poet-ridden, a people, as the word ‘poet’ implies, of
makers, shapers, artists.”27 Or again, “From religion Greek mythology

banished fear, fear which poisons and paralyzes man’s life.… Some rites of
fear and repulsion they kept, for ritual is always conservative, but their
mythology and theology, their representations of the gods, was informed
throughout by reason, lighted by beauty.”28
After Harrison
Harrison’s Mythology was the last time, for a long time, that a classicist
invoked the ritualist approach.29 Amongst Semiticists, however, it was just
then gathering speed—and so, too, was the comparativism that always
seems to accompany it. Its main proponent after Smith, Samuel Henry
Hooke, was a generation younger than Harrison. He started his academic
career late in life and spent a good part of it away from British academia—
he returned to England with a professorship in Old Testament Studies at
University College London only in 1927, at the age of fifty-three. His first
major publication on the topic, an edited volume entitled Myth and Ritual,
appeared in 1933, twelve years after Harrison’s Epilegomena, and his first
monograph on the topic, The Origins of Early Semitic Ritual, appeared in
1938.30
Like Harrison in her later years, Hooke understood a myth to narrate the
actions of a ritual as it was being performed, and therefore he understood
the two to be of virtually simultaneous origin (for this reason, the approach
as it was developed by Semiticists is formally called the “myth-and-ritual
approach” rather than the “ritualist approach”). Hooke also embraced the
paradigm of the dying-and-reviving god / king and gave pride of place
within it to the ancient Babylonian Akitu festival, a New Year’s celebration
during which the human king was first deprived of office and then
reinstalled while a priest recited the Enuma Elish, a cosmogonic poem that
culminates in the installation of Marduk as king of the gods.31 Hooke argued
that it was with this particular myth-and-ritual pairing that the paradigm of
the dying-and-reviving god / king had originated and then traveled wide and
far, expressing itself through many other myth-and-ritual pairings. In other
words, Hooke presented the story of Marduk as the Ur-myth par excellence,
from which all others had descended and then been diffused around the
world. As Hooke’s work spread, it fanned not only the flames of the myth-
and-ritual approach and of comparativism but also the flames of
essentialization: scholars felt newly encouraged to purge away any

inconvenient details in a given myth or ritual, in order to show that its
structure was parallel to what Hooke had described.
It wasn’t only Semiticists who continued to use and develop the models
that the classicists had pioneered, however. At midcentury, the literary critic
Stanley Hyman, for example, pushed Harrison’s ideas so far as to argue that
ritual underlay not only all myths but also all literature. In his view,
literature had sprung from myths whose connections to rituals had been
weakened. Hyman was enthused about Gilbert Murray’s 1914 attempt to
trace Hamlet to the same ritual origins as the story of Orestes, and he also
admired Jessie Weston’s 1920 use of the theory to illuminate Arthurian
legend.32 The ritualist approach to myth was also taken up by Lord Raglan.
In his 1936 The Hero: A Study in Tradition, Myth and Drama, Raglan
extended the thesis of the dying-and-reviving god into hero tales (Oedipus,
Heracles, Moses, and Robin Hood were amongst his examples). One of
Raglan’s most enduring contributions is his observation that twenty-two
traits frequently appear in hero stories throughout the world (although
almost no story includes all of them). For instance, the circumstances of a
hero’s conception are often unusual and he often meets a mysterious death.
Raglan’s neatly enumerated formulation makes it clear not only that
ritualists are almost always comparativists, too, but also that the sort of
comparison they prefer focuses on similarity—that is, on collecting as many
instantiations as possible of a given “type” but paying relatively little
attention to their differences. This, again, encourages the essentialization of
any single myth so as to make it fit whatever pattern a scholar is
championing.
The ritualist approach was taken up by anthropologists, as well, although
hesitantly at first. Bronislaw Malinowski and Clyde Kluckhohn agreed that
there was a tendency for myths and rituals to depend upon each other, but
Kluckhohn warned that the variety of ways in which they might do so made
monolithic theories about how and why it had happened dangerous. He also
rejected the idea that there could be any all-encompassing thesis as to which
had come first. In a 1942 essay that presaged some aspects of Walter
Burkert’s work, Kluckhohn universalized only so far as to say that myth and
ritual “both are symbolic processes for coping” with the anxieties that life
presents and that they are a “cultural storehouse of adjustive responses for
individuals” that “supply fixed points in a world of bewildering change and
disappointment.”33 In 1954, nonetheless, Edmund Leach confidently

asserted that “myth implies ritual, ritual implies myth, they are one and the
same,”34 and it was his voice that carried the day: the paradigm was firmly
implanted amongst anthropologists for a long time to come.
The Classicists Return to the Fray

In 1959, the classicist Joseph Fontenrose criticized the version of the
ritualist approach that literary scholars had developed. In his Python: A
Study of Delphic Myth and Its Origins, Fontenrose argued that although
myths were “traditional stories that accompany rituals,”35 myths did not
necessarily originate in rituals but might instead become “attached” to
rituals at a later stage, in one way or another. He also took up the
diffusionist banner of the myth-and-ritual school, but with a new emphasis:
he assumed that once a myth had begun to spread from its place of origin, it
might join forces with a variety of different rituals that it encountered and
itself be changed in the process. Much of the book is spent demonstrating
that what Fontenrose called the “combat myth”—that is, the tale of a hero
defeating a monster—enjoyed just this sort of diffusion, linking itself now
to the Babylonian New Year’s ritual, then to certain Egyptian rituals and
myths, and then again to the Delphic Septerion festival, for example. “It is
simpler to suppose,” he wrote, “that a well-known type of story was
introduced in many places to serve as the primeval precedent of the rituals
than to believe in so many places the rituals spontaneously generated a
uniform pattern of myth.”36 Like others, then, Fontenrose assumed that
there were Ur-myths (or at least Ur-types of myths), but the way in which
he understood them to have spread throughout the world meant granting
them much more independence and flexibility than earlier scholars had
been willing to. His version of comparativism left room for differences, as
well as similarities.
Stanley Hyman reviewed Python in 1960, in the first issue of a magazine
called the Carleton Miscellany. It’s worth pausing on the venue of this
review: during its twenty years of publication, the slender and highly
selective volumes of the Miscellany brought together essays, poems, and
works of criticism that were contributed not only by academics but also by
poets laureate and winners of the Pulitzer Prize: the Miscellany aimed to
please and stimulate the broadly educated, intellectually inclined person. In
other words, the myth-and-ritual question was once again perceived as
something central to the life of the mind.

Hyman’s review marshaled the ghosts of Smith, Frazer, Harrison,
Murray, and Cornford (who collectively, he said, had definitively
established the real, ritual origins of myth) to support his contention that
every aspect of Fontenrose’s approach was utterly wrong. What was under
threat for Hyman and for many other people as well, we might guess, was
the treasured idea that “if myths arise out of ritual, [then] they are an
expression of deepest human needs, of leaping for crops and flocks and
goodly themis, and [they] have a profound sociological and psychological
truth, whatever the literal unreality of their stories.”37 If one took away the
direct link between myths and rituals, then perhaps the whole assumption
that myths meant something—something primal and genuine about our
deepest selves—might collapse. For Hyman and others like him, this turn of
events threatened to pull the rug out from under a lot of literature, too.
Fontenrose gave as good as he got in the next issue of the Miscellany,
concluding his response to Hyman, with:
I know of no competent anthropologist, folklorist or classicist (or none

at all, in fact, unless you call Lord Raglan an anthropologist) who
accepts the view that Mr. Hyman considers to be well established. But
it does seem to be the rage at present among literary critics, none of
whom has done any of the spade work in mythological study.38
More importantly, earlier in his response Fontenrose brought out an

underlying assumption of Python more effectively than he had in the book
itself: “once you have myths,” Fontenrose asks, “why can’t they suggest
new myths?”39 In his next book, The Ritual Theory of Myth (1966),
Fontenrose renewed his attack against the ritualist approach by further
emphasizing the imaginative contribution that narrators make to myth.
Myths do not exist in some abstract form, he insisted, but rather are
promulgated through literary and artistic narrations, and in the process they
are continually and productively re-created to suit their audiences. In other
words, when one essentializes a myth—when one simply boils it down to
the Ur-myth from which it supposedly emerged and leaves it at that—one
loses a great deal of what actually constituted the myth as it was received
by people. This is a crucial point, as is a related plea that Fontenrose made
for studying myths “not in a vacuum, but in their institutional settings.”40

For the Greeks, this meant studying myths as they were performed at
festivals, first and foremost. Even as he rejected the ritualist approach to
myths, in other words, Fontenrose returned myths to their ritualized
settings.
Shortly after Fontenrose’s The Ritual Theory of Myth appeared, Geoffrey
Kirk published his Sather Lectures on myth, in the course of which he
briefly addressed the myth-and-ritual question (Myth: Its Meaning and
Functions in Ancient and Other Cultures, 1970). Kirk’s return to the topic
in 1974, in chapter 10 of The Nature of Greek Myths, presents a more
developed version of his ideas. Like Fontenrose, Kirk thought that the
connection between myth and ritual had been exaggerated, at least for the
Greeks, and therefore argued strenuously against its use by classicists. He
pointed out that in both ancient Greece and other cultures, there were stories
that anyone would be apt to call myths that had no apparent connection to
ritual at all. His prime example was the Greek theogonic myth: “the ancient
Greeks did not carry out actions designed to imitate or reproduce the
separation of sky and earth.”41 He also distinguished between “significant”
myths and “insignificant,” “trivial,” “feeble,” or “half-baked” myths. For
Kirk, the first category included myths that either had been associated with
a given ritual over a long period of time (which for him meant aitia that
were hoary with age) or were well-known through important literary
sources (the theogonic succession myth being a good example, as it comes
to us through Hesiod). The second category constituted myths that, in
Kirk’s opinion, had neither literary importance nor a long-standing
attachment to ritual, such as an aition for the Anthesteria that tells of
Orestes’s arrival in Athens on the day of the choēs. Kirk judged this to be a
late addition to the festival and “a feeble affair.”42
This mode of categorization meant, in turn, that for Kirk there were
“significant” links between myths and rituals (although the only two Greek
instances to which he granted this designation were those connected with
the Lemnian women and those connected with the daughters of Cecrops),
and there were “insignificant” links (into which category fell the vast
majority of links that scholars had proposed in the past). Regarding the
myth of Orestes that purported to explain the ritual performed during the
Anthesteria, Kirk concluded that in this and many other cases, “pre-existing
myths, or faintly plausible details from them, are dragged in as aitia [for
rituals], which is a very different process from solid mythical invention.”43

In a similar vein, the fact that there were several different aitia for the
Buphonia disqualified all of them, in his view: in this case, the “degree of
myth involved was minimal and the generation of real narrative extremely
slight.”44 The story of Orestes biting off his finger and then establishing cult
to the Eumenides was similarly judged “a superficial aetiology.”45 The myth
of Demeter and Persephone was labeled a “crude aition for seasonal
agriculture [that in adapted form] was applied to a particular fertility cult
that had been known in Eleusis at least from the Mycenaean period,” and
the story of Neoptolemus’s death at Delphi was judged a “minor invention
that drew on the ritual use of knives there in an unimportant way.”46 More
generally, in Kirk’s opinion, what rituals frequently had done was to
“encourage half-baked aitia in the form of loosely applied or ill-chosen
details from other and obviously independent tales.”47
In other words, Kirk rejected the ritualist approach to myth not so much
because he had a problem with the approach itself as because he saw little
evidence of its relevance to the analysis of Greek myths. In making this
call, ironically, Kirk presented aitia as being more central to the ritualist
approach than some other scholars had (recall that Harrison, in fact, found
the aitiological link between myths and rituals to be relatively late and
secondary). Kirk also seems to have thought that he knew better than any
ancient author what made for an appropriate myth-and-ritual combination.48
Myths, Rituals, and Programs of Action

To some extent, Fontenrose and Kirk were flogging a dead horse—or at
least a horse that most classicists had stopped riding a long time ago.
Fontenrose initially revived discussion of the approach because his own
topic of interest (the combat myth) had motivated him to develop a new
variation of it. Subsequently, he critiqued the approach as it was
traditionally applied because he had come to realize how deeply it was still
influencing literary studies. Kirk dissected the approach as part of a broader
criticism of monolithic approaches to myth during the twentieth century
(although he seems to have taken particular delight in attacking this one).
At about the same time that Fontenrose and Kirk were speaking out
against the ritualist approach, it was finding its way back into classics by an
unexpected route. In 1972, Walter Burkert published Homo Necans and
then, in 1979, Structure and History in Greek Myth and Ritual. One of the

most important ideas that Burkert explored in these two books, as well as in
a series of articles,49 was the premise that both myths and rituals are
symbolic expressions of “programs of action” that have deep roots in
biologically determined events such as puberty and the seeking of a mate or
in social realities such as the hunt. Given this, he suggests, myths and
rituals usually arise independently of each other, although they sometimes
end up functioning in tandem. Like Fontenrose’s insights, this laid the
groundwork (as Burkert himself signals several times)50 for better
appreciating the contributions made by specific narrations of myths, and yet
the two greatest effects of Burkert’s work were to spur onward again the
search for myth-and-ritual pairings and, correspondingly, to encourage
further still the essentialization of myths.
The revival of the search for pairings can be traced to two aspects of
Burkert’s work.51 First, Burkert used programs of action to offer a
particularly captivating revival of Jane Harrison’s argument that the
initiation paradigm underlay many Greek myths and rituals—which
inevitably revived her ritualist approach to myth as well.52 Renewed
attention had already been brought to initiation in the ancient world by
Henri Jeanmaire in 1939, and then again by Louis Gernet and Angelo
Brelich in the 1960s, but by combining initiation with ideas borrowed from
ethologists such as Konrad Lorenz and with the idea (shocking at the time)
that violence lay at the heart of many religions, Burkert galvanized the topic
like no one else had before.53 His timing was right, moreover; the younger
classicists who took their cues from Burkert, such as Hendrik Versnel, Fritz
Graf, Jan Bremmer, and Christiane Sourvinou-Inwood, had come of age
during the 1960s, a period when the Western world developed a heightened
awareness of society’s power to enforce normative expectations of behavior
during adolescence.54
Second, while endorsing the idea that myths and rituals could operate
independently, Burkert nonetheless confirmed that they were more effective
when working together and offered attractive interpretations of myths that
relied on their partnership with rituals. Odysseus’s sufferings in
Polyphemus’s cave could be shown to echo a werewolf story pattern—
which in turn reflected initiatory rituals. The tale of Proetus’s maddened
daughters and their gruesome sacrifice of a child could be linked to a
festival that marked the temporary dissolution of society and suspension of

its rules. The story of Demeter’s angry withdrawal and mollified return
could be paired with widespread rituals in which images of deities were
ceremoniously taken away and then retrieved.55 Although Burkert had
already reminded his readers of an important point regarding Greek myths
in an earlier publication that laid some of the groundwork for his books:
“[In ancient Greece] we are not dealing with myths that perform a direct
ritual function. In [the present] case, nothing suggests that the myth of
Kekrops’ daughters was officially recited during the festival of the
Arrhephoria, as the creation epic would have been at a given time and place
during the Babylonian new year festival,”56 this and other cautionary
statements paled in the light of his exciting analyses. In the more than forty
years since Burkert first published on the topic, numerous scholars,
following what they take to be his lead, have proposed further myth-and-
ritual pairings that pay relatively little attention to the issue of how, exactly,
the myths were narrated or enacted in their particular ancient settings.57
The second effect of Burkert’s work was to further encourage the
tendency towards essentializing myths and rituals. Scholars who embraced
the ritualist approach were already liable to do this—in part because they
also tended to be comparativists, as we have seen, and in part for another
reason that I will discuss shortly—but Burkert’s ideas added fuel to their
fires for an additional reason. Building on the work of Vladimir Propp, who
had proposed that a limited sequence of thirty-one actions (or “functions”)
underlay each of the 100 Russian wonder tales he had examined and that a
limited pool of actors (“hero,” “villain,” “donor,” and so on) populated
those tales,58 Burkert developed his own sequences, which he suggested
reflect “the reality of life”—that is, they narrate or enact biologically driven
programs of action. For example, Burkert proposed that there was a
connection between a program of action experienced by virtually all women
in premodern societies (puberty / loss of virginity / pregnancy / birth) and a
sequence of story motifemes that he called “the girl’s tragedy”: (1) a girl
leaves home, (2) she is secluded from others, (3) she is raped and
impregnated, (4) she is severely punished by her parents or other relatives,
and (5) she is rescued, sometimes by the son to whom she gave birth. Given
the universality of this program of action, Burkert reasoned, we should not
be surprised to discover that “the girl’s tragedy” is a nearly universal type of
tale.59

Burkert is careful to remind his readers that such sequences cannot tell
us, on their own, what a myth “means.” Early on in Structure and History in
Greek Myth and Ritual, he offers the following definition: “myth is a
traditional tale with secondary, partial reference to something of collective
importance,”60 and he reminds us several times that a significant part of any
myth’s relevance stems from its application in a specific place, at a specific
time, in response to specific circumstances. He rejects the search for
“origins” in the traditional sense, and in the course of the book (which, after
all, includes the word “history” in its title), he demonstrates several times
how a myth can “crystallize” in a certain form that includes particular
details and then “re-crystallize” with new details under new
circumstances.61 For Burkert, tracing a myth to a more general type or to a
program of action is only a first step in understanding it.
And yet it was Burkert’s programs of action—the primal modes of
behavior hiding behind the niceties of history, the hermeneutically alluring
“structures” that provided the first word in the book’s title—that grabbed
his readers’ attention and inspired a new flurry of publications on myth.
One of these is especially interesting for our purposes. In 1993, in a detailed
review of the ritualist approach, Hendrik Versnel demonstrated that its
advocates had long been fixated on two paradigms, each of which was
believed by its backers to be the interpretive key that could reveal the true
meaning of most, if not all, myths and rituals. The first was the New Year’s
paradigm, which included Frazer’s dying-and-reviving god, for example,
and the second was the initiation paradigm, as explored most famously by
Harrison and then again by Burkert. The fervor to apply these paradigms
had tempted scholars into jettisoning inconvenient details, adjusting
plotlines and tailoring myths and rituals in various other ways in order to
streamline them into one paradigm or the other and thus reveal their real,
core “meaning.”62
Versnel tried to avoid doing this by looking beyond the two paradigms.
He noted that they each shared structural affinities with a third pattern that
he called the “primordial crisis”—that is, a sequence in which a character
leaves familiar territory, confronts dangers in a landscape marked by
marginality (in some cases to the point of death), and triumphantly returns
in a state of renewal. The Odyssey, which advocates of both the New Year’s
paradigm and the initiation paradigm claimed to be able to decipher,
provided Versnel with a perfect example of the primordial crisis in action

and an opportunity to bring the warring factions together. Perhaps, he
suggested, the Odyssey was able to be read as both a New Year’s tale and
an initiation story because at heart it was really a tale of primordial crisis.
Many other contested myths and rituals could be similarly understood in
this way, he claimed—and one needn’t stop there: the plot is also behind
numerous fairy tales, other kinds of stories and even the dreams of
individuals. The pattern is widespread, he concluded, because its roots are
planted in Burkert’s programs of action—that is, the pattern reflects
biological and social drives of an exceedingly old and elementary nature.
Thus, in order to avoid the two paradigms that ritualists had embraced for
decades, Versnel introduced another paradigm that essentialized myths in a
different way—for what is left of a myth once it has been condensed to a
three-part primordial crisis? Versnel was well aware of this problem; on the
penultimate page of his essay, he said, “Of course, whoever thinks all this
much too vague and prefers to sit down and reread the Odyssey itself is
right, too”—a nod to the importance of the well-wrought tale itself.63
Meanwhile …
Over the past twenty pages or so, I’ve sketched the effects that the ritualist
approach has had on scholars of ancient Greek myths. Born out of the
comparative method, the ritualist approach was susceptible to essentializing
the myths it treated, since comparison itself had a long history of stripping
myths down to what was perceived to be their cores and either ignoring or
explaining away inconvenient details. Hand in hand with this sort of
comparison went the assumption that there were Ur-myths, from which
other myths had evolved. To reconstruct an Ur-myth, one had to cut away
the extraneous matter that had accumulated around its descendants in their
different cultural contexts. For a comparativist, this meant that Greek
myths, like all other myths, needed to be pruned before they were useful.
The ritualists’ own agenda further encouraged the essentialization of
Greek myths for two reasons. First, because the ritualists thought of myths
as ancillary phenomena, created to serve as librettos for the rituals that
constituted “true” religion, they were willing to refashion myths to fit
whatever patterns they saw in the rituals. Second, because they assumed
that poetry and art had embellished Greek myths to the point of disguising
any genuine religious feeling that they might have once expressed, the

ritualists were willing to extricate what they decided the myth really was
from what they claimed were merely its artistic or literary vehicles.
In the 1960s and 1970s, Burkert tried to put myths and rituals on an equal
footing by suggesting that they both had emerged from biological programs
of action. In doing so, he endorsed Fontenrose’s suggestion that myths
could develop and operate independently of rituals, even if they sometimes
joined forces, and Fontenrose’s emphasis on the imaginative contribution
that individual narrators could make to myths. This should have pushed
scholars of ancient religion to reconsider the narratives through which
myths were expressed in ancient Greece, but Burkert’s work instead
inspired a new spate of myth-and-ritual pairings that, just as in earlier times,
tended to focus on finding the “true essence” or the “original meaning” of a
myth. This quest, once again, led to treating what was understood to be the
“myth” separately from what were understood to be its narrative “vehicles.”
For scholars whose training and interests lay primarily in ancient
religions, the ritualist approach has long been the most popular way of
studying Greek myths. Other approaches to myth emerged during the
twentieth century as well, however, including, most prominently, the
psychoanalytical and structuralist approaches. The psychoanalytical
approach assumes that myths reflect universal concerns of the human
psyche; it follows that myths are bound to share a universally valid system
of symbols, too, which can be recovered from individual myths if one looks
beneath their surfaces. This, in turn, means discarding or reinterpreting
details in the ancient narratives.
The structuralist approach requires one to discard, or at least to disregard,
surface details, too. According to Propp’s brand of structuralism, everything
except the plot and some basic character roles is disposable; one can
change, for instance, the name of the hero, the destination to which he
travels, or the very goal of his quest without changing the core meaning of
the story. According to Claude Lévi-Strauss’s version of structuralism, in
contrast, plot is irrelevant; what matters are the smaller units of a story that
he called “mythemes” and how they articulate a culturally embedded
system of binary oppositions.64 What Propp and Lévi-Strauss had in
common was that each thought that the particular narrative through which a
story was told was of little importance; each of them worked from a
simplified version of whatever story they chose to treat. Lévi-Strauss
created his simplified version by homogenizing as many possible variants

of the story as he could find into a sort of abstract. Propp implicitly did the
same, deriving a model from the 100 exemplars that he studied, against
which every individual exemplar could then be measured.
Most scholars who adopted a structuralist approach tended to treat the
myths that they dealt with in one of these two ways. Jean-Pierre Vernant
was an exception. Rather than working from abstracts of myths, Vernant
started from individual narratives. He paid close attention to the authors’
choices of words, to the ways in which the stories unfolded in narrative
time, and to the chains of causality that the authors chose to emphasize. In
other words, for the first time in the modern study of myth, Vernant was
interpreting myths in the specific forms through which they had been
conveyed to ancient audiences. Furthermore (under the influence of his
teacher, the sociologist Louis Gernet), Vernant contextualized narrations of
myths within their historical and social settings, so that he might more fully
recover the codes of meaning through which characters, objects, and actions
were used to convey a narrator’s ideas. This required Vernant to do close
readings not only of the myths themselves but also, alongside them, close
readings of a variety of other ancient texts and artistic representations. This
enabled him to show, for example, that Hesiod’s description of Pandora
having a “mind like a bitch” drew its force from a much broader Greek
collocation of female lust, dogs, hungry stomachs, and the burning heat that
the Dog-star Sirius brought at the height of summer, when women were
most wanton.65
Vernant’s method was carried forward by other scholars; I will be
drawing on some of their work in later chapters. Vernant’s student Marcel
Detienne needs to be mentioned now, however. Detienne turned away from
Vernant’s practice of interpreting a single, specific mythic narration and
back towards Lévi-Strauss’s approach, constructing each of his objects of
study as an “ensemble of its variants disposed in a series to form a group of
‘permutations.’ ”66 His particular rationale for doing this was that,
throughout the history of ancient Greece, myths had been continually
remade to suit their immediate environments. There were no canonical
versions, and in his opinion, one would get at what he called a myth’s
“hidden system” only by standing back and taking a panoptic view of all the
variants. Indeed, “the more numerous the variants, the better structural
analysis works.”67 For Detienne, then, a myth was both constituted by, and
disappeared into, its own multiformity.

It was perhaps Detienne’s fascination with multiformity that led him, a
few years later, to dismantle the very concept of myth as a unity, as scholars
had understood it. In L’invention de la mythologie (1981), he argued that
the category of myth as a distinctive form of discourse had been invented
during the fifth century BCE, by men such as Plato, Herodotus, and
Thucydides, who needed something to set in opposition to their newly
conceptualized category of “reason.” The modern study of myth, he
asserted, starting with figures such as Bernard le Bouvier de Fontenelle in
the eighteenth century, had replicated this move. In the Western world, then,
both “myth” and “reason” (or “science” or whatever else one chose to set in
opposition to myth) had never been more than cultural constructs. Greek
myth per se could not be isolated as a definable phenomenon, much less
studied as such.
Detienne’s book was just the first shot in what quickly became a larger
assault on the integrity of myth as a category. A year later, Luc Brisson
published Platon, les mots et les mythes, which looked more closely at
Plato’s innovative use of the word mythos to characterize stories that were
not true, and the year after that saw both Paul Veyne’s Les Grecs ont-ils cru
à leurs mythes?, which argued against the assumption that the Greeks had
understood “myth” as an essential mode of thought, and Fritz Graf’s
Griechische Mythologie, the first two chapters of which went even further
than Detienne had in tracing the development of the modern European
conception of myth as a particular type of story. For the next fifteen years or
so, attention shifted away from what had long been assumed to be Greek
myths (that is, all those stories in which Zeus, Athena, Heracles, and their
ilk were characters) and towards the very question of whether, or how, one
could genuinely isolate, much less define, any such thing.68 Although the
study of individual myths did not completely stop, methodological
innovation slowed down; the default approach continued to be that of
looking for myth-and-ritual pairings, even if with more flexibility.69
There was one important exception. Claude Calame had been amongst
those who came to the conclusion that “there simply is no ontology of
myth,”70 but he took things further. In his 1990 Thésée et l’imaginaire
athénien,71 and even more so in his 1996 Mythe et histoire dans l’antiquité
grecque: La création symbolique d’une colonie, Calame insisted that, far
from being a mode of thought, discrete or otherwise, for the Greeks myth

was a process of symbolic production that came to exist only in the course
of narrative production itself (for this reason, he called his approach
“semionarrative”). In other words, Calame insisted that scholars should
neither neatly excise “a myth” from its assorted “versions” nor magically
reconstitute it by agglomerating those versions into a homogenized whole.
Each “version” was in effect its own myth, arising in a particular form in
response to particular needs on particular occasions.72 Moreover, rejecting
the traditional relationship between myth and ritual, Calame argued that
myths themselves could have a pragmatic effect upon their audiences and
become instruments for change—affecting the emotions of the audience and
thereby shifting public opinion, for example. I’ll return to these ideas in
Chapter 3, where I’ll be asking how myths affect emotions and to what sorts
of ends they do so.
Aitia
In ending this chapter, however, I first need to return briefly to the ritualist
approach and look more closely at one of its most persistent features: the
tendency to emphasize myths’ aitiological functions. That there were plenty
of Greek myths that undertook to explain why this or that ritual existed is
not in doubt. The issue, rather, is that scholars have seldom stopped to ask
when and why these aitiological myths were narrated or performed.
Typically, there has simply been a vague understanding that a given aition
was linked to its partner ritual in some way—perhaps it was performed
during the ritual, for instance. Occasionally, this presumption led a scholar
to “recover” an underlying structure that a myth shared with other myths
that were more straightforwardly aitiological. Sometimes, it led to the
hypothetical reconstruction of rituals that no longer existed in historical
Greece but that scholars perceived as aligning both with the recovered
myths and with ritual paradigms that were borrowed from other cultures. In
the end, these sorts of operations required a good deal of imaginative re-
creation of both myths and rituals and a good deal of circular
argumentation.
In more recent decades, scholars have also paired myths and rituals by
looking for thematic resonances between the two (thus, for example, my
own proposal that the Homeric Hymn to Hermes was performed at boys’
athletic festivals, which relies on understanding both the myth and the
festival as articulations of male social maturation)73 or by looking for

contextual motivations that would have prompted a poet to narrate a
particular myth at a particular festival or at a particular time (thus, for
example, the various interpretations of Pindar’s second Olympian that rely
on assuming that its recipient, Theron of Acragas, belonged to a particular
religious cult).74 Certainly, both of these approaches have their virtues, but
they still leave us with a large number of myths that have no obvious
aitiological, thematic, contextual or other sort of connection to the
occasions on which they were narrated. Why tell the story of Deianira’s
accidental murder of Heracles during a dithyramb performed in Delphi, as
Bacchylides did, for example?75
Certain genres of poetry do seem to have had a tendency to take aitia as
their topics, the paean being a prime instance. Several paeans composed for
performance at Delphi tell us about how the Oracle was established there
(as does the Homeric Hymn to Apollo, which includes, towards its end, a
paean-like refrain).76 Alcaeus’s paean narrates Apollo’s first trip to
Hyperborea, after which he traveled to Delphi and assumed his duties there
as an oracular god. The story reflects cultic reality insofar as Apollo was
understood to leave Delphi in Dionysus’s care for three months each winter
while he enjoyed an annual vacation amongst the Hyperboreans. Pindar’s
sixth paean explains that Neoptolemus has a tomb in Apollo’s Delphic
precinct because he had been killed there in a fight over how the priests
were distributing sacrificial meat. Apollo’s conception and birth were also
appropriate subjects for paeans.77
But as a group, these paeans raise two questions that can also be asked of
many other myths that were narrated or performed in cultic settings. First,
what are we going to count as an aition? Broadly speaking, a poem that
narrates the foundation of the cult site probably qualifies, given that it
explains how the site became sacred—but such stories do not fit the ritualist
paradigm in the stricter sense insofar as they do not tell us how a particular
ritual that was performed at the site had been established. Similarly, divine
birth stories are aitiological insofar as they tell us how, where, and
sometimes why a god came into existence, but they do not usually explain
why specific rituals were performed for that god: the closest we typically
get is some indication of the god’s future talents or responsibilities, as when
the newly born Hermes stole cattle.78 Both birth stories and site-foundation
stories fit into a broader category that we might call “exaltation of the god

and his or her deeds,” which is common in ancient Greek poetry of many
types.
Second, even when we get something that looks like an aition in the
stricter sense of the term, how often can we link it to a specific ritual setting
in the manner that the ritualist approach demands? We have no evidence for
a formal festival celebrating Apollo’s annual return to Delphi, for which
Alcaeus’s paean might have been composed, for example.79 That paean,
moreover, portrays Apollo as arriving in Delphi at midsummer and brings
cicadas, swallows, and nightingales onstage to sing their distinctively
seasonal tunes—whereas Delphic tradition said that Apollo arrived each
year in early spring. If the paean were performed at a festival celebrating
Apollo’s return, this discrepancy would have jarred, as would have the
poem’s meter, which is more suited to a symposiastic setting.80 We know
that Pindar’s sixth paean was composed for performance at a Delphic
theoxenia (“feast for the gods”). Ian Rutherford is surely right that this
paean’s story of a disrupted sacrifice that led to Neoptolemus’s death was
thematically appropriate for such an occasion, but in no sense does it
provide an aition for the theoxenia.81
When poets do narrate what looks like an appropriate aition (that is, one
suited to the ritual at hand), they seem to do it out of sheer choice, with no
indication that they are following a tradition and no expectation that other
poets will follow their lead. For example, we might point to Pindar’s first
Olympian as an aitiologically satisfying poem. It narrates Pelops’s chariot
race against Oenomaus, which according to one strand of tradition led to the
establishment of the Olympic games, at which the victor honored by
Pindar’s poem (the Sicilian tyrant Hieron) had just won a crown in the
single-horse race. Thus, in this case, an Olympic aition was used to
celebrate an Olympic victor and, more generally, to glorify the Olympic
games themselves. But Bacchylides was also commissioned to celebrate
Hieron’s victory that year, and he did so with an ode that told the story of
Heracles’s descent to Hades—which has nothing to do with the Olympic
games as far as we can see. And Pindar himself, in two Olympian odes
written during that same year for the winner in the four-horse chariot race
(Theron, another Sicilian tyrant and a rival of Hieron), told a different story
associated with the foundation of the Olympic precinct in one of them—
how Heracles had provided shade for the first competitors by fetching olive
trees from Hyperborea—but in the other one offered complex ruminations

on the afterlife, which, again, had nothing to do with the foundation of the
Olympic games. The poet of the Homeric Hymn to Apollo, who tells the
story of the Delphic Oracle’s foundation, prominently expresses his
uncertainty about what he should sing about—he toys with the idea of
singing about Apollo’s love affairs instead of the Oracle’s foundation, for
example. Whatever the authorial intention behind that statement may have
been (I will return to that question in Chapter 7), it shows that the poet’s
audience could imagine hearing non-aitiological stories at the same event
where they would hear aitiological stories.82
These examples could be multiplied, but my point is probably clear by
now: even if poets sometimes chose to narrate aitia that were more or less
appropriate for the occasions on which their compositions would be
performed, they felt no obligation to do so or to stick to any other
aitiological guidelines that would have satisfied Jane Ellen Harrison.
Geoffrey Kirk was right then, even if for the wrong reasons: there were
more “significant” aitiological myths in ancient Greece than he was willing
to grant, but their significance did not necessarily mean that they would be
narrated in connection with rituals, as the ritualist paradigm expected.
How else, then, did Greek aitia get passed down, if not in poetry
composed for performance alongside the rituals? As in most societies, oral
transmission must have been important, not only for aitia but also for other
kinds of myths: the stories told by grandmothers and nursemaids that Plato
disparaged, the tales with which women wiled away time in front of their
looms, those told by men at symposia, and others that were passed along in
all kinds of conversational situations, such as when Socrates, strolling along
the river close to an altar to the wind god Boreas, told Phaedrus that they
were nearing the spot where the Boreas had once snatched away the
Athenian princess Oreithyia.83 Aitia were also repeated in more formal
situations. Isocrates mentioned the aition for the Proerosia as part of his
praise for Athens, for example,84 and the tragedians incorporated aitia as
well, if it suited their programmatic purposes. The ending of Euripides’s
Medea, for instance, tells of the foundation of Hera’s cult in Perachora, and
the ending of his Iphigenia in Tauris gives us aitia both for a ritual
bloodletting performed in the Attic demos of Halae Araphanides and for the
practice of dedicating the clothes of women who died in childbirth to
Iphigenia at Brauron.85

And then there were what we would call the writers of history (even if
the ancients would not have divided “history” from “myth” in the same
hard and fast way that we would). Starting by at least the fifth century, a
great number of myths, including aitia, were incorporated into historians’
chronicles. From Hellanicus, we get the story of how the court of the
Areopagus was established, for example. From Androtion, we get the aition
for the Buphonia.86 Thucydides mentions the origin of the Synoikia during
his discussion of how Theseus had reorganized Athens. The Hellenistic
poets, particularly those who, like Callimachus, worked at the Alexandrian
library where the historians’ writings were collected, found in these aitia
ample material for a new form of learnéd poetry. Callimachus, in his Aitia,
an elegiac poem in four books, narratively traversed a wide swath of Greek
cities, explaining the origins of their more puzzling rituals along the way:
why, for example, is it customary for maidens in Elis to be visited before
their marriages by armed warriors? The practice looks back, Callimachus
tells us, to an occasion when Heracles and his soldiers sired new children
upon the Elean women, whose husbands they had just killed.87
All of this aitiological material was passed further along in each
generation by other historians, poets, travel writers, mythographers,
scholiasts, and lexicographers. In other words, there were numerous ways in
which cultic aitia could be transmitted in ancient Greece that lay outside of
performance during the cults themselves, and there are no indications that
performing aitia in close association with rituals was obligatory, or even
standard. The transmission of Greek aitia in no way depended on their
narration during rituals, and the successful performance of Greek rituals in
no way depended on the recitation of their aitia in the same way as, for
example, the successful performance of the Sumerian Akitu festival
depended upon the annual recitation of the Enuma Elish, or as the
successful performance of the Catholic mass depends on the priest reciting
the Words of Institution, which repeat what Christ said at the Last Supper.
The Greeks just didn’t understand the relationship between myths and
rituals to be like that which the ritualists proposed two and a half millennia
later.
In fact, the Greeks were remarkably silent about this relationship. To the
best of my knowledge, no ancient author tells us why myths are narrated in
connection with rituals or what types of myths they should be. The closest
we come to an answer to the first question are the statements that we find at

the end of some hymns and dithyrambs when a poet hopes that his narration
has pleased the god. Or in other words, the professed reason that myths
were narrated at festivals was to entertain and honor the gods whose
festivals they were. Of course, we can be sure that human audiences were
on the poets’ minds as well: entertaining them was the way to accrue fame
and commissions.
As for the other question—what types of myths were suitable—the
closest we come to an answer is found in Aristotle’s Poetics. He doesn’t
really tell us which types of myths were appropriate, however, in the sense
of what relationship, if any, they should have with the festival at which they
were performed or with any other outside factor. Rather, he tells us that
their plots should include reversals of fortune, changes from ignorance to
knowledge, and suffering—all of which should arouse strong pity and fear
in the audience. He also tells us how the narrator should present these
elements: preferably, as arising from unexpected events that will surprise
the audience. He notes, as well, that tales of certain old mythic families (the
House of Atreus, the House of Laius, and so on) are best suited to fulfilling
these criteria.
Aristotle, in other words, stresses the idea that the myths narrated by
poets should engage the audience, emotionally and cognitively, as deeply as
possible, and should be about well-known, favorite characters. These may
seem like desiderata too obvious to need mentioning, but the point I am
trying to make is just that: rather than looking for other explanations of why
the Greeks narrated myths in connection with festivals, we need to accept
that the Greeks understood their myths, first and foremost, as a means of
entertaining both themselves and the gods. With this in mind, we can step
back and ask ourselves what other effects such engaging narrations might
have had—on their mortal audiences, at least. Chapters 3–5 will do just
that.

3
Narrating Myths
IN CHAPTER 2 we saw how, under the long shadow of Sir James Frazer, the
study of Greek myths has tended to start from the assumption that a myth
must be extracted from the narrative vehicles that conveyed it before a
scholar can do anything with the myth qua myth. For Frazer, this
assumption grew naturally out of his zeal for comparative work: until he
had set aside what he perceived as the ornaments in which an Aeschylus or
a Homer had draped a myth, he could not demonstrate how the myth was
similar to other myths that he had gathered from the four corners of the
globe and how they were all, as he believed, derived from a single original
myth. Although many of Frazer’s other ideas have been discarded by later
generations, his tendency to essentialize myths has survived—not only
amongst those who, like Frazer, focus on the connections between myths
and rituals, but also by those who embrace structuralist or psychoanalytical
approaches to myth. If myths are to reveal the lost rationales behind
mysterious rituals, the universal concerns of the human psyche or the basic
structures of the human mind, it has seemed necessary first to pare away
their surface details, however enchanting those details might be (indeed, the
more enchanting the details, the more they threaten to obscure the “real”
myth underneath).
I do not reject these approaches completely—each of them has enhanced
our understanding of what myths can do in significant ways. But I do want
to suggest that, having fallen into the habit of excising Greek myths from
their narratives, most scholars have long overlooked one of the most salient
and significant features of mythic narratives: their ability to engage their
audiences emotionally and cognitively. I argue that this habit not only has
prevented us from understanding some of the most important reasons that
myths were able to help create and sustain ancient Greek beliefs in the
gods, the heroes and the divine world more generally (and, thus, one of the

reasons that it was appropriate to narrate myths in connection with rituals)
but also has excluded Greek myths from a larger contemporary dialogue on
the power of narratives. Wendy Doniger, an insightfully comparative
scholar of myth, once described the stories told by the Greeks as
“mythological zombies”—that is, she felt that the Greeks had killed
anything that was really alive in the myths in the course of polishing them
into the beautiful literary and visual forms that still survive today. This isn’t
true, as I will show, but I can understand how Doniger came to such a
conclusion: scholars of Greek myths have done almost nothing along the
lines of what Doniger has done so well for Hindu myths: demonstrate the
vibrant way in which they permeated their audiences’ daily experiences,
thereby keeping the characters and their stories vigorously alive.1
In this chapter, I’ll try to capture some sense of those ancient Greek
experiences. I’ll begin by laying down some context that will help us
understand how Greek myths were different from those of many other
cultures. First, I’ll look at the historiola, a type of myth that was very
common in the ancient Mediterranean. Historiolae are performative
utterances in the sense that what happens in them is expected to start
happening in the real world as well, if the historiola is properly narrated.
Next, I’ll take up the fact that the Greeks, in contrast to their neighbors, do
not seem to have adopted the habit of using historiolae. This suggests that
the idea of narrating a myth in order to cause something directly to happen
in the real world did not align well with what the Greeks thought myths
were and what myths could do. I’ll also show that in the single, seemingly
experimental historiola that they did create, the Greeks shifted away from
the genre’s typical reliance on performative utterance towards what George
Lakoff and Mark Johnson have called the “conceptual metaphor”—or, in
other words, that the Greeks expected that the myth comprising this
historiola would affect the real world through the figurative force of its
language. I’ll suggest that such an expectation is characteristic of Greek
mythic narration more generally: it relied on figurative language and other
rhetorical techniques such as deixis to draw connections between events in a
story and events in the world outside the story. This quality contributed to
what Claude Calame has called the “pragmatic effect” of Greek mythic
narration—that is, the potential of mythic narratives to bring about changes
in the outlook and behavior of its audience members. Such changes were
less directly tied to the specific actions of the myth and less focused on a

specific goal than were the changes expected when a historiola was
narrated, but I suggest that, by the same token, Greek mythic narratives
carried a greater potential for more broadly making the audience feel as if
their world and the world portrayed by the myth were one and the same and
as if they lived on a continuum with the events that the myths narrated.
After this preliminary discussion of how Greek narratives merged the
world of the myths with the world of their audiences, I’ll move on to the
topic that occupies the rest of this chapter: how recent work on the specific
effects that engaging narratives can have on emotion and cognition will
help us better to understand how the narration of myths in ancient Greece
conditioned listeners to open themselves to realities that could not be
perceived by normal means—to the idea that gods and heroes were present
amongst them, listening, watching and affecting the course of mortal lives.
I’ll begin by expanding upon Joshua Landy’s study of how narratives can
change not only what audience members think, but more importantly the
ways in which they think, priming their minds to accept as truths what
might otherwise be dismissed as fiction. Then I’ll look at Tanya
Luhrmann’s recent work on the ways in which narratives are used in
contemporary American religion to prepare worshippers to experience
divine presence more regularly and more intensively than they otherwise
would, thus creating and sustaining belief. More generally, I’ll also review
work by social scientists showing that there is a widespread human
propensity to develop emotionally and cognitively rich relationships with
characters encountered in fictional venues. Throughout these discussions,
I’ll pause to reflect on how the insights gained from these studies can be
applied to the narration of myths in Greece.
Following that, I’ll borrow from recent work by media scholars and look
at one particular characteristic of Greek myths that conditioned audience
members to believe in the gods and heroes who populated them: the fact
that Greek myths were narrated in an episodic fashion. Finally, I’ll look at
the styles of narration used in myths and make some suggestions about how
they would have contributed to belief in the gods and heroes, as well. I’ll
take my general inspiration here from the recent turn in narrative theory
towards rhetorical poetics—the attention given to how narrators selectively
use the resources of storytelling to affect their audiences in particular ways.
I will have two specific models, however. The first is the informal stories
that people of our own times tell one another about encounters with the

supernatural. The second is the literary ghost story as it was developed
beginning at the turn of the last century. In applying what we can learn from
these two models to my ancient Greek materials, I will draw on what
contemporary folklorists have discovered about the ways in which such
stories help to create and sustain belief in the phenomena that the stories
describe.
Historiolae
Historiolae are brief myths that are recited in order to solve an immediate
problem—most commonly, illness.2 Typically, at least some of the
characters in a historiola are gods, heroes or other supernatural agents.
These characters have names that are familiar to their audiences; in and of
themselves, therefore, such names already evoke histories, actions, and
personal characteristics that are relevant for the work that a historiola will
do. Thus, for example, Isis, a goddess whom myths portrayed as healing her
son Horus, is a central character in many historiolae that are intended to
heal humans.3 The characters in historiolae, moreover, inhabit a realm (I
will call it the “mythic realm”) that is rife with powers greater than those
available in the “quotidian realm”—the realm of everyday people and their
everyday abilities. Therefore, although historiolae are narratively compact,
they are potentially tales of great power.
The actions that are performed by the supernatural agents in a historiola
are understood to affect the world directly, at the very moment that the
words of the historiola are recited. That is, narrating a historiola constitutes
a performative utterance. The following is an example from an Egyptian
papyrus dating to about 1325 BCE. A mother whose child has a fever is told
to say:
Isis came out of the spinning house [at the hour] when she loosened
her thread. “Come, my sister Nephthys! See, my deafness has
overtaken me! My thread has entangled me! Show me my way that I
may do what I know [how to do], so that I may extinguish him with
my milk, with the salutary liquids from within my breasts. It will be
applied to your body, Horus, so that your vessels become sound. I will
make the fire recede that has attacked you!”

While the mother recites these words, she applies her own milk to her child
just as Isis applied hers to Horus. The child’s fever is expected to break, just
as Horus’s fever broke.4
But historiolae are not performative utterances in quite the same sense
that the words “I now pronounce you husband and wife” or “we find the
defendant guilty as charged” are performative utterances. The main point of
a historiola, rather, is to establish a paradigm to which the situation
confronted in the quotidian realm is expected to adjust itself: when the
historiola about Isis’s milk is recited, the mortal child is expected to follow
the paradigm that the historiola established and recover just as Horus
recovered.5 Historiolae are performative utterances that work by means of
persuasive analogy, in other words. To take another example, if a baby has
a headache, then its mother might invoke the paradigm of “banished
headache” by telling of how Christ pushed the Evil Eye off a rock to stop it
from giving headaches to another baby, thus “persuading” her own child’s
headache to go away as well.6 The relationship between the paradigm and
its application in all of these historiolae can be represented this way:
A:B : a:b
:
Isis’s actions : Horus’s fever : narration of Isis’s actions : present child’s
: fever
Christ’s actions : mythical : narration of Christ’s actions : present
headache : headache
mythic power : mythic crisis : narration of mythic power : present crisis
:
Historiolae built on this analogical principle were common in most ancient

Mediterranean cultures, and in many other cultures as well.7 Frequently, the
effectiveness of the historiola itself is enhanced by narrating it under
various “felicity conditions”—that is, it is narrated at certain times of day
when its power will be greatest, it is narrated by a certain type of person
(midwives, healers, ritual practitioners), or its narration is accompanied by
the application of a salve, an amulet or some other material (such as breast
milk), whose discovery, invention or use by a supernatural agent is
sometimes described within the historiola.8
Historiolae from many different cultures share another feature, too: an
emphasis on the place where the action occurs. Here are just the first lines

of two:
Isis comes down from the mountain at midday, in summer, the dusty
maiden; her eyes are full of tears and her heart is full of sighs.9
Saint Lazarus and Our Lord were going for a walk in our town.10
In the first example, Isis is crying at the base of a mountain, and in the
second, Lazarus and Christ are walking in “our town.” My earlier example
was similarly set in a spinning house where Isis and Nephthys were
working, and many other historiolae are set in specific places as well—
places that are virtually always terrestrial in nature and sometimes even
places in the speaker’s immediate environment. Such emplacement of the
paradigm narrated by the historiola subtly reiterates one of the genre’s
underlying principles: what is described as happening in the mythic realm
will also happen within our own, quotidian realm. Indeed, the terms
“mythic realm” and “quotidian realm”—heuristically useful though they
may be for discussing what historiolae do—have the unfortunate side effect
of suggesting that the actions narrated by the myths occur in places that are
completely separate from our own world.11 This is contradicted not only by
the insistently familiar settings of historiolae but also by the ontological
underpinnings of the cultures that produce them: gods, heroes, saints and
other supernatural entities may be understood to make their homes in a
place that is geographically distant (higher or lower than where humans
make their homes, beyond the western ocean, and so on), yet that place is
not understood to be cut off from our places; sacrificial smoke can rise from
our altars to their nostrils, prayers uttered by our mouths can reach their
ears, and when they choose to visit us for whatever reason, they find it quite
easy to do so. Setting the paradigmatic action of a historiola within familiar,
everyday geography in effect conjoins the mythic and quotidian realms;
each time the historiola is spoken, the distance between the two is
narratively erased so as to momentarily create a shared realm and thus to
bring the paradigm into immediate contiguity with the situation that is
meant to replicate it.
The Getty Historiola

Remarkably, the ancient Greeks have left us almost no historiolae of their
own design, and we do not find mention of anything like them in Greek
authors, in spite of the fact that we hear a lot about other ritual behaviors
that make use of persuasive analogy and performative utterances—for
example, katadesmoi (“curse tablets”). The Greeks used the basic concepts
that underlay historiolae, in other words, and were undoubtedly exposed to
historiolae during their interaction with other Mediterranean cultures, but
they don’t seem to have themselves adopted the idea that by narrating a
myth they could directly cause something to happen in the real world.12
There is an exception to that statement, however, from which we can
begin to learn more about the Greek view of what myths can do. It is the so-
called Getty Hexameters, which were inscribed on a lead tablet dating to the
fourth century BCE (the tablet is now owned by the J. Paul Getty Museum
in Malibu). The tablet was found at the site of the ancient city of Selinus, on
Sicily—an island through which we suspect that ritual techniques from
foreign lands often entered the Greek world. Both the Getty Hexameters
and several abbreviated copies of them (three from Sicily and five from
other places) were derived from an older archetype that probably goes back
to the late sixth century BCE.13
The historiola on the Getty tablet is set within a double frame.14 The
innermost of these presents the god Paean (well-known for healing illnesses
and for saving people in other dire circumstances) as the original narrator of
the historiola; in the outer one, an unidentified speaker tells us that the
“sacred words” that Paean is about to speak (that is, the historiola) will
guarantee protection from a broad variety of ills on land and at sea. Later in
the text, after the historiola has been concluded, additional lines, apparently
narrated by the unidentified speaker, specify that these ills include war,
sickness, and dangerous animals. What the Getty historiola offers, then, is
wholesale safety and well-being (lines 4–6, 23–32, 46–50).
The latter part of the historiola itself survives only in fragments, but the
first part, which seems to constitute the heart of the matter, is almost intact:
As down the shady mountains in a dark-and-glittering land,

a child leads out of Persephone’s garden, by necessity for milking,
that four-footed holy attendant of Demeter,
a she-goat with an untiring stream of rich milk

laden; and she follows, trusting in goddesses, bright
with torches …15
As I’ve shown elsewhere, the story here is borrowed from Egypt, where
we find many myths about both Isis and also the goddess Hathor using
either their own milk or the milk of a gazelle (a bovid closely related to the
goat and similar in appearance) to do remarkable things—including curing
Horus of fever, as in the example of a historiola that I provided in the
preceding section, and curing Horus of blindness.16 So, on the one hand, we
could say that the Getty historiola is Greek in language, but otherwise
Egyptian. On the other hand, from earliest times, milk had been
independently associated in Greek culture with blessedness and security:
“falling” or “jumping” into milk meant to arrive at a state of the highest
possible happiness, rivers of milk were common features of paradisiacal
times or places and milk was an important healing agent in a wide range of
cures discussed by Greek and Roman medical writers.17 Therefore, the
agent of efficacy that is central to the Getty historiola would have made the
story instantly resonant for Greek audiences, as would the presence of the
Greek goddesses Demeter and Persephone.
But there is an important quality that sets the Getty historiola apart from
those that we looked at in the preceding section—a quality that, I suggest,
was introduced by the Greeks themselves when they adapted what was a
foreign technique to their own use. Remember that the analogical
relationship in historiolae is typically like this:
A:B : a:b
:
mythic power : mythic : narration of mythic power : present
crisis : crisis
In the Getty text, instead, we have something like what the linguists George
Lakoff and Mark Johnson have called a “conceptual metaphor.” In a
conceptual metaphor, an abstract concept, such as “love,” is compared to a
concrete image, such as “journey,” producing, in this case, the familiar
expression “love is a journey,” as well as other expressions that draw on the
journey metaphor, such as “we’ve come too far together to give up on our
marriage now.”18 As in the case of all metaphors, in conceptual metaphors

the two things that are compared must share one or more salient
characteristics but must stop short of being identical or even close to
identical—otherwise, the descriptive power of the metaphor would be nil (it
is hardly metaphorical to compare mandarin oranges to tangerines, for
example). In the Getty text, we have an abstract concept (“wholesale
protection and prosperity”) and a concrete image (“abundant milk”) that
together constitute an implicit conceptual metaphor that we might express
as “prosperity is milk”—a metaphor that was already familiar to the text’s
Greek audience. Situating the power of a historiola within a metaphor
(conceptual or otherwise) rather than within an analogical paradigm is
unparalleled as far as I know, which makes a certain sense: the aim of a
traditional historiola, after all, is to cause something in the quotidian realm
to pattern itself after something in the mythic realm not in only one or two
salient ways, but rather as closely as possible.
The Getty text complicates things yet further. The outermost frame of the
historiola includes lines that describe the “felicity conditions” under which
it should be narrated—that is, the ritual that will help to activate its power:
Whoever hides in his house of stone the notable letters

of these sacred words [of the historiola] inscribed on tin,
As many things as the broad Earth nourishes shall not harm him
nor as many things as much-groaning Amphitrite rears in the sea.
The felicity conditions for using traditional historiolae frequently include

manipulating materials that are cognate to those manipulated in the
historiola itself: the mortal mother’s milk in our first example, which is
cognate to Isis’s milk, is an instance of this. In the Getty example, however,
the goat milk described by the historiola does not seem to have any cognate
in the quotidian realm. Instead, the story of the goat’s milk becomes
efficacious when it is inscribed on tin and “hidden in a house of stone”
(whatever “house of stone” means).19 In other words, the Getty historiola
operates on a strictly figurative plane; it requires the manipulation of a
semantic substitute for the efficacious agent rather than the manipulation of
a quotidian cognate. Perhaps such a change was inevitable, given another
way in which the Getty text runs counter to traditional historiolae. Namely,
when compared to the goals of other historiolae—to cure a fever, a scorpion
sting, a headache or another immediate problem—the wholesale protection

from a broad variety of ills that the Getty text promises constitutes an
enormous, rather vague and ongoing aim. Even if one wanted to apply real
goat’s milk to something in the quotidian realm as the Getty historiola was
being recited, what would that something be?
On the single occasion when we catch the Greeks experimenting with a
historiola, then, they change the basic rules by which the technique worked.
Instead of adopting the idea that the narration of a myth could directly cause
something similar to happen in the everyday world, they employ a myth as
a figurative discourse: the provision of abundant milk described in the Getty
text was a metaphor for the provision of protection and prosperity.
Leaving aside the Getty Hexameters, the closest that Greek mythic
narration ever comes to a mode of operation like that which we see in
historiolae is in the pars epica of certain Greek prayers. There, a god might
be reminded of what he or she did in the past in order to persuade him or
her to do something similar in the present. Yet the words of such prayers are
not understood to be directly efficacious in the performative sense; we are
still far away from the ideology that underlies historiolae.
How to Do Things with Metaphors

This observation shouldn’t come as a complete surprise; it aligns well with
the norms of Greek myth telling as we know them already in the Archaic
and Classical periods, when metaphor and other forms of figuration played
important roles in linking a myth to the occasion or purpose for which it
was performed. The story of Persephone’s annual return from the world of
the dead, for example, when narrated in connection with the Eleusinian
mysteries, was not meant to suggest that initiates into the mysteries would
similarly return to the upper world for a portion of each year after they had
died, but rather reminded them that initiation ensured them happier
existences in the Underworld, once they had gotten there. Persephone’s
experiences were a metaphor for those of the initiates, in other words; the
two shared the salient characteristic of being partial triumphs over death but
differed insofar as Persephone annually returned to the world of the living
and the dead initiates did not. When narrated in connection with the
Thesmophoria, the same story metaphorically expressed the celebrants’
hopes that crops would once again rise from the dark earth into which seeds
were cast. The two situations shared the salient characteristic of anticipating
the annual return of something desirable but differed insofar as Persephone

returned each year in her own right and the crops “returned” only in the
sense that their seeds generated new plants to replace those that were dead
and gone (an idea that, in turn, served as a metaphor for the Thesmophoria’s
other focus, the successful conception and birth of new children). The fact
that some stories, such as this one, could serve as meaningful
accompaniments for two different festivals with different primary goals
underscores myth’s metaphorical nature. Had the relationship between the
myth and the two rituals I just described been one of straightforward
analogy, such double service would not have worked very well.20
Epinician odes, and especially those composed by Pindar, relied heavily
on the metaphorical uses of myths as well. It was unusual for an ode to
narrate a myth that was directly connected to the victor or the athletic event
at which he had won. Rather, the myths narrated in epinicians typically
evoked the nature of the victors’ accomplishments and the effects that their
attendant glory would have upon the victors, their families and their cities
by telling stories about heroes and gods. A victor could not hope to gain
actual immortality, but victory—especially when celebrated publicly by a
poem in which a hero’s or god’s story was embedded—could bring an
everlasting fame that was as close to immortality as any ordinary human
could get.21 Pindar’s fourth Isthmian, for example, encourages us to
compare the young victor, Melissus, to Heracles (Pindar goes so far as to
point out that both were small in build; lines 49–53), but leaves it up to us
to choose exactly how we read Heracles’s story into that of Melissus. When
skillfully done, the narration of myth plays a crucial role in bringing about
the social benefits that athletic victory could garner, but it does so without
drawing explicit, one-to-one equations between the circumstances of the
myth and the victor’s own circumstances. These myths are not historiolae,
in other words, but rather extended metaphors that invite us to consider the
salient characteristics shared by two individuals or situations.
Epinicians bring us to Claude Calame’s observation that mythic
narratives could have a pragmatic effect upon their audiences, for his
explorations of this idea frequently draw on Pindar’s epinicians for their
examples. (In fact, Calame suggests that because they were performed in
ritualized community settings, epinicians and other forms of melic poetry—
e.g., paeans, dithyrambs, maiden songs, wedding songs, and funeral laments
—provided particularly good contexts for myths that were meant to have a
pragmatic effect).22 In the case of an epinician ode, this change usually

meant elevating the victor’s status within his community, but it might also
include, for instance, persuading a king to let an exile come home.23 In the
case of a tragedy, the change might mean altering the audience’s opinions
about a war, politics, or other current events. In other situations, the change
might affect just a single person: Achilles’s narration of the story of Niobe
in Iliad 24 causes the grieving Priam to finally break his fast.24
Calame’s realization that narrating myths can have a pragmatic effect is
important, as is his observation that this effect is bound to be more powerful
when a myth is narrated within a ritualized, communal setting. But two
issues remain. First, exactly how does the narration of a myth bring about
this pragmatic effect? In the case of historiolae, explicit, one-to-one
analogies are expected to bring the quotidian world sharply into line with
the mythic world; a historiola is essentially a speech act that relies on the
principle of performative utterance. In the case of the Getty Hexameters, the
myth works metaphorically, building on milk’s salient characteristic of
being able to cure, protect, and bring blessings to everything it touched. Yet
there is still a strong element of ritual efficacy present in the Getty
Hexameters, as there is in traditional historiolae: the words of the myth
have to be carved on a particular type of metal and then deposited in a
particular place in a particular way for the words of the myth to take their
metaphorical effect. This is not the mode in which we see the majority of
other Greek mythic narratives being composed and narrated.25 What
happens in those cases, then? What qualities make them potent enough to
have pragmatic effects?
Second, Calame’s pragmatic model of mythic narration works best in
cases where the similarities between the two things being compared are
fairly obvious, even to us—modern readers who are far removed from the
social and cultural contexts in which the myths were first presented. We can
guess from the start that Euripides’s Erechtheus, a play narrating the tale of
a war waged during Athens’s earliest days, was likely to provoke thought
amongst its audience members about the Peloponnesian War in which
Athens was engaged at the time of its production. The examples I offered of
using Persephone’s experiences as a metaphorical expression of hopes for a
better afterlife or abundant crops are less straightforward than the linkage
between the Erechtheus and the Peloponnesian War, but they are not

obscure, and the idea that athletic victors were understood as the heroes of
their day makes intrinsic sense.26
I don’t mean to diminish the importance of Calame’s insight that mythic
narratives could have pragmatic effects (or to ignore his close readings of
some mythic narratives whose metaphorical frameworks are less readily
apparent), but I do want to raise the question of how we are supposed to
understand the many mythic narratives for which we can discern no
pragmatic goal other than the general one of praising a god and thus
winning his or her favor (I think here of the Homeric Hymn to Aphrodite,
for example, and of one of Bacchylides’s dithyrambs, which tells a gripping
story of Theseus’s plunge to the bottom of the sea and then closes with the
hope that Apollo has enjoyed the story and will reward those who
performed it).27 It’s not that entertaining gods (and mortals) was
unimportant, but in theory there were other ways of doing that than by
narrating myths. We know that it was possible to use a non-mythic story as
the kernel of a tragedy, for instance: Aeschylus’s Persians proves that,
although the fact that it floats almost alone amongst a vast ocean of
tragedies that do use mythic stories suggests that such subject matter was an
exception. Indeed, the comic poet Antiphanes complained that tragic poets
had a big advantage over comic poets: the audience already knows the basic
plot of a tragedy the moment the central character’s name is mentioned, he
charges, because tragedies are always based on well-known myths.28 All of
the epics that remain to us (which were re-performed at festivals for
centuries after their composition) take myths as their subject matter, too.
Shorter hymns and melic compositions sometimes abstain from narrating
anything at all, offering only praise and entreaty, yet by and large, when
they do narrate something, it’s a myth. In sum, narratives performed in
honor of the gods and heroes almost always took myths as their subject
matter—even if those myths had little or nothing to do with the specific
gods, heroes or rituals with which their performances were connected.
Again, we must wonder, how and why did the habit of narrating myths at
festivals become so engrained that a wide variety of subjects were judged
appropriate, as long as they were myths? What desires did narrating myths
fulfill? What purposes (conscious or unconscious) did it serve?
For the moment, I want to put this bigger issue on hold (we will come
back to it later in this chapter) and return briefly to the first question I
raised, which will help to prepare the way: how, exactly, does the narration

of myth sustain a metaphorical connection between the mythic and
quotidian worlds? For Calame and others who have worked on the topic,
the metaphorical connection between (for example) a victor and a hero in
an epinician myth is activated by a number of consciously deployed
narrative techniques that bring the two worlds together. The place where the
myth is being narrated or performed may be mentioned in the narrative
itself—perhaps the events of the myth are even set in the very place where
the narrative is delivered. The action of Euripides’s Erechtheus, as I
mentioned earlier, takes place on the Athenian Acropolis that loomed above
the theater in which it was first performed. One of the goals towards which
the Argonauts strove, in Pindar’s fourth Pythian, was the establishment of
Cyrene, where the poem itself was presented many generations after the
actions that the myth describes.29 A mythic narrative might also be
delivered in, and perhaps indeed composed for, a setting that visually cued
its events: Pindar’s Isthmian 4, which tells the story of Heracles and
Antaeus, was performed in front of a Theban temple of Heracles that
included a pedimental relief showing the two characters wrestling.30 Verbal,
and perhaps physical, gestures on the part of the poet (deixis ad oculos or
deixis ad phantasm or some blend of the two) could bring the two worlds
together, as well.31 Lines 362–363 of the Homeric Hymn to Apollo, for
example, suggest that its performer, standing in Delphi, directed the
audience’s eyes to the very spot where Apollo had once left the body of
Python to rot, as the performer repeated what Apollo had said on that
occasion—“Now rot here!” Such an emphatic statement suggested, even if
only for a split second, that the snake was rotting there, still. Apostrophe
worked similarly: when, in Olympian 1, Pindar directly addresses Pelops in
the second person, it is as if the poet, and also we, his audience, are
suddenly in the hero’s presence.32
The two worlds may also be presented as sharing the same values,
limitations, and advantages: Calame suggests, for example, that in both the
(mythical) Athens of Erechtheus’s day and the Athens in which the
Erechtheus was performed, citizen culture was construed as being rooted in
maternal nature.33 The poet may express such shared conditions by applying
to each the same metaphorical figures (what Calame calls isotopies,
adopting the term from A. J. Greimas). In Pythian 4, the isotopies of
matrimonial union and agricultural cultivation are used to explore both the

experiences of the Argonauts and those of Cyrene’s later colonists.34 In
Isthmian 4, Heracles and the victor’s family are described as undertaking
travels (either real or metaphorical), to “the west,” a place that Pindar
associates, disquietingly, with both darkness and achievement.35 Sometimes,
a narrative may leap back and forth across generations so rapidly and
repeatedly as to conflate them, blurring any chronological distinction
between the mythic and the contemporary worlds. An extreme case is
Pythian 4, which jumps back and forth amongst seventeen generations of
the Battiad dynasty and its ancestors that stretch from the Argonaut
Euphemus to his descendant Arcesilas, the victor whom Pindar celebrates.36
All of the techniques I have mentioned so far help to draw together a
specific figure or event from the mythic realm and a specific figure or event
in the quotidian world. Kennings (circumlocutions whose sense depends on
the listeners’ knowledge of myths) work to momentarily immerse listeners
in the mythic world more generally. When Pindar says that a victor won a
crown at “Nisus’s hill with its lovely glens” (i.e., in the city of Megara), we
perceive the victor, if only for a moment, against the backdrop of the
remarkable story of King Nisus, his treacherous daughter, and his fatal lock
of purple hair.37 To a certain degree, then, kennings work like historiolae
insofar as they use narrative elements, and particularly names, to
temporarily collapse the distance between the mythic and the everyday
realm. But there is an important difference. In contrast to historiolae,
neither kennings nor any of the other techniques that I have just surveyed
are performative utterances—that is, they are not expected to cause to
happen in the everyday realm what once had happened in the mythic realm.
(Indeed, one probably would not wish to cause to happen again what had
happened to Nisus.)38 By using kennings and other techniques, the poet is
able to merge worlds without ever promising—or threatening—to make
anything in the quotidian world align precisely with a particular paradigm
that has been set in the mythic world. In fact, all of the techniques that I’ve
been discussing for the past few paragraphs might be said to help merge the
two worlds not only spatially (as in the case of the historiolae, where, for
example, Saint Lazarus and Our Lord take a walk in our town) but also
temporally: the Acropolis under whose shadow a tragedy is presented
morphs into the Acropolis where Poseidon once struck the earth with his
trident and Medea once offered Theseus a poisoned cup of wine. In this way

of thinking, there was no moment at which the mythic world decisively
changed into the world that we know today; the deeds described by the
myths existed on a continuum that flowed uninterruptedly into the time of
the listeners. A well-narrated Greek myth would leave those listeners
feeling not that they were repeating paradigmatic actions of the gods and
heroes that had been performed eons ago (as is the case with historiolae),
but rather that they were living amongst the gods and heroes, even if as
lesser partners.
Narrating Fictions
It did take some preparation to feel that way, however. I don’t mean only
the preparation of crafting a narrative that included deictic references,
complex chronologies, kennings and the other devices I have just reviewed,
if you were a poet, or the preparation of going to the festival where the
narrative was to be recited, if you were an audience member. That was
important, of course; situating yourself in a time and place given over to
worship and celebration of the gods and heroes, amongst other people who
were there for the same reason, was certainly conducive to putting you in
the right frame of mind.
What I’m interested in here, rather, and what will eventually bring us
back to the second issue that I raised earlier (why myths were so
persistently narrated at religious festivals), are the ways in which Greek
mythic narratives cumulatively prepared their audiences to feel as if they
were living amongst the gods and heroes. Recent work by scholars of
narratology, anthropology, social psychology and cognitive science has
shown that some types of fictional narratives, when they are well designed
and well executed, not only teach their audiences certain facts (in the case
of Hilary Mantel’s Wolf Hall, this might include details about life at court in
Cromwellian England) or certain values (in the case of The Wonderful
Wizard of Oz, this might include the idea that collaboration with friends will
get you further than you would have gotten alone) but also teach them new
modes of thinking or new ways of looking at the world, which they
subsequently apply, consciously or unconsciously, to other narratives they
consume and to situations in real life.
Joshua Landy has called such narratives “formative” fictions—to stress
the distinction he wants to make between their role in “forming” or “fine-
tuning” our mental capacities and habits and the emphasis that traditional

approaches to literary criticism have tended to put upon fiction’s
“informative” role (its role in conveying facts to its audience).39 Amongst
Landy’s test cases are the parables of Jesus, as recorded in the Gospel of
Mark. Landy suggests (as have scholars of early Christianity such as David
Brakke)40 that these were not meant to be interpreted in ways that decoded
their metaphors once and for all (the long history of varied interpretations
of most of the parables in itself argues against this idea), but rather to give
followers practice in reading the physical world metaphorically until,
eventually, they came to view everything that they experienced here and
now as nothing but a metaphor for the higher plane of existence to which all
Christians were expected to aspire. In other words, reading parables trains
people to engage in a form of abstract, figurative thought that enhances
faith in an idea that is central to Christianity.41 Another of Landy’s test cases
focuses jointly on the illusions of the stage magician Jean Eugène Robert-
Houdin and the works of the Symbolist poet Stéphane Mallarmé, each of
whom sought to re-enchant the secular world of nineteenth-century Europe
by training his audiences in the skills of detached credulity—a state of mind
in which distrust of what their senses perceived, and yet conviction that
those perceptions must be true, were able to coexist. By helping them hone
this capacity for “lucid self-delusion,” Robert-Houdin and Mallarmé taught
their audiences to generate fictions that made their lives more interesting
and restored a touch of the miraculous to a world that science had
threatened to disenchant completely.
Well-constructed narratives that are effectively delivered, then, can
change the way in which people decide what is real and unreal—and more
importantly, change the way in which people decide whether, and when, to
accept the evidence of their senses as the only guides to distinguishing
between the two, or rather to admit into consideration another model of the
world, in which parts of reality are understood to lie outside the reach of the
conventional senses. With regular exposure to such types of fiction, it
becomes easier and easier to slip into the new mode of thinking that those
fictions encourage. It may become an individual’s dominant mode of
thinking, as in the case of (idealized) Christian parabolic thinking, or an
alternative mode into which an individual slips when prompted to do so by
the right circumstances—by opening a book of poems, for instance, or
entering a theater where an illusionist is performing. In either case, the
effects are cumulative: as Mark’s Jesus said about the difference between

those who understood his parables and those who did not, “to him who has,
more will be given.” Or in other words, those whom fictions have begun to
train in a new mode of thinking will enter into what Landy calls a
“formative circle,” in which their proficiency in the new mode of thinking
continues to increase, making it easier and easier to enter into that mode of
thought again.42 Christopher Partridge, a scholar of contemporary religions,
and Michael Barkun, a political scientist who studies contemporary
apocalyptic Christianity, have made similar observations about how the
products of popular culture (films, television shows, novels, music) give
people new perspectives on theological and metaphysical issues and
“provide resources for the construction of religious and paranormal
worldviews”—which then fuel the further production of popular narratives
on these topics, eventually leading to what Barkun calls “fact-fiction
reversals” (a situation in which something understood by its creator to be
fiction is eventually moved by its audience into the realm of fact). Jeff
Kripal similarly has studied the important contributions that are made to
contemporary Western belief systems by science fiction, superhero comics
and other popular representations of the paranormal. He stresses the crucial
role that “fiction” plays: “It is almost as if the left brain will not let the right
brain speak … so the right brain turns to image and story to say what it has
to say (without saying it).”43
Fictional narratives—as opposed to ostensibly fact-based narratives such
as news reports or documentary films—offer an ideal venue for formative
training and other types of cognitive adjustment such as Partridge, Barkun,
and Kripal describe, for several reasons. Whether a fictional work is the
creation of single mind (a novel, a poem, a stage illusion) or drawn from
common cultural property (a fable, an urban legend), it can be tailored
almost infinitely to suit the tasks that its narrator intends it to serve.44 And
when it is tailored and narrated skillfully, a fictional work can present
something that we have never experienced before so convincingly that our
minds accept it—thus enlarging our sense of what might be possible.45
Therein lies, at least partially, the popular success of any number of fantasy
authors—J. R. R. Tolkien, Ursula K. LeGuin, Madeleine L’Engle, Neil
Gaiman, and so on. Or to switch to a darker genre and some startlingly
concrete results, The Exorcist enlarged our sense of what might be possible
so effectively as not only to cause some readers to throw the book away

unfinished and some moviegoers to faint and vomit, but also to “stimulat[e]
an unprecedented demand for Catholic exorcisms” for decades to come.46
William Peter Blatty’s novel and William Friedkin’s film were able to make
even doubting minds believe in demonic possession at least temporarily,
and in some cases much longer. (As a friend of mine once said about The
Exorcist, “I don’t believe in possession, but if I did, I know that’s what it’d
be like.”)
Moreover, by definition, as audience members we leave our day-to-day
expectations at the door when we enter a narrative and tacitly agree to be
pulled along by the story without continually comparing it to what happens
in our everyday lives. This means that we temporarily dismantle the
cognitive barriers that are patrolled by mundane logic, the five senses, and
our customary value systems in order to open ourselves up to scenarios that
may differ significantly from what we encounter outside the narrative.
Consumers of vampire fiction enter a universe where some of the most
basic postulates of normal life (such as death being final) are no longer
valid. Less alarmingly (perhaps), as twenty-first-century readers of Jane
Austen, we must immerse ourselves in a world that lacks telephones and the
right of females to inherit. Still other fictions depart from our normal
experiences in only the mildest of ways—most viewers of American TV
sitcoms can empathize with their main characters by doing little more than
imagining that they live in a different part of the country, have a different
family structure or are employed in a different profession. Yet even these
minor departures from our personal reality require us to drop our cognitive
guard and open ourselves to the realm of what a good narrator has made
plausible rather than sticking to what is provable.47 And from that point on,
we become putty in the hands of that narrator, until we put down the book,
leave the theater, or turn off the TV.
Narrating God
But what if you can’t turn off the TV because you don’t even think it’s
there? What if the materials that train the mind to think in certain ways and
to accept alternative realities are not understood by the audience—and
perhaps not by the authors, either—to be fictions, at least in the usual sense
of that word?
Tanya Luhrmann spent four years with members of the Vineyard Church
(an Evangelical church with congregations throughout the United States),

observing how they train themselves to feel the presence of God in their
daily lives—how a form of cognitive self-training enables them to cultivate
God as a daily reality that is as vivid as their friends and family.48 One of
the first major hurdles they face in doing so is overriding a basic feature of
human psychology in the modern West: the assumption that our minds are
private. Once congregants come to accept that the barriers between their
individual minds and God are porous (once they have adopted what
Luhrmann calls a “participatory theory of mind”), they practice specific
techniques to enhance their ability to experience the presence of God.
One of these techniques is kataphatic prayer, which is also called
“Ignatian” prayer, after the sixteenth-century saint who developed it.
Kataphatic prayer encourages worshippers to fill their minds with vivid
images, symbols, and ideas while praying. By doing this, they gradually
condition their minds to perceive things that are not present to the ordinary
senses. During prayer, they feel that they are able to experience events
described in the Bible with remarkable detail—down to the “dust motes in
the sunlight when Mary heard the news from Gabriel.” A study conducted
by Luhrmann, in which non–church members were trained in kataphatic
practices for thirty minutes a day during a four-week period, showed that,
even over this relatively short period of time, participants could
significantly enhance their minds’ abilities to generate sensory experiences
that were similar to their memories of real events.49 Another technique
practiced by Vineyard congregants involves treating God like a flesh-and-
blood friend—setting out a coffee cup for God once a week, for example, or
an extra dinner plate and chair. Behaving as if God is present (although
Vineyard members would not use the phrase “as if”) encourages
congregants to view God not as a “packet of rules and propositions,” but as
someone directly involved with their lives, just as a friend or family
member would be.50 Through these and other activities, congregants
practice suspending disbelief and learn to trust that what is ordinarily
judged by the senses to be “unreal” is actually more real than the physical
world around them. Consuming fiction—the right kind of fiction, at least—
is understood to help with this precisely because fiction also requires the
mind to suspend disbelief and (like kataphatic prayer) gives a vivid reality
to things that might otherwise be hard to conceptualize. The Chronicles of
Narnia by C. S. Lewis are popular amongst Vineyard congregants for this
reason. Lewis’s portrayal of God in the form of the lion Aslan takes its cue

from certain metaphors in the Bible, but it gives those metaphors a palpable
reality that sustains readers as they school themselves in more sophisticated
ways of conceptualizing God. For readers of Narnia, God becomes a vivid
presence whose roar you can hear and whose fur you can touch.51
What are the rewards of all this? The title of Luhrmann’s book is When
God Talks Back, and that is exactly what he does, eventually, according to
those who have diligently followed the path I’ve just described. Sometimes
God helps them decide important issues in their lives—and sometimes he
answers such workaday questions as what to wear.52 From an outsider’s
point of view, it looks as if members of the Vineyard Church have learned
to interpret some of the experiences of their minds and bodies as being not
their own, but God’s—a remarkable demonstration of how reading fiction,
talking to others about suprahuman entities and engaging in cognitive
practices that stimulate the sensory and social regions of the brain can
reinforce the belief that those suprahuman entities are present and paying
attention to us.53 Like Robert-Houdin’s audiences, members of the Vineyard
Church have learned how re-enchant their world, but without any intention
of “leaving the theater.”
Parasocial Interaction
Those of us outside the Vineyard Church could choose to interpret the
congregants’ experiences as self-conditioned instances of what
psychologists call parasocial interaction (PSI). Parasocial interaction—as
opposed to social interaction—occurs when one person thinks about
another person, perhaps even communicates with another person,
unilaterally, without receiving a response. An everyday example of PSI
would be someone noticing an attractive colleague, imagining in some
detail how to strike up a conversation and what that conversation would be
about, but never taking the next step towards actually starting a social
relationship. If parasocial interaction is more than transitory—if it continues
over a sustained period of time—then a parasocial relationship (PSR) is
established. A good example of a PSR would be a fan idolizing a rock star,
attending concerts, and writing fan letters for years, while the star remains
oblivious to the fan’s existence.54
Recent research has shown that all of us engage in PSI to some degree—
with both celebrities and people in our immediate environments—and that

most people develop one or more PSRs in the course of their lives.
Research has also shown that it is very common for psychologically normal
people to develop PSRs with fictional characters—feeling grief, for
example, at a character’s death that is not measurably different in quality
from the grief felt at the death of a real person.55 A famous instance
involves Dickens’s serialized novel The Old Curiosity Shop (first published
in 1840–1841). When readers realized that Dickens was about to kill off a
character named Little Nell, Dickens was inundated with letters begging
him to spare her. Dickens did not relent, however, and after the fatal episode
was published, his audience went into mourning. A member of Parliament,
Daniel O’Connell, began to cry while reading the episode on a train and
threw his copy of the journal out the window. Ships arriving in New York
from England carrying copies of the serial were met with shouts of “Is
Little Nell dead?”56 A similar, although more intense, outcry arose
following Arthur Conan Doyle’s extermination of Sherlock Holmes in
1893. Doyle, however, unlike Dickens, eventually capitulated, resurrecting
Holmes ten years later. Even today, people still write to Holmes at 221B
Baker Street, asking for help.57 More recently, J. K. Rowling’s characters
have inspired a flurry of PSRs amongst both children and adults. Online
chat rooms and blogs discuss questions such as whether Severus Snape was
actually fond of Harry Potter or only protected the boy because of the love
he had once borne for Harry’s mother. Fans offer arguments on either side
that extend well beyond anything that the books actually say. The case of
Snape, incidentally, demonstrates that PSI is not restricted to fictional
characters whom we love or admire. Indeed, actors and actresses who play
villains on screen are chastised and sometimes even physically attacked by
viewers who have grown to detest their characters. (Fans of Breaking Bad
expressed vitriolic hatred for the character of Skylar White that
occasionally spiraled into homicidal rage directed at Anna Gunn, the actress
who portrayed Skylar.)58
And yet, even if a few people forget themselves long enough to accost an
actress, the vast majority of people who experience PSI or develop PSRs
with fictional characters, when asked about their existential status,
immediately concur that the characters do not really exist. As David Giles,
a scholar of media psychology, puts it, “PSI takes place because the figures
are encountered in a narrative context that makes a ‘humane’ response a

logical one.”59 In other words, we engage parasocially with fictional
characters because their creators have made them so convincing that they
prompt the same emotional and cognitive reactions in us as real people do.60
It is probably clear by now why I think that we could choose to call the
relationship that the Vineyard congregants develop with God a self-
conditioned form of PSR: the congregants train themselves to experience
sustained and personal interaction with an individual who, as far as an
outside observer can tell, never responds and perhaps does not exist. Such a
relationship looks just as unilateral as a fan’s relationship with a rock star or
a reader’s relationship with Little Nell. But it is probably also clear why this
characterization is a problematic one: as I have just emphasized, ordinarily,
people engaging in PSIs and PSRs understand that the subject of their
fantasies is not responding and in some cases is not even real. In contrast,
for Vineyard congregants, and for that matter for many other Christians and
members of other faiths as well, God or the gods are decidedly not fictional
and are expected to interact with their worshippers occasionally—through
prophecies, lightning strikes, miraculous cures and a variety of other means.
Which opinion really matters, anyway, when it comes to distinguishing
between social and parasocial interaction—our scholarly opinion or their
experiential opinion? If we can’t see, hear or feel the god with whom
someone is communicating, does this automatically make the relationship
parasocial, whatever the recipient says? This is a very basic question,
which deserves much more research and discussion than it has been given
by scholars (oddly, the similarities between PSRs and religious beliefs have
barely been acknowledged by either psychologists or scholars of religion).61
I cannot entertain it adequately here, but I do want to emphasize the
necessity of not allowing our own lack of belief in particular suprahuman
entities to prevent us from accepting that they can be vividly real to others
and from understanding the variety of means through which that reality
might be constructed. This is particularly important for—but also a
challenge to—scholars of antiquity, due both to the distance that separates
us from those we study and to the fact that ancient gods have had long
careers as subjects of Western art, which has imposed additional layers of
fictionality upon them.
There is, then, a widespread human capacity to form strong emotional
and cognitive attachments to figures with whom there can be no social
relationship in the normal sense of that term (that is, no relationship that is

reciprocal in the eyes of outside observers). I am suggesting that we should
extend this observation to figures of belief whom we usually call “gods,”
“heroes,” “angels,” “saints,” and so on, and that we try to use some of what
researchers have learned about this capacity to understand the ways in
which vivid narrations of myths, centering on vibrant characters, might
have enhanced the ancient Greeks’ relationships with their gods and heroes.
The more skillfully developed are the narratives through which such figures
are presented, the higher are the chances that people will experience PSIs
and PSRs with them.
Under the term “narratives,” I would include a wide variety of things—
not only traditional media such as books, theater, films, and TV programs,
but also worship (during which the figure’s history, attributes, and activities
may be presented, celebrated, or discussed), the viewing of artistic
representations of the figures,62 and personal conversations (in the Vineyard
Church, congregants’ stories of their own encounters with God are
important factors in other congregants’ development of a capacity to
envision and experience God; a similar situation must exist in any other
community where people talk about the Invisible Others that they have
experienced).63
Under “skillfully developed,” I would include not only traditional
narrative skills such as those needed to write a gripping novel, direct an
exciting film, create a captivating statue or give a stunning performance as
an actor, but also a variety of contextual considerations such as where the
narrative is performed and anything else that might accompany it—
costumes, music, scenery, and the like. As I have already mentioned, simply
entering a space that one knows to be given over to encounters with
supernatural entities (a sanctuary, a temple, a mosque, a church—even a
theater, in ancient Greece) helps one to enter into a new state of mind that
entertains the possibility that those entities exist. “Props” (fonts of lustral
water, saints’ or heroes’ relics, lamps in the dark interior of a sacred
building, divine images, bells-and-smells) can help, too. As Landy notes in
his account of nineteenth-century theatrical re-enchantments, the suggestive
use that Robert-Houdin made of what he told the audience were bottles of
ether (at that time, a new wonder drug) lent both a touch of reality and an
aura of mystery to his illusions.
I am not suggesting, however, that the ancient Greeks purposefully
sought to train themselves, cognitively or emotionally, to encounter the

gods and heroes in the way that the Vineyard congregants train themselves
to encounter God (or at least the vast majority of the Greeks didn’t do that
—we might entertain the possibility that something more or less like
kataphatic prayer was used by figures such as Aristeas or Abaris, for
example, who claimed that their souls journeyed while their bodies slept
and then returned to report on what they had seen). Nor am I suggesting that
ancient Greek sanctuaries were frequented by figures such as Robert-
Houdin, who sought to “re-enchant” the populace by suggestive sleights of
hand (some of the itinerant ritual specialists whom we hear about already in
the early fifth century were probably skilled in what we now call stage
magic, but I can’t imagine them choreographing illusions to accompany
public narrations of myths—nor can I imagine the sponsoring institutions
allowing such things). Rather, my aim has been to offer examples of the
human capacity—indeed, I would say the human desire—to open the mind
and the heart to being transported by well-crafted and well-delivered
narratives into a state in which figures who are not part of the ordinary
world seem real, and in which one’s view of how the cosmos works is
altered. Some people, ancient or modern, walk away from such narratives
with nothing more than a temporary uplift of the spirits. Others walk away
with their existing belief in an alternative reality newly confirmed. Others
yet again find themselves changed to some degree—perhaps so affected by
a narrator’s vision that they are more inclined than they were before to start
believing, or to believe in a deeper sense, what they had been hearing about
all their lives—from priests, ritual experts, parents or friends. During the
moment of engagement itself, I suspect, very few remain unaffected
altogether.64 In fact, this power to affect an audience lay at the root of the
objections that Plato and later philosophers made to tragedy.
Episodic Narratives
Good narratives, then, when well delivered, can be powerful stuff—indeed,
a good narrative leads to at least a temporary, and sometimes a longer-term,
erasure of the line between fiction and reality. We can begin to see how the
very fact that Greek myths were often delivered as polished narratives lent
credibility to the worlds that they constructed—worlds in which mortals
and immortals interacted freely—and therefore made those narratives fitting
partners for rituals that were intended to elicit divine response.

In this and the next section, I want to look more closely at some specific
features of Greek myths that encouraged such an erasure of the line and that
therefore helped to sustain belief even after a festival was over. Some
features, such as deixis and the other techniques I mentioned earlier, were
wielded intentionally by skillful narrators—even if not with the specific
intention of effecting such results as those I’ve just been describing. Others,
I suspect, evolved alongside Greek mythic narration itself and became
standard features (although good narrators, of course, might learn to deploy
them in particularly effective ways). Overall, I want to suggest, Greek
myths can be distinguished from many other sorts of narratives by the fact
that their characters became known to the audience in a particularly vivid
and intimate way, which forged an especially strong bond between those
characters and the audience members and nurtured the idea that the
characters could have a real effect on audience members’ lives.
To talk about some of these features, I’ll borrow insights from scholars
who work on narratology and, in particular, those who seek to understand
how episodic narratives work—that is, narratives that extend their story arcs
across installments that are temporally separated from one another. Thus,
for example, if we had read Dickens’s Old Curiosity Shop as it was
originally published, we would have had to read eighty-eight installments,
each of them appearing a week apart, to learn how Little Nell’s story turned
out. If we had read George Eliot’s Middlemarch as it was originally
published, we would have had to read eight installments, each of them
appearing two months apart, to learn how Dorothea Brooke’s story turned
out. If we watched the first season of Homeland when it was originally
broadcast in 2011, it took us twelve weeks to discover whether Nicholas
Brody was a hero or traitor (which is not to say that this question wasn’t
reopened in the second and third seasons and episodically treated again).65
Following the fates of Downton Abbey’s characters required us to view
fifty-two episodes, stretched out over six years.
One of the most interesting effects of cutting up a story into episodes is
that audience members continue to think about the story in between
installments. During these periods, they look forward to meeting the
characters again and begin to contemplate the finer points of character
development and plot. They might speculate about what a favorite character
is going to do in the next installment, or even about what the character
would do in a situation unrelated to the story. Such engagement between

episodes gives characters a life that is independent of the narrative and its
creator and thus increases the chances that the characters will become the
objects of PSIs and PSRs.66 Particularly since the advent of broadcast serials
that simultaneously reach a large number of people through radio or TV,
discussion and speculation about characters have often taken place within
groups as well as in the minds of individuals. Colleagues at the copy
machine ruminate about what will happen to Lady Mary in the next episode
of Downton Abbey or what Eleven will see in the next episode of Stranger
Things. But even before broadcast technologies arose, the characters in
episodic stories were objects of public discussion, as the stories I told about
the death of Little Nell attest. Such group involvement heightens the effects
of any single individual’s PSI with a character. If we are all “believing” in
Little Nell together, our grief at her death is more intensely experienced.67
Many episodic narratives are “series,” in which each episode presents a
self-contained story that has few connections to other stories in the series,
although the characters continue from episode to episode (most situation
comedies work this way—for example, Modern Family). Other narratives
are “episodic serials”: that is, although the main story arc extends across a
number of installments or episodes, each installment or episode includes a
smaller story arc that is resolved within it.68 Within an episode of Mad Men
that was first broadcast in 2012, for example, the character Betty was told
by her doctor that she might have cancer, but by the end of the episode
Betty knew that she was healthy after all. The larger, ongoing narrative arc
in which Betty took part—the ups and downs of her relationship with her
ex-husband, Don—scarcely got any closer to resolution than before,
however. Ancient bardic and rhapsodic recitations fit this model well. On a
given occasion, a rhapsode might narrate the story of Odysseus and
Polyphemus, for instance—but this would have been recognized by most
members of the audience as being just a part of the longer story arc
constituting Odysseus’s return to Ithaca. Indeed, many episodic narratives,
both ancient and modern, characteristically remain incompletely resolved
even when they are formally “complete” (Odysseus will set out again from
Ithaca soon after landing there, as we learn from Tiresias in book 11 of the
Odyssey, before that epic itself is even half over). Other stories are revisited
by new authors, who offer other endings that may (or may not) be more
satisfying than the earlier ones. (Must Antigone and Haemon always die

instead of wed? Must Electra remain unmarried?)69 The mixture of
satisfaction and dissatisfaction with which many “completed” serials leave
their audiences, as Sean O’Sullivan notes, is an extended manifestation of
the very hunger that episodic narratives cultivate on a weekly (or monthly,
or some less regularly articulated) basis: we yearn to hear more about the
characters who have roused our interests, to whom we have become
attached, and with whom the serial has provoked us to form bonds.70
The difference between ancient mythic narratives and the modern
episodic narratives I have mentioned would seem to be one of chronological
control. An ancient listener might have heard the story of Odysseus and
Polyphemus only after he had known for some time that Odysseus
eventually made it back to Ithaca—and thus he knew, before the rhapsode
even began to narrate Odyssey 9, that Odysseus would escape from
Polyphemus’s cave and make it back to his ship. When viewers watched a
new season of Downton Abbey or read a new serialized novel, they
consumed its episodes in a predetermined order, over the span of which the
larger story arc developed in exactly the way that its creator intended. The
difference is not as extreme as we might initially think, however. As
twenty-first-century showrunners know quite well, not all viewers are
“ideal viewers”—that is, some viewers haven’t faithfully followed the
stories as the episodes unfolded over the course of a season.71 Moreover,
once a season is over, viewers may consume episodes in a different order by
watching reruns or downloading episodes from the Internet. Under these
circumstances, some elements of the larger story arc and the rewards of
watching it unfold are suppressed, just as they would be in the case of a
rhapsodic recitation heard out of sequence—most obviously, the element of
suspense as to how the larger story will turn out will be gone. Audience
members can nonetheless enjoy each episode as part of a longer story
through the practice of habits that are deeply engrained in most of us, of
which I will mention just two that are especially common. The first might
be called “willful forgetfulness.” That is, we suspend our knowledge of
what will happen in order to experience it again—we jump with fright
every time that Margaret Hamilton makes her initial appearance as the
Wicked Witch of the West even though we know that she will later end up
as a puddle. Moreover, as Alfred Hitchcock once observed, the effect of
what we commonly call suspense comes at least as much from knowing
what is about to happen and being unable to intervene as it does from not

knowing what is about to happen—a sentiment that any member of a Greek
tragic chorus would endorse. It is the way that the story is told that
generates emotional response in both “spoiled” and “unspoiled” audience
members, as much as the story itself.72 The second habit involves treating
an episode that one consumes out of sequence as a backstory that explains
how the characters ended up where they did or that offers new insights to
their natures. The original readers of serialized novels such as Middlemarch
sometimes missed an installment that they could read only later, in the
home of a friend who had been scrapbooking the serial or in the completed
novel itself, once it had been bound between covers. Such a person might
read chapter 21 only after she already knew that Dorothea Brooke had
married Will Ladislaw at some undescribed moment between chapters 84
and 85, but chapter 21’s narration of an unexpected meeting between
Dorothea and Will in Rome, years earlier, adds interesting insights as to
why their marriage eventually took place.
Consuming episodes of a novel or a television show out of sequence is a
good analogy for the way that Greek audiences experienced mythic
narratives. People who listened to Odyssey 9, to Bacchylides’s dithyramb
about Theseus’s trip to Crete, to Pindar’s tale of Jason’s adventures,73 to
Sophocles’s Philoctetes and to many other mythic narratives were thrown
into each hero’s story in medias res (perhaps into a part of the story they
hadn’t heard before or even into a part of the story that didn’t exist until the
poet invented it), but most of the listeners brought to those performances a
basic knowledge of the heroes’ larger histories. And in fact, particularly if
we think about the body of Greek mythic narratives as a whole—a body to
which many different authors independently contributed episodes of gods’
and heroes’ stories—we can more fully appreciate that this mode of
consumption brings its own rewards: if the single episode is imaginatively
composed, then it inevitably exists in a productive tension with the larger
story arc as the audience knows it. Good narrators play with this fact. Later
poets often presented the smaller story they were narrating in a way that
challenged the larger arc that had been provided by epic, for example.
Helen as we meet her in Euripides’s Helen is quite a different woman, who
has had different experiences, from the Helen we meet in the Iliad.74 Hesiod
states outright something that Homer never mentioned: Odysseus had three
sons by Circe and another by Calypso.75

But to return to the main point I’ve been making, Greek mythic
characters, like Dorothea Brooke, Little Nell, Nicholas Brody, Lady Mary
Crawley, Eleven, and other serialized characters, were served up to their
audiences in small doses, a circumstance that (if the narratives were
effective) whetted listeners’ appetites to hear more about them and
encouraged them to think about those characters—even to develop PSRs
with those characters—during the intervals in between. The fact that, in
contrast to all of the modern characters I have just named, Greek mythic
characters endured through the centuries enhanced this effect: almost
everyone knew at least parts of their stories from an early age.
Narrating Belief
In the preceding section, I looked at what happens when a narrative is
delivered episodically, as ancient Greek myths were: audiences tend to
become more personally engaged with the characters and their story lines.
In the section before that, I discussed an aspect of human psychology that
makes it easy for audiences to form strong attachments with characters,
including the characters of myths, with whom they cannot interact in the
ways that they do with people in their everyday lives.
In this section, I look at certain features of narratives themselves that
make them good at persuading their listeners to believe in phenomena that
are not accessible to our everyday senses. Generally speaking, in doing so I
align myself with a recent emphasis on rhetorical poetics in literary
criticism, as represented by the work of James Phelan, for example. Phelan
has urged us to pay more attention to literary narratives’ affective, ethical,
and aesthetic effects and to the way that narrators use “the resources of
narrative in order to accomplish certain purposes in relation to certain
audiences.”76
To date, however, rhetorical poetics has not been interested in unpacking
the ways in which literary narratives affect beliefs. To get at that, I begin
from two folklorists’ studies of how people in contemporary cultures
informally narrate experiences with the supernatural. One is Gillian
Bennett, who studied beliefs in visitations (that is, visits from dead relatives
and friends) experienced by women in 1980s Manchester, England. The
other is Kirsi Hänninen, who studied beliefs in supernatural encounters
experienced by Finnish people in the early 2000s. Bennett and Hänninen
have identified a number of techniques that people telling stories about such

experiences—with ghosts, with God and with other entities—employ to
help persuade listeners that the stories are true.77 Folklorists usually call this
type of narrative a “memorate,” to distinguish it from “literary” ghost
stories. The term springs from the assumption that the narrators of
memorates believe that the stories they are narrating are genuinely drawn
from their memories, whereas the authors of literary stories acknowledge
them to be their own “fictional” creations. A criterion that so sharply
divides “fiction” from “truth” is problematic for the materials that this book
addresses, but the terms “memorate” and “literary ghost story” are
conveniently concise and so I will use them here.78
In the majority of Bennett’s and Hänninen’s cases, the techniques used by
those who narrate memorates are employed unconsciously; the narrators
have internalized the characteristics of persuasive narratives about the
supernatural that they have heard and then replicated them themselves. The
circumstances of presentation are usually informal; although by definition,
the subjects in Bennett’s and Hänninen’s studies have been prompted to
narrate their stories, memorates typically are told amongst groups of friends
and relatives, with little or no preparation beforehand on the part of either
the narrator or the audience. With memorates, then, we seem to be moving
from Noyes’s first category of cultural forms, which are sought out and
receive focused attention, into her fourth category of forms, which also
receive focused attention but which are unsought. These modes of creation
and presentation are so different from the very conscious polishing that
Greek poets put into their narratives about the gods and heroes as to make
comparison between the two seem impossible, perhaps. However, some of
the most successful authors of literary ghost stories from the turn of the last
century—the period during which ghost stories were developing as a genre
and being purposefully honed to what is considered to be their peak of
excellence—employed some of the same techniques as do the narrators of
memorates. These techniques persuade us to believe that the stories they are
telling are true, at least during the time that we are reading them, and
perhaps longer. This should encourage us to look within ancient literary
narratives about the gods and heroes for variations of these techniques as
well and to try to gauge what effect they had on ancient audiences.
In the next part of this section, I’ll describe some of the techniques that
Bennett and Hänninen have discussed and sketch how M. R. James (1862–

1936), a master of the literary ghost story, employed them. I’ll then turn to
passages from Greek narratives that I suggest worked similarly.
The X/Y Format
Many of the techniques used by the narrators of memorates contribute to
what the sociologist Robin Wooffitt has dubbed the “X/Y format” of
narration.79 These narrators need to persuade audiences not only that the
extraordinary experiences they relate really happened (what Wooffitt calls
the “Y” factor) but also that the narrators themselves are sane, normal
people who function successfully within the familiar world (Wooffitt’s “X”
factor). Therefore, memorates frequently begin with statements that
describe the familiar world and the protagonist’s ordinary life within in it
(the X factor), such as this one, taken from Bennett’s work:
Ned was working at the time of the story for a local farmer, Sam
Black, at the Manor Farm at Dell, and he used to have to go to market
with these cart horses, and he was going to Bradbury market one
terrible frosty day.
The story continues with Ned’s lead horse slipping on the ice and dragging
the other horse down as well; Ned finds himself in a fix on a lonely country
road. Then the supernatural element (the Y factor) enters the narrative, as
Ned’s dead father speaks to him:
[Ned] said that Dad’s voice CAME TO HIM QUITE CLEARLY,80 said “Cut the
girth cord, Ned! Cut the girth cord!”81
X factors needn’t come at the beginning of a story, however, and they

needn’t focus only on the normalcy of the outside world. Narrators’
descriptions of their own thoughts and intentions can help to establish their
credentials as normal, sensible people, as well. One Finnish woman in
Hänninen’s study described an attack by a threatening supernatural entity in
detail, and then concluded with:
I woke up and shivered in horror for several hours. My husband and I

went to the emergency room and I got sedatives. We decided that if

this event happened again, we would move to another apartment, but it
didn’t.82
The narrator emphasizes the real-world, goal-oriented logic that she used to
respond to her experience: her visit to the hospital, her taking of sedatives,
her plans for relocation. By doing so, she establishes that she did what any
ordinary, sensible person would do in any emergency.83 In addition, the
reference to her husband is a subtle call to witness that endorses the
narrator’s story by suggesting that another person took it seriously.
At times, the X factor in a memorate can be developed to a high degree,
including such things as banal descriptions of the weather at the time the
incident occurred or what the people involved were planning to have for
dinner that evening. This is what Bennett calls “evidential scene-setting.”
Some of the best turn-of-the-century ghost stories do this to a luxurious
extent—which, given that they were destined for publication over many
pages of a book or feuilleton, was possible in a way that it isn’t for the
narrator of a briefer memorate. Characters in these stories engage in
conversations about train timetables and upcoming charity bazaars; they are
described going about their daily business as clerks, curators of museums,
headmasters on holiday, and so forth, when the supernatural bursts upon
them. As M. R. James himself said,
If [a ghost story] is to be effective, I think that as a rule the setting

should be fairly familiar and the majority of the characters and their
talk such as you may meet or hear any day. A ghost story of which the
scene is laid in the twelfth or thirteenth century may succeed in being
romantic or poetical: it will never put the reader into the position of
saying to himself, “If I’m not very careful, something of this kind may
happen to me!”84
In a variation of the X/Y strategy, the narrator may express ambiguity

about what he or she is describing, insisting first (for example), “I was fully
awake” and “[the uncanny visitor] stood in front of me,” but then declaring,
“I don’t know whether I was dreaming or not,” before returning to the
assertion, “and he [the uncanny visitor] was there”—all within a short
narrative space. Bennett suggests that this strategy is adopted in order to

save face—narrators try to distance themselves from their remarkable
declarations even as they make them, so as to align themselves with
sensible people85—but the strategy simultaneously serves to reassure
listeners that the narrator strives to operate under the same “commonsense”
principles as they do and strives to understand his or her experience in that
way.
Interjections and the Dialectical Relationship
There are several types of interjections that a narrator might make while
telling a memorate. Endorsements and justifications are amongst the most
frequent used by Bennett’s and Hänninen’s subjects. Most common is
endorsement by personal assertion (“I know I experienced this!”) or
emphasis (“Oh yes, yes, yes. Oh, yes. I do [believe], yes”), but some
narrators have recourse to religious principles, adding justificatory remarks
such as “God wouldn’t allow it” or “It was Saint Paul, wasn’t it, said we’re
encompassed with a great cloud of witnesses?”86 And so it is in literary
ghost stories as well. Near the end of James’s “Canon Alberic’s
Scrapbook,” the narrator, a friend of the man who suffered the supernatural
experience, says,
I never quite understood Dennistoun’s view of the events I have

narrated. He quoted to me once a text from Ecclesiasticus: “Some
spirits there be that are created for vengeance, and in their fury lay on
sore strokes.” On another occasion he said “Isaiah was a very sensible
man; doesn’t he say something about night monsters living in the ruins
of Babylon? These things are rather beyond us at present.”87
Interjections play another important role, in addition to endorsing,

justifying or doing whatever they individually do. When a narrator
interrupts a story, he or she also interrupts the integrity of the diegetic88
world that the story has been creating. Such interruptions remind listeners
that there is a difference between the “Narrating I” (the person relating the
memorate, who exists in the “Story-Realm,” that is, the time and place in
which the story is being told) and the “Experiencing I” (the person who
lived through the experience that the memorate describes, who exists in the
“Tale-World,” that is, the time and place in which the events took place).89

Anything that asserts the difference between these two “I’s” helps to
establish that the person telling the story has distanced him- or herself
sufficiently from the extraordinary experience as to be able to carefully
consider whether it really happened. In a sense, this works towards the
same result as does the X/Y format; it helps to show that the narrator is
trustworthy while nonetheless allowing the narrator free rein to tell a
remarkable tale.
Moreover, those who narrate memorates in the X/Y format or with
interjected comments are implicitly inviting audience members into a
dialectical relationship, using the story not simply to persuade listeners to
believe in what they describe but also to explore the nature of that
experience along with the narrator—questioning it, doubting it, trying to
understand it, and eventually, if the narrative is successful, believing it. The
introduction of ambiguity, which I mentioned earlier as a means of
establishing the X/Y format, helps to set up this dialectical relationship, as
well, given that it compels listeners to make a choice between (or amongst)
the different possible messages that the narrator is sending.
As Bennett says, traditions of both belief and disbelief are transmitted
through interactive processes such as face-to-face communication, the
sharing of information and the telling of stories.90 Storytelling is particularly
crucial here for a reason that I discussed in Chapter 1: a story can describe
events that have not been experienced by the listeners themselves so
persuasively that the events come to seem credible, thus enlarging the sense
of what might be possible beyond the quotidian witnesses of the usual five
senses. It is no surprise therefore that, as Bennett says, “people customarily
respond to questions of belief with narrative answers. The more
controversial the topic, the likelier it is that the conversation will include a
lot of narrative.”91
Ancient Narrations of Remarkable Incidents

From antiquity, we have no true memorates—there were no folklorists to
record them. We do have a few first-person narratives of supernatural
encounters within formal literary works, however: Hesiod’s encounter with
the Muses on Mount Helicon, for example.
We also have some “vicarious memorates”92—that is, secondhand
accounts of someone else’s experience. (We have already seen two
vicarious accounts from the twentieth century: the story about Ned and his

father’s voice, told by a relative and recorded by Bennett, and the end of
“Canon Alberic’s Scrapbook,” a literary ghost story that is narrated by the
friend of the person to whom the supernatural experience happened.) From
antiquity, we have, for example, Herodotus’s report that the runner
Phidippides met the god Pan on Mount Parthenium. The god told
Phidippides to instruct the Athenians to pay more attention to him; if they
did so, he promised to help them. Phidippides must have been a persuasive
narrator, for the Athenians duly built a shrine to Pan and instituted an
annual festival in his honor.93 Herodotus doesn’t give us the kind of details
we’d like to have about this incident, however; he is more interested in
telling us about interaction between the Spartans and Athenians on the eve
of the Battle of Marathon, in which Phidippides played a role—
Phidippides’s encounter with Pan is just a side story. Ideally, we would like
to know what Phidippides saw, heard and felt when he met Pan and what
words and techniques he used to tell his story to the Athenians so
convincingly.
We also find vicarious memorates tersely preserved in inscriptions that
were displayed in sanctuaries, recording miraculous things that the
indwelling god had done. The Epidaurian iamata (“cure stories”) are the
best known of these. Patients who slept in the abaton (a special inner room
of the sanctuary) at Asclepius’s sanctuary in Epidaurus would be visited by
the god or one of his divine children in their dreams and wake up healed.
Some of the most remarkable of these cures were recorded by the priests
and engraved on tablets that were erected in the sanctuary.94 Patients would
read these (or have them read aloud) before they entered the abaton
themselves, which, hopefully, would bolster their belief that the god would
help them, too. Some of the iamata include vivid details—one patient
dreamt that Asclepius ordered some servants to bind him tightly to a door
knocker to keep him from moving while Asclepius cut his abdomen open to
remove an abscess, for instance. Afterwards, when the patient woke up,
cured, he found that the floor of the abaton was covered in blood.
These cure stories are short and have been stylized to fit particular
narrative formulas.95 As a result, the patient’s voice has been lost almost
completely. It is interesting to note, however, that the priests who
configured the stories before they were inscribed often included physical
proofs that what the patient said happened had really happened: the door
knocker to which one patient had been bound, for instance, or the blood on

the floor of the abaton, the kidney stone that Asclepius handed to a patient
after removing it from the man’s urinary tract, the arrow pulled out of
another patient’s lung and the statement that before the arrow was pulled
out the patient’s wound had filled sixty-seven basins with pus.96 Although
the X of the X/Y format in these stories was implicitly established by the
fact that the miraculous cures happened in a building that was within sight
of where tablets were erected, details such as these must have contributed to
establishing veracity as well—to a sort of evidential scene setting. One can
imagine a priest, reading aloud these stories, gesturing towards the door
knocker to which the patient had been bound or showing his audience the
arrow pulled from a lung. Many of the iamata include another element that
evokes Bennett’s work, as well. The omniscient narrator of the story often
says, “it seemed to so-and-so that Asclepius appeared in a dream and did
such-and-such.”97 The phrase “it seemed to” subtly distances the narrator
from the event, thus helping to establish that the narrator is trustworthy
before the rest of the remarkable tale unfolds.
The iamata and the story of Phidippides are set in what was, for the
Greeks, contemporary time—that is, these stories involved individuals who
lived at the same time that their audiences did or in the very recent past.
Some of the people who heard or read these stories knew the individuals to
whom the experiences had happened or had parents or grandparents who
knew them. The stories that most interest me in this book—the myths
composed by poets—are different in two ways. First, they are set during an
earlier age when, it was thought, heroic humans were still walking the earth
and encountering gods firsthand. The poets needed to show that at least
some of the people in that world were like their audience members in order
for the audience to feel a connection to them, but the poets also had to show
that the heroes were exceptional figures—figures who did remarkable
things that helped or hurt other humans while they were alive and who, now
dead, could continue to do remarkable things that would either help or hurt
humans. This sometimes required breaking the guidelines that M. R. James
set for himself, as expressed in the earlier quotation, as we’ll see.
Sometimes, effective poetic narratives needed to convince the audiences
that such things had been possible in a world that operated somewhat
differently from the way that theirs did.
Second, the poets’ stories have been very consciously shaped to affect
their audiences—in this sense, they are a lot like James’s ghost stories. We

will never know, however, whether a given poet purposefully adopted
narrative techniques that he or she recognized as being conducive to belief,
unconsciously replicated what he or she had heard other convincing
storytellers doing, or something in between, because the poets (unlike
James) have not left us any comments about how they thought such stories
should be shaped. My presumption, moving forwards, is that the poets were
probably aware that certain modes of telling a story and certain features
within a story had greater effects on their audiences than others did, and so
used those modes and features when it was appropriate. They probably
would not have articulated this in anything like the terms that James did,
however. In other words, I assume that although they were consciously
crafting effective narratives, usually the poets were not consciously setting
out to induce belief per se in the same direct sense that the priests who
composed the iamata were.
To pursue these ideas further, I’ll take as my representative samples the
beginning of Hesiod’s Theogony, portions of the longer Homeric Hymns,
some portions of the Odyssey, some selections from poems by Bacchylides
and Pindar and the messenger’s speech from Euripides’s Hippolytus.
Hesiod’s Theogony
At the beginning of Theogony, Hesiod informs us that he was tending sheep
at the foot of Mount Helicon when the Muses burst upon him, spoke to him,
taught him how to sing beautiful songs, and handed him a branch of
flowering laurel as a token of his new skills and responsibilities.98 Before
they did all of this, however, they berated him and the class of shepherds as
a whole: “Field-dwelling shepherds! Ignoble disgraces! Mere bellies! We
know how to say many false things similar to genuine ones, but we also
know, when we wish, how to proclaim true things.”99
Suggestions as to how we are to understand this encounter have run the
gamut.100 Was it a dream? A waking vision? Is there even a significant
difference between the two? Or is the scene, when all is said and done, just
a literary topos, anyway? Tilting the scale towards the latter assumption are
all the other cases from the ancient Mediterranean world in which a poet or
a lawgiver receives instructions during an encounter with a god on or near
the mountain where the god dwells (for instance, Moses) and all the other
cases in which a future poet or holy man is caring for animals when the

Muses or some other extraordinary experience descends on him (for
instance, Archilochus, Epimenides, and Moses).
But as M. L. West said, “the presence of these and other typical elements
in Hesiod’s vision need not mean that it was not genuine.”101 Or, to shift the
viewpoint slightly, the presence of typical elements in Hesiod’s account
need not mean that those elements were simply rote. Indeed, if they
continued to be included in such stories, and especially in Hesiod’s poem,
which was much admired already in antiquity, then they must have had
some effect on their audiences. Carolina López-Ruiz has urged that “rather
than assessing the experiential nature of Hesiod’s self-representation (which
we shall never know), we need to be aware of how, as a narrator, he
presents himself in a way that predisposes his audience to take his account
seriously and that elevates its contents to those of exceptional
revelations.”102
Interestingly, the lines provide a setting that nicely fills out the X part of
an X/Y format. “There I was, just minding my own business on Mount
Helicon, tending my sheep as I did every day (X), when suddenly the
Muses appeared and gave me an incredible gift and their divine advice!
(Y).” The Muses’ description of the class of shepherds—and thereby
Hesiod—as “mere bellies” heightens the effect, given that it accentuates the
difference between the human and the divine that have just come together
so precipitously.103
The coda to these lines is important, too: “But what is this to me, about
an oak or a rock?” asks Hesiod rhetorically, before he launches into the
poem proper, leaving his personal experience with the supernatural behind.
The expression, proverbial in ancient Greece, meant something such as
“Why have I bothered to talk about all of this?104 What he is saying, of
course, matters a great deal—most of his poem will constitute a saga of the
gods’ early days, how they battled to bring the cosmos into its final state
and eventually produced the generation of heroes. Therefore, how and from
where Hesiod got the information that he tells us also matters a great deal.
By inviting us into an implicit dialogue about how to assess the
supernatural experience that he has described, Hesiod makes us pay close
attention to what is to come, therefore; we will bear some of the
responsibility for deciding whether it is trustworthy, which will make our
commitment to its truth all the stronger if we choose to accept it.105

The Homeric Hymns
The longer Homeric Hymns give us stories in which divinities invade the
human world unexpectedly, with terrifying ramifications for the humans
whom they visit. Demeter comes to Eleusis in disguise, Apollo kidnaps a
boatload of Cretans, Aphrodite seduces Anchises and Dionysus terrorizes
some pirates who kidnapped him.106 In each of these cases, the poet shows
us not only how humans react to the presence of a god (the Y factor) but
also something of what normal life in the everyday world was like before
the god arrived (the X factor).
This is done most extensively in the Hymns to Demeter and Apollo. In
the Hymn to Demeter, daughters of the Eleusinian king’s household are
going about their daily business, fetching water from a well, when they
encounter the goddess Demeter, who at this point in the narrative has
disguised herself as an old woman. The girls take Demeter home to their
mother, Metanira, who has recently given birth to a son and is therefore in
need of a nursemaid—a post that the disguised Demeter has told the girls is
ideally suited to a woman of her age.
Metanira is sitting in the palace amongst her serving women when
Demeter steps onto the threshold, filling the doorway with divine radiance
as her head touches the lintel. In spite of the fact that Demeter is still
disguised as an old woman, Metanira is seized by “awe, reverence and
sallow fear.” Some time later, after Demeter has been working in the
household for a while, Metanira peers out from her bedchamber one night
to see Demeter placing her infant charge in the hearth fire. Metanira “slaps
her thighs and cries out in alarm and lament.” Angered by this, Demeter
casts off her disguise, which fills the house with “a brilliance like
lightning.” Metanira gives way at the knees, is struck speechless and fails to
notice that her beloved son lies crying on the floor, where Demeter has
dropped him. Trembling, the other women spend the rest of the night trying
to propitiate Demeter.107
In the Hymn to Apollo, a boatful of Cretan sailors—“fine men” who are
going about their daily affairs, “sailing on business towards sandy Pylos and
Pylos’ people”—are struck “silent in terror” when Apollo manifests himself
on board ship as a huge dolphin, an “enormous and terrible beast.”
Whenever anyone tries to touch the dolphin, it “tosses him off in any old
which-way, shaking the whole ship down to its timbers.” When the sailors
try to steer the ship, it will not obey the rudder. When Apollo subsequently

goes ashore and enters his temple, filling it with radiance, the wives and
daughters of the local town “scream under [Apollo’s] force, for the god has
filled everyone with terror.”108
In the Hymn to Aphrodite, Anchises, a cowherd on Mount Ida, is playing
his lyre in his hut when he is approached by Aphrodite, disguised as a
human maiden. She convinces him to make love to her in his own humble
bed, strewn with animal skins. When, upon waking from his postcoital nap,
he sees her in her true form and learns that he has slept with a goddess, he is
afraid, and covers his face with a blanket.109
In the Hymn to Dionysus, Dionysus, disguised as a handsome, well-
dressed young man, is spotted by some Tuscan pirates who happen to be
sailing past, “speeding over the wine-dark sea,” as pirates are wont to do.
They stop, leap from their ship, grab the boy and drag him aboard,
presuming that he comes from a wealthy family. As the pirates are
tightening the sheets, Dionysus causes grapevines and ivy to climb all over
the ship, a sight that “seizes the pirates with astonishment.” When the god
changes himself into a lion and causes a bear to appear on deck, as well, the
pirates flee to the stern, where they halt in terror. When the lion eats the
captain, the other pirates jump into the sea, where they are transformed into
dolphins.110
The juxtaposition of the everyday lives of these humans with the
terrifying advent of the gods creates X/Y formats. The effect in these cases,
however, is not so much to establish that the narrator is a normal, reliable
person whose remarkable story must therefore be true (as in the cases
studied by Bennett and Hänninen) but rather to establish that the humans in
the story are very much like those in the audience, in spite of the fact that at
least some of them (Anchises and the Eleusinian royal family) formally
qualify as heroes (as entities who deserved worship) from the audience’s
point of view. The time in which the myths are set and the time in which the
poet is narrating them are telescoped by the details of familiar life that the
poet includes, making it easier for the audience to imagine themselves in
these circumstances and to vicariously experience a god’s presence. If
scholars are correct that some of these hymns were composed to be
performed at cult sites that had connections to the stories being told, then
the settings would have helped to accomplish that elision.
Before we leave the encounters narrated by these Hymns, I want to
underscore how unpleasant most of them are. It’s true that Anchises is

“seized by desire” when first he sees the disguised Aphrodite and that the
Eleusinian infant, Demophon, grows “like a divine being” while under
Demeter’s care.111 It’s also true that Demeter establishes the Eleusinian
mysteries and Apollo the Delphic Oracle, both of which are of great benefit
to humanity. But in the Hymns, we hear most often about “fear,” “alarm,”
“terror,” “screaming,” “astonishment,” and the necessity of hiding under
bedclothes even when the gods in question intend no actual harm. For the
audiences who listened to these hymns, the gifts that the gods had bestowed
upon mortals were valuable assets, but the conditions under which they
were bestowed were reminders of the fact that gods were very different
from mortals, that a face-to-face encounter with one was likely to be
terrifying—and that at any moment gods might be walking through the
human world in disguise.
The Odyssey
Most of the Odyssey is narrated by the poet, who omnisciently describes
what is happening to the characters, but books 9 through 12 are narrated by
Odysseus himself, as he tells the Phaeacians about his amazing adventures,
and part of book 4 is narrated by Menelaus, who had an amazing adventure
of his own with Proteus, the Old Man of the Sea. This presents us with two
different kinds of reports about extraordinary experiences. In the first, we
should look for the ways in which the poet presents Odysseus, his
companions, and his family members reacting to such experiences. The
other kind of reports come close to being memorates (although, like the rest
of the Odyssey, they are delivered in a formally determined meter, diction,
and style). We should be alert within these reports for the ways in which
Odysseus and Menelaus use something like the techniques described by
Wooffitt, Bennett, and Hänninen to persuade their listeners of what they are
saying—particularly Odysseus, who had quite a reputation as a raconteur.
Prime examples of the first kind of report are episodes such as those that
Richard Buxton has recently discussed: scenes in which a god, typically
Athena, changes into an animal, typically a bird.112 I will take as an example
the incident with which Buxton opens his own discussion, a passage from
Odyssey 3 that is set in Nestor’s palace in Pylos. Telemachus and Athena,
who is disguised as Mentor, are visiting Nestor, and “Mentor” has just
finished advising Nestor about what he should do to help Telemachus.
Having finished this task, Athena decides to depart for Olympus: “So

saying, green-eyed Athena went away, likening herself to a vulture, and
astonishment [thambos] took hold of all those who were looking, and the
old man was amazed [thaumazen], as he saw it with his eyes.”113
Interpreters have sometimes tried to wriggle out of what the text so
clearly says—that Athena turned herself into a vulture—by arguing that, for
example, the poet is speaking only metaphorically: Athena is merely “as
swift as a vulture.” After Buxton has rehearsed these arguments (which he
suggests are anchored in the discomfort of thinking about an
anthropomorphic divinity lowering herself to such a disguise), he concludes
that “such a move, like other attempts to rescue the poem from fantasy and
render it more ‘real,’ is not only unsustainable but also unnecessary.”114 I
agree with him, but what interests me most at the moment is a topic that is
more broadly taken up in his book: the amazement with which this
remarkable occurrence is greeted. Indeed, in this brief passage, the
amazement is signaled twice: once through the noun thambos and again
through the cognate verb thaumazen.
Up until this point in the story, the visit of “Mentor” and Telemachus has
been pleasant and productive but completely as one would expect: they
have been entertained, they have been present at a sacrifice that is narrated
in familiar detail, they have talked, and Nestor is preparing to make his
visitors comfortable for the night. The X of Wooffitt’s X/Y format has been
generously laid out, in other words. And then the astonishing thing happens
—our Y factor!—as “Mentor” transmogrifies himself before their very
eyes. Telemachus, Nestor, and the other people who see this are
understandably surprised, and their surprise serves to stress the
remarkability of what has just happened for the people listening to the
poem, disposing them to be amazed as well. The final phrase in the passage,
“as he saw it with his eyes,” is a variation of Bennett’s call to witness, as it
insists that Nestor, whom the Odyssey and other epics have firmly
established to be a reasonable person, was not dreaming or half asleep but
fully awake and in control of his senses when he saw what he claims to
have seen.
There are several examples of the second kind of extraordinary
experience in books 9 through 12 of the Odyssey, as Odysseus and his crew
encounter various monsters, an enchantress and the ghosts of the dead and
then, finally, witness the weird portent of the skins of the slaughtered oxen
of the Sun crawling across the sand, while their flesh bellows and moos as it

rotates on spits over a fire. The descriptions of remarkable experiences in
these books run contrary to the typically terse way that epic poetry
describes such things; in the Iliad, descriptions of the Chimaera, Cerberus,
and Typhon are all dispensed in just a few words, for instance.115 Two of the
reasons for this difference seem obvious: Odysseus is narrating his own
story and wants to make himself look as good as possible—the greater the
foe, the greater the hero. And, given that he is a hero known particularly for
his way with words, we would be surprised if he didn’t linger over
extraordinary events.116
Some of Odysseus’s more extended descriptions include a feature for
which Wooffitt, Bennett, and Hänninen have not prepared us: Odysseus
knows in advance exactly what awaits him, for Circe has told him about
what he will encounter and has advised him about how to deal with it. We,
his listeners, therefore know all of this in advance, as well. There is little or
no element of surprise when the encounter actually takes place, therefore,
either for the internal audience (the Phaeacians) or for us. This stands in
contrast to the amazement when Athena turns into a vulture and the
amazement or horror evoked by the events described in modern memorates
and literary ghost stories. But even as Circe’s words dull surprise, they
create a proleptic call to witness: however fantastic the experience may be
about which Odysseus tells us, it has already been confirmed by a goddess
who knows about everything that is going on in her particular section of the
world, a goddess for whom the Sirens, Scylla, and Charybdis are normal
parts of life.
The case of Scylla and Charybdis is especially interesting in this regard.
Circe’s preview of Scylla’s physique and modus operandi is extensive—a
bit longer, in fact, even than Hesiod’s description of the great monster
Typhon117 and longer by far than any other description of a monster in
Archaic or Classical literature. Scylla is an “evil monster” who “howls
terribly,” although her voice itself is like that of a new puppy. She has
twelve feet that wave about in the air and six gangly necks, each topped
with a horrible head that has three rows of close-set teeth that are “full of
black death.” From the waist down, her body is hidden by the cavern in
which she lurks, but she pokes her heads out the door in order to fish over
the cliff, looking for dolphins, dogfish or one of the sea monsters that
Amphitrite tends. If she spots a ship passing by, all six heads dart out at
once, gobbling up six sailors at the same time. Later, when Circe and

Odysseus discuss the best strategy for getting through the strait where
Scylla and Charybdis lurk, Circe adds that Scylla “is no mortal thing, but an
immortal evil—dread, dire, ferocious, un-fightable, against whom there is
no defense.” Circe’s description of Charybdis is much shorter but still
terrifying: three times a day, Charybdis sucks down the black water and
three times a day she vomits it up again. Not even the Earth-shaker himself
could rescue a ship caught in that whirlpool, she warns.
The next day, after a close encounter with the Sirens, the men in
Odysseus’s ship are almost immediately “struck with terror” again because
they already see smoke and heavy surf ahead and hear a thundering sound
—indications that Charybdis is near at hand. In fear, they let their oars drop.
The blades dangle free in the oarlocks, drifting about and banging up
against the side of the ship—a bit of evidential scene setting. The vessel is
now dead in the water, dangerously stranded. Odysseus walks back and
forth between the lines of rowers, giving them a pep talk, reminding them
of the dangers they’ve already survived, telling them to move forward, and
instructing the helmsman about how to handle the steering as they get
nearer to Charybdis—a sensible, real-life response to the danger they are
about to confront and very characteristic of what we already know of
Odysseus.
And then Odysseus tells his audience what he was thinking at the time
that all of this was happening and how he had prepared for it. He reveals
that he had told the men in advance only about Charybdis and not Scylla, “a
problem that could not be dealt with,” lest they leave their benches and
cower fearfully in the belly of the ship. In other words, Odysseus testifies to
the fact that, even under extraordinary circumstances, he operated in the
same way as he always did, in a rational, goal-oriented manner. Less
rationally perhaps (although just as characteristically), in spite of Circe’s
admonition to stay under cover himself, Odysseus puts on his armor, takes
up two spears and stands at the ship’s prow—preparing to confront Scylla.
The X that precedes the Y in this episode is somewhat less developed
than in the examples we have looked at from Hesiod and the Hymns and
less developed than in some of Odysseus’s own adventures.118 This is
mostly because Odysseus has moved straight from one extraordinary
encounter (the Sirens) to the next; the dangling oars banging against the
ship provide a touch of everyday reality, however, as do Odysseus’s
practical instructions to the men and his own interior considerations. And so

they draw nearer to the strait on either side of which Scylla and Charybdis
lurk, and Odysseus describes what he saw next:
Charybdis churns up the water around her, vomiting it forth like a

cauldron over a strong fire, the whole sea boiling up turbulently and
the sea-foam spattering the tops of the surrounding rocks. When
Charybdis sucked down the salty sea-water, the turbulence revealed
what lay beneath the sea, and the rocks all around her groaned terribly,
and you could see the ground at the bottom of the sea, black with sand.
Green fear took hold of the men. In terror of destruction, we kept our
eyes on Charybdis.119
But while they are staring at Charybdis, Scylla darts from her cave.
Odysseus says,
[She] snatched up six of my men, the strongest and most able of them,
and when I turned round to look at my ship and my other crew, I saw
their feet and hands from below, already lifted high above me, and
they cried out to me and called me by name, the last time ever they did
so, in heart’s sorrow.120
This is a graphic, evocative description, more than enough to paint a

horrifying picture in our minds. But Odysseus adds an extended metaphor
that pulls us back, for a moment, both to Circe’s earlier description of
Scylla (thus tacitly confirming its evidential value) and into the normal
world that offers such a contrast to the scene he is witnessing. By doing so,
he reminds us that even in the midst of this extraordinary experience he is
capable of a certain calm reflection that allows him to draw on his everyday
experiences—and that, therefore, he is a man whose description of events
can be trusted:
And as a fisherman with a very long rod, on a jutting rock, will cast his
treacherous bait for the little fishes, and sinks the horn of a field-
dwelling cow into the water, then hauls them up and throws them on
the dry ground, gasping and struggling, so did my men gasp and
struggle as they were drawn up the cliff. Right in her doorway Scylla

devoured them. They were screaming and reaching out their hands to
me in this sickening struggle. That was the most pitiful thing I have
ever seen during my sufferings as I travelled the seas.121
The metaphor also serves in somewhat the same way as do the shorter
interjections that Bennett studied, moving us away from an extraordinary
scene for a moment, giving us some room to decide whether we will accept
the story that Odysseus is telling us.
Odysseus’s immediate audience, the Phaeacians, reacted to this and the
other strange tales that he told them by sitting in silence, spellbound by his
narrative but believing it; the king, Alcinous, promises to take Odysseus
home, at last, in order to escape from all of his many earlier sufferings.122
An audience listening to a rhapsode recite this tale in historical Greece
would not, perhaps, have been able to experience what Odysseus did, even
vicariously, with quite the same depth of verisimilitude as I have suggested
they experienced what the Homeric Hymns described, but the features that I
have stressed in my reading of the episode involving Scylla—the proleptic
call to witness, the evidential scene setting and the extended metaphor that
drew on normal life—helped to persuade them that these things had
happened once in an earlier time, in a faraway place, to a man who was not
so different from themselves.
Bacchylides and Pindar
The characters in Bacchylides’s and Pindar’s myths do not live the lives of
normal people in historical Greece, either.123 These characters—the heroes
—travel to Hyperborea, Crete, and Hades; capture animals such as the
Cerynitian Hind, Pegasus, and Cerberus; kill monsters; fight in the greatest
wars of all time; and marry minor goddesses—or clouds. The two poets,
however, have a habit of breaking into the diegetic worlds of the myths that
they are narrating with interjections similar to those employed by the
narrators of memorates and modern ghost stories, in order to endorse the
stories they are telling and their own interpretations of what happened.
Often, they have what Bennett calls “recourse to religious principles” for
their endorsements. Thus, Bacchylides interrupts his story of Theseus’s
return from a thrilling visit to the undersea palace of his father with the
exclamation, “Nothing that the gods wish is beyond the belief of sane
mortals!” He cuts into the story of Croesus, who is rescued from certain

death on a pyre by a timely rainstorm sent by Zeus and then carried off by
Apollo to live amongst the Hyperboreans, with the affirmation, “Nothing
that the planning of the gods brings about is past belief!”124 Pindar
introduces the story of Bellerophon’s capture of Pegasus and his triumph
over the Chimaera and the Solymoi with the declaration, “The gods’ power
easily brings into being even what one would swear to be impossible and
beyond hope”; he concludes the story of Ixion’s intercourse with a cloud
(which eventually leads to the birth of the centaurs) with “The god
accomplishes every purpose just as he wishes, the god who overtakes the
winged eagle and surpasses the seagoing dolphin, and bows down many a
haughty mortal, while to others he grants ageless glory”; he follows the
story of Perseus’s beheading of Medusa and his use of that head to petrify
his enemies by insisting, “but to me no marvel, if the gods bring it about,
ever seems beyond belief”; and he interjects into the story of Coronis’s
infidelity and Apollo’s revenge, “The anger of Zeus’ children is no vain
thing.”125 Pindar is fond, as well, of endorsement by emphasis, using such
interjections as τοι or καί τοι or ἦρα καί (meaning something such as
“indeed!” “and surely!” or “verily!”) to assert the truth of a tale.126
At the same time as such interjections endorse these myths as accurate
accounts of the gods’ and heroes’ power, they remind us, by breaking into
the diegetic worlds of their stories, that the poets are inevitably standing at a
distance from what they are narrating. The same effect comes from many of
Pindar’s gnomic statements and from comments such as “Wonders are
many, but then, too, I think, in mortals’ talk stories are embellished beyond
the true account and deceive by means of elaborate lies. For Charis, who
fashions all things pleasant for mortals, by bestowing honor makes even
what is unbelievable often believed; yet days to come are the wisest
witnesses,”127 which is virtually a programmatic statement of Pindar’s view
of myths and their narrators. Even as these statements support Pindar’s
version of events and more generally his operational mode as a poet who
deals in myths, they remind us that he is mediating the myth for us. Inspired
or not, neither he nor any other poet can claim to have been an eyewitness
to the events they describe. And inspired or not (as Pindar concedes in the
lines that follow those I just quoted), a poet might choose to speak well of
the gods to escape divine blame and its concomitant retribution, even if
speaking well meant speaking less than the truth.128 Recognizing this invites

us to enter into a dialectical relationship with the poet and the story he tells,
to explore the nature of the experience that it describes, to try to understand
why and how it happened, what that might tell us about the gods and
eventually to decide to what degree we think the poet has portrayed the
experience accurately. If we do believe the poet’s story (or even only
portions of it), our commitment to believing it will be more strongly held
than it would have been, precisely because we have reached it on our own.
Pindar’s transitions are especially interesting in this respect. Pindar has a
habit of gliding seamlessly from the world of the victorious athlete whom
he has been hired to celebrate and into the world of the hero whose myth he
will narrate by using a single, seemingly inconsequential word such as a
particle or relative pronoun to pivot between the two. In his first Olympian,
for instance (from which the lines that I quoted in the previous paragraph
were taken), Pindar uses the relative pronoun τοῦ to move from a mention
of the Peloponnesus, where the athletic victory that he is celebrating had
taken place and which he characterizes as the “colony of Pelops,” into the
myth of the hero Pelops that will consume most of his poem’s remaining
lines.129 These pivoting transitions work well to collapse the distance
between the here-and-now of the victor’s achievement and the there-and-
then of the mythic hero’s accomplishments, which makes the hero seem
more like one of us, potentially, or at least more like the victor being
celebrated. Yet the line between the two worlds that is dissolved by such
devices is soon reestablished by an interjection or gnomic statement,
reminding us of our distance from the heroes and the fragility of our
knowledge about them.
The Messenger Speech from Euripides’s Hippolytus
Messenger speeches in Greek tragedies are a venue through which
remarkable things can be narrated—things that couldn’t, or shouldn’t, be
shown on the stage itself: the miraculous rejuvenation of Iolaus in
Euripides’s Heracleidae, or Heracles wrestling with Death in Euripides’s
Alcestis, or the mysterious disappearance of Oedipus into the grove of the
Semnai Theai in Sophocles’s Oedipus at Colonus.130 These speeches must
describe extraordinary eruptions of the divine into the human world in a
manner that makes them thrillingly real but also in such a way as to make
them credible for at least their internal audiences, the other characters in

these plays. About the external audience, sitting in the theater, I will say
something shortly.
I take as an example here the messenger speech from Euripides’s
Hippolytus, which describes the sudden appearance of a gigantic bull that
comes out of the sea, scaring Hippolytus’s horses so badly that they wreck
his chariot and gravely injure Hippolytus himself.131 The internal audience
of this speech is the members of the chorus and Hippolytus’s father,
Theseus.
The speech does a good deal of evidential scene setting through its
description of quotidian activities and local geography: the messenger tells
us about how he and other servants were combing their horses’ manes on
the shore when they were joined by Hippolytus, who had just been banished
from Athens by Theseus. Theseus had additionally called down a curse
upon his son, asking his own father, Poseidon, to kill Hippolytus.132
When Hippolytus arrived on the shore, the servants readied a chariot for
him and he mounted it: he put his feet into the driver’s rings, took the reins
from the rail into his hands, picked up the goad and touched the horses with
it. With the servants following on foot, he left town by a road that ran along
the shore of the Saronic Gulf, leading first to Argos and then to Epidaurus
—a road that would have been familiar to many of the men in Euripides’s
audience. The X of the X/Y format is set firmly in place, in other words.133
And then comes the Y: a terrible rumble was heard, deep within the earth,
which caused the horses to prick up their ears. “Violent fear came upon all
of us,” says the messenger. Looking towards the water, they saw an
enormous, uncanny wave blotting out the sight of Sciron’s coast, the
Isthmus and Asclepius’s Rock (once again, points of local geography that
would have been familiar both to the internal audience and to Euripides’s
audience). When the wave broke on the shore, it cast up a monstrous,
savage bull, whose bellowing filled the land. The horses panicked, and
Hippolytus tried to control them by pulling back on the reins, but they
refused to be steered by him, paying no heed to either their harnesses or the
chariot itself. Maddened with frenzy, they crashed, taking their driver along
with them. And then, all was utter confusion. Axles and linchpins flew into
the air, and Hippolytus, tangled in the horses’ reins, was dragged along the
road, his head bouncing on the rocks.134

This is a vivid scene that works to firmly anchor the extraordinary within
the ordinary; the detail of axles and linchpins flying through the air in the
very midst of the supernatural crisis is especially effective in this regard. By
the standards laid out by Wooffitt, Bennett, and Hänninen, the messenger
would have had a high chance of persuading his immediate listeners that
what he described had really happened (and, in any case, the arrival of the
dying Hippolytus soon after the messenger finished speaking, carried on a
litter, would have been an extraordinarily effective instance of Bennett’s
call to witness). But what about the other listeners, sitting in the theater, just
across the gulf from where Hippolytus’s accident occurred? They probably
did not worry that exactly such a thing could happen to them—they were
not the sons or grandsons of gods after all, as Theseus and Hippolytus were
—but the scene would have been a formidable representation of an incident
that was accepted as part of Athenian history.135 This had happened to their
ancestors. This was what the gods, and the heroes, were capable of.
Powerful forces had once been at play in the cosmos, and there was no
reason to doubt that they still were. To adapt M. R. James’s formula, if they
weren’t very careful, something awful (albeit less spectacular) might also
happen to them.
A Final Question
All in all, then, myths as they were publicly disseminated in Greece,
through the skillful performance of highly engaging narratives, were ideal
partners for festivals because the myths helped to sustain the basic premises
on which the festivals depended: that the gods and heroes existed and that
they might choose to interact with mortals, for better or for worse. Myths
were particularly able to do this, I have suggested, because, amongst other
things, they were narrated episodically, which encouraged individuals in the
myths’ audiences to develop closer, more personal engagements with their
characters, and because the poets used techniques that were conducive to
belief, such as Wooffitt’s X/Y format and the call to witness, evidential
scene setting, ambiguity, endorsements and justifications that were
described by Bennett and Hänninen. These features, as well as some that
are more familiar from literary studies of Greek literature, such as deixis
and apostrophe, and the habit of performing the narratives at ritually
important times and places, imbued the myths with what Landy would call
a formative effect. They conditioned the audience members to experience

the gods and heroes as a daily reality and to forge enduring relationships
with them.
There is one more issue that is central to our understanding of how Greek
myths worked in concert with festivals, which I have left dangling up till
now. Granted that well-narrated myths about Theseus, for example, helped
to sustain belief in his existence and his availability to respond to
worshippers, how does that help us to understand the contribution that such
a myth made when it was narrated, for instance, at a Delian festival in
honor of Apollo (as was Bacchylides’s seventeenth ode)? Traditionally, such
questions have been answered contextually or thematically when they could
not be answered aitiologically: narrating a story about Theseus in honor of
Delian Apollo, for example, has been said to make sense because Delos, the
home of the god, was presented in myth as one of the places where Theseus
and his friends stopped on their victorious way home from Crete and
because Apollo was generally regarded as the divine protector of young
men. This is certainly a valid approach, but it risks a potential circularity.
Bacchylides’s eighteenth ode, which describes Theseus’s journey from
Troezen to Athens from the point of view of his father, Aegeus, is assumed
by scholars to have been written for performance by Athenians simply
because it concerns the royal family of their own city. What if our
manuscripts did not tell us that the seventeenth ode was composed for the
Ceans to perform on Delos? Would we invent a different way for it to
“make sense,” perhaps assuming that it, too, was meant for the Athenians,
and for performance in Athens, simply because it concerned Theseus and a
group of other young Athenians? Or what about cases where even our most
imaginative scholarly logic can’t contrive a connection, such as the brief
dithyramb that Bacchylides composed to be sung at Delphi during the
winter months when Dionysus, not Apollo, was in charge there?136 How
does the story that Bacchylides tells us, about Heracles’s accidental murder
at Deianira’s hands, help sustain belief in anyone other than Heracles
himself? And how (famously) could so many tragedies get away with
having nothing to do with Dionysus, the god at whose festivals they were
originally performed?137
Context and theme (as well as the personal tastes of the poets and their
patrons) certainly could matter when a poet chose a myth to narrate, but
they cannot answer the question of why, seemingly, a wide variety of myths
were appropriate for narration at a wide variety of festivals. For this, we

have to think about myths not in the individual sense but as a whole—as a
corpus. This is what I will take up in Chapter 4.

4
The Greek Mythic Story World
IN CHAPTER 3, I took up the question of how the highly polished nature of

Greek mythic narratives—the vivacity and expressive power that earned so
many of them an enduring place in the pleroma of world literature and art—
contributed to the creation and sustenance of belief in the gods, the heroes,
and a divine world more generally. I focused particularly on how Greek
myths evoked emotional and cognitive responses from their audiences that
were virtually indistinguishable from those evoked by real-life people and
situations and on how the ancient modes of narrating myths helped to keep
the stories and their characters alive in an audience member’s mind and
heart long after a narration was over, thus further sustaining the beliefs that
the stories had nurtured.
One question that I temporarily set aside was why the narration of almost
any myth, focusing on almost any character, was appropriate for recitation
at almost any festival dedicated to almost any god—even when there was
no obvious aitiological, thematic or contextual link between the myth and
the festival. The general answer must be that the Greeks cared less about
always making tightly “logical” connections between festivals and myths
than we have imagined—or, to put it otherwise, that the contributions that
mythic narratives made to creating and sustaining belief in the gods and
heroes must have been more broadly based than we have previously
acknowledged. More specifically, I want to suggest that an important
element enabling this breadth of applicability was the tightly woven story
world that was cumulatively created, on a continuous basis, by the myths
that were narrated. This story world validated not only each individual myth
that was part of it, but also ta palaia more generally—all the stories about
what had happened in the mythic past, the characters who had lived then
and the entire worldview on which the stories rested. Because a myth told
about Heracles, for example, was deeply embedded in this story world, it

had the power to sustain and enhance belief not only in Heracles himself
but also in the entire cadre of the divine world and the possibility of their
interaction with mortals, if it were well narrated.
I’ll begin with a discussion of what makes story worlds in general
coherent and credible and will then move on to ask whether the story world
created by Greek myths fulfilled those criteria or operated in a different
manner. Along the way, I’ll discuss some characteristics that Greek mythic
narratives share with narratives that are familiar from our own times, which
will further heighten our appreciation of the way that the Greek mythic
story world created and sustained belief. Of course, an important backdrop
to my project as a whole, as sketched in this chapter and the preceding one,
is the fact that the places and times in which Greek mythic narrations were
frequently performed were in themselves conducive to creating and
sustaining belief—that is, the narrations were performed in sanctuaries
dedicated to the gods and heroes, during festivals dedicated to the gods and
heroes. The audiences were primed by these conditions to open their minds
to the ideas that the myths conveyed.
Story Worlds
One of the first scholars to theorize about how story worlds are created was
himself the creator of a very famous story world: J. R. R. Tolkien. In an
essay titled “On Fairy-Stories,” which was delivered in 1939 as the annual
Andrew Lang Lecture at the University of St. Andrews and then published
in 1947 in a Festschrift for the fantasy writer Charles Williams that was
edited by C. S. Lewis, Tolkien introduced the term “Secondary World,”
which he contrasted with “Primary World,” the world in which we live. The
virtue of these terms, he suggested, is that they allow us to avoid the terms
“real world” and “fantasy world,” which obstructed an idea that Tolkien and
a number of other authors and scholars wanted to emphasize (including
Lewis, in his own essay in that collection),1 namely, that a well-constructed
fictional world elicits responses from us that are almost indistinguishable
from the ways in which the real world affects us, even if that fictional world
has fantastic elements. Samuel Taylor Coleridge’s well-worn phrase
“willing suspension of disbelief,” which he invented in 1817 to discuss
poems that included elements of the fantastic, such as his own Rime of the
Ancient Mariner,2 misses the mark completely as far as what well-

constructed fiction does to readers, according to Tolkien. What really
happens, he says, is that an author makes
a Secondary World which your mind can enter. Inside it, what he
relates is “true”: it accords with the laws of that world. You therefore
believe it while you are, as it were, inside. The moment disbelief
arises, the spell is broken; the magic, or rather art, has failed. You are
then out in the Primary World again, looking at the abortive little
Secondary World from outside.… [T]hen disbelief must be suspended.
But this suspension of disbelief is a substitute for the genuine thing.3
The media scholar Michael Saler has recently coined the term “willing
activation of pretense” to describe this process that Tolkien discussed,4 but I
think that Tolkien meant something more than that: a truly well-constructed
story world requires no conscious decision at all on the part of audience
members who participate in it—neither the suspension of disbelief nor the
activation of pretense. It immerses readers or viewers so completely and yet
so subtly that they pass into it without even noticing that they are doing so.
In addition to credibility, a Secondary World requires something else. As
the media scholar Mark Wolf puts it, what it needs is a
distinct border partitioning it from the Primary World, even when it is

said to exist somewhere in the Primary World, or when the Primary
World is said to be part of it, as in the case of the Star Trek universe
containing earth. It is connected to the Primary World in some way but
also set apart enough to be a world unto itself.5
“Partitioning with a distinct border” can refer to devices such as wardrobe

doors, rabbit holes, deadly deserts over which cyclones carry houses and
distances that only warp-speed vehicles can traverse, but it also refers to
giving the Secondary World a sufficient number of sufficiently striking
features—geographical, botanical, zoological, technological, and so on—
that make it different from the Primary World. For example, the world in
which the Oz stories are set has Munchkins, flying monkeys, kalidahs,
talking trees, mangaboos, and a queen who can change her head as easily
other women change their hats (to name just a few of its oddities).6 Wolf

argues, in fact, that one can’t really be said to have created a truly
Secondary World if one simply introduces a single element—vampires or
space aliens, for example—into a world that in all other respects aligns with
our Primary World.7 One might nonetheless have created a gripping story—
after all, Bram Stoker’s Dracula essentially does what I just described: it
introduces vampires into a world that is otherwise identical to that of late
nineteenth-century Europe. One might even make such a story coherent and
credible—Stoker did this particularly through the introduction of diary
entries, letters written by one character to another and newspaper clippings,
for example. But it is doubtful that the original readers of Dracula, who
were, after all, the same sorts of Victorian folk who populated the novel,
felt that they had entered a wholly new story world in the same way that
readers of the Oz books or the viewers of the Star Wars movies do.
That takes care of the word “Secondary” in Tolkien’s term: a Secondary
World must be different enough from the Primary World to merit the
adjective. But what about the word “World”? When we talk about a story
world, we typically mean something that goes beyond the narratively
constructed space in which a single story is set, something that constitutes a
space where many stories, whether they be directly connected to each other
or not, can be set and that is perceived by its audience as consistent and
coherent. The story world of Oz, which slowly developed over the course of
sixty-plus volumes, is again a good example: it includes not only the
country of Oz itself but also some neighboring lands, such as Ev and Ix,
which were explored both in some of the later Oz books and also in books
that formally lay outside the Oz corpus. Characters transfer from one book
to others, sometimes traveling from one country within the Oz world to
another.8 Tolkien’s story world, including the land called Middle-Earth, is
the setting not only for The Hobbit and The Lord of the Rings but also for
stories narrated in the Silmarillion, which take place in lands contingent to
Middle-Earth, all of which are contained in a world called Arda, which is
itself part of a cosmos called Eä. Some of these stories, especially those
from the final section of the Silmarillion, serve as prequels to The Lord of
the Rings. There are story worlds outside of fantasy and science fiction, too.
William Faulkner’s Yoknapatawpha County, in which twenty-two of his
novels are set, qualifies as one, for example, as does Garrison Keillor’s
Lake Woebegone. These latter two worlds are not distinctively secondary—
that is, they don’t differ significantly from the Primary World in their

geographic, botanical, zoological, and technological features—but they are
well enough described to hold together as discrete places over the span of
the different stories. You can even map them, as Faulkner himself did.9 (In
fact, the creation of maps and other paratextual materials to accompany a
series of stories—lexicons of invented languages, timelines, genealogies,
bestiaries, and other guides to its infrastructure10—is a sure indication that a
story world has taken on an existence of its own.)
With these two sets of guidelines in place—what it takes to make a story
world a world and what it takes to make it secondary—let us turn now to
Greek myths and consider what kind of story world underlies them.
The Greek Story World

My first question is whether we find strong characteristics of a Secondary
World in Greek myths. Are there an adequate number of sufficiently odd
geographic, botanical, zoological, and technological features and the like to
qualify? Let us take stock—trying to be representatively inclusive without
becoming pedantically exhaustive. In our bestiary, we have some
remarkable creatures: Gorgons, griffins, harpies, sphinxes, centaurs,
Cyclopes, Sirens, Typhon, the Hydra, Scylla, Pegasus, the Chimaera, the
Minotaur, a couple of dragons and Cerberus. But many of the other
creatures who appear in myths are just larger or more vicious versions of
creatures found in the Primary World: the Nemean Lion, the Erythmanthian
and Calydonian Boars, the Crommyonian Sow, the Mares of Diomedes, the
Teumessian Vixen, the Cretan and Marathonian Bulls, Python, and Sciron’s
giant turtle.11
Botanically, there is very little: from the Odyssey, we get moly and
lotuses, and Epimenides mentions a plant called alimos (“banishing
hunger”) that kept Heracles from starving while in the desert,12 but I can’t
think of other examples. Technologically, we have the various automata
built by Hephaestus, Perseus’s (or rather Hermes’s) winged shoes and
Hades’s cap of invisibility. As for unusual geographic features, I can offer
only the Symplegades and Charybdis.
Aside from the zoological catalogue, this is not a particularly strong
record of weirdness. And we must remember, moreover, that the catalogue I
just presented is a pasticcio, gathered together from many different works.
In single episodes, Greek myths usually introduce only one remarkable

feature—a monster, most often. This amounts to the same thing as a
seriatim introduction of vampires and aliens into an otherwise Primary
World. Usually, moreover, ancient authors seem relatively uninterested in a
monster’s remarkable features, describing them only briefly. Consider this
passage from the Iliad, for example, in which the story of Bellerophon and
the Chimaera is told:
So off went Bellerophon to Lycia, under the excellent escort of the

gods. And when he reached the river Xanthus, the king welcomed him
and honored him with entertainment for nine solid days, killing an ox
each day. But when the tenth dawn spread her rosy light, the king
questioned him and asked to see the tokens that his son-in-law Proetus
had sent. And when he saw the evil tokens, he ordered Bellerophon to
kill the furious Chimaera, a creature that was not human but divine; a
lion in front, a serpent in the rear and a goat in the middle, and
breathing fire. Bellerophon killed her, trusting in signs from the gods.13
Homer doesn’t fail to mention the Chimaera’s triple physiognomy and fiery
breath, but he does not choose to take full advantage of their narrative
possibilities (nor does Pindar, for example, who allots to her a single
adjectival phrase, “fire-breathing”).14 Similarly, in Sophocles’s Trachiniae,
Deianira, quoting the centaur Nessus, describes the Hydra only as “the
monstrous beast of Lerna”—no mention even of the Hydra’s nine heads,
much less their ability to duplicate themselves when cut off. The chorus of
Aeschylus’s Agamemnon refers to Scylla simply as “dwelling in the rocks, a
harmful thing for mariners,” and Euripides’s Medea describes her as just “a
dweller on the Tuscan cliff.”15 This is not to say that Greek authors could
not focus more closely on the wondrous or the horrible when they wished
to: Aeschylus’s portrait of the unnamed goddesses who invade Apollo’s
Delphic Oracle, for example, is detailed and chilling: black and utterly
abominable, they snore forth repulsive breath and foul ooze drips from their
eyes. But that is the point, after all: the epiphany of these disgusting
goddesses is meant to be understood as a sudden irruption of maddened rot
into a space of reasoned light; the contrast has to be striking. Similarly, if
for different reasons, Odysseus’s own narrations of the monsters he
encountered and their dreadful modes of attack are hair-raising—but what

else would we expect from such a talented, self-aggrandizing storyteller?16
Hesiod’s description of Typhon, the king of all monsters and the greatest
threat to Zeus’s world order, which constitutes the topic of Hesiod’s poem,
also goes on at some length: from Typhon’s shoulders sprang a hundred
serpents’ heads, all licking with their dark tongues; fire shot forth from his
eyes and terrifying voices echoed from his many mouths, mimicking gods,
bull, lions, puppies, and horrible hissings. The battle between the two is
prodigious and vividly described.17
But generally, the monstrous and the marvelous are treated with restraint
by ancient authors; the one big exception is metamorphosis (the
transformation of a person into an animal, plant or mineral), which happens
frequently in Greek myths and which I’ll treat in Chapter 6.18 Overall, the
effect of introducing the monstrous and the marvelous into Greek myths is
much like that produced in the modern genre of magic realism, where
elements that, in isolation from their narratives, might stand out as magical
or fantastic—ghosts conversing with the living, telepathy, and extraordinary
longevity, for instance—are integrated into the everyday world in such a
way as to be accepted by audience members. This does not rob them of
their marvelousness; rather, it enables them to contribute to an expansion of
the narrative world’s possibilities.19 We should note the contrast between
this way of narrating the marvelous and that found in many examples of
fantasy writing, in which remarkable elements are accentuated by vivid
description, such as this passage from Neil Gaiman’s Neverwhere:
Upon [Hunter’s] arrival, it comes through the underbrush, a fury of

brown and of white, undulating gently, like a wet-furred snake, its red
eyes bright and peering through the darkness, its teeth like needles, a
carnivore and a killer. The creature is extinct in the world above. It
weighs almost three hundred pounds and is a little over fifteen feet
long, from the tip of its nose to the tip of its tail. As it passes her, it
hisses like a snake and, momentarily, old instincts kicking in, it
freezes. And then it leaps at her, nothing but hate and sharp teeth.20
This giant weasel creature enhances the strangeness of a subterranean world

that Gaiman has already painted as being weird, dangerous and utterly
disconnected with “normal” existence above ground.

My initial conclusions, then, are that the story world of Greek myths is
not a strongly secondary one, that the secondary qualities that it does
possess focus upon single events or single characters, and that those events
or characters are often integrated into descriptions of the Primary World in
such a way as to expand the possibilities of the latter, rather than to
highlight the extraordinariness of the former. This isn’t because the Greeks
weren’t able to envision or create what we would call more truly Secondary
Worlds, according to Tolkien’s standards: some examples that prove they
could are Lucian’s True History, the latter parts of the Alexander Romance,
those portions of the old Argonautica that are set in Colchis or on the
journeys to and from Colchis, and books 9 through 12 of the Odyssey.
The latter two instances are particularly interesting because their
reception histories once again demonstrate the Greek preference for keeping
their story world closer in nature to the Primary World than to a Secondary
World. Over time, the fantastical lands that Jason and Odysseus visited
began to be pulled back into the category of the unremarkable, as the
Greeks repeatedly tried to map them onto the world that they knew.
Phaeacia became Corcyra, for example, the Symplegades became the straits
of the Bosphorus, and the island of the Cyclopes became Posillipo. We
should note as well that, leaving aside these travels of Odysseus and Jason,
when Greek heroes came from or journeyed to exotic places, they typically
were places that nonetheless had a firm location in the world as the Greeks
knew it: Cadmus and Europa came from Phoenicia, Theseus went to Crete,
Bellerophon journeyed to Lycia and Iphigenia ended up amongst the
Taurians in Scythia, for example. Perseus and Heracles traveled to more
fantastical lands for some of their labors—each of them went to the Garden
of the Hesperides, for example (see Figure 4.1 for Heracles’s trip there)21—
but “Hesperides,” after all, is nonetheless a firmly geographic name: even if
no one who told or listened to these myths had themselves been there, the
very word “Hesperides” (Westworld) insisted that the heroes had gone west,
and thereby kept them within a world articulated by familiar cardinal
points. The name of the longest-running fantasy land that the Greeks
invented—Hyperborea (Beyond-the-North)—reveals a similar
determination to keep exotic realms tethered to the known world, even if on
a longer leash. Perhaps, as Pindar once said, you couldn’t get to Hyperborea
by ordinary means such as walking or sailing, but you could pin
Hyperborea to a spatial relationship that was essentially no different from

that to which you would pin Oropus if you lived in Athens. If you found the
right mode of transportation and just kept going north, you’d eventually
arrive there. Perseus and Heracles each managed to do so, after all.22
4.1 Heracles approaches the Garden of the Hesperides in the cup of Helios. Atlas and
one of the Hesperides look on; Athena and Hermes flank the scene (further on the
image, see Chapter 4, note 21). Attic red-figure bell krater by the Nikias Painter, dated
to between 420 and 400 BCE (detail). Now in a private collection. Photo credit:
Christie’s Images / Bridgeman Images.
Greek myths, then, clung to familiar geographic templates, familiar

botany and familiar technology, and they limited the degree of strangeness
to which their monsters could aspire. They do not constitute a clearly
Secondary World by Tolkien’s terms. Nonetheless, I suspect that many of us
feel it in our bones that Greek myths do, collectively, have a distinct story
world. One reason for this is that we grew up surrounded by books that
corral individual myths into anthologies. Translated or renarrated by the
voice of a single author, illustrated by the pen or brush of a single artist and
then bound between two covers, the myths are perforce given cohesiveness.
The practice has a long history: Ovid and Apollodorus were already doing
it, as were predecessors such as Pherecydes and Hellanicus. Centuries of
European art cemented the idea, not only insofar as artists took particular
pleasure in illustrating Ovid, the greatest of all unifiers, but also insofar as

there was an implicit agreement that Greek myths—virtually any Greek
myths, whether they came through Ovid or not—were, along with stories
from the Bible, amongst the few appropriate subjects for upper-class décor.
Walking today through a museum gallery within which such works have
been gathered together is like strolling through a mythographic handbook.23
The Mythic Network

But another, and more important, reason that the Greeks seem to have a
mythic story world is that the gods, heroes and monsters whom we meet in
the individual stories are always part of a network.24 There is no such thing
as a Greek mythic character who stands completely on his or her own; he or
she is always related to characters from other myths, and the narrators take
some pains to tell us that (and, one assumes, to invent such relationships
when they need to). Python may have been a new monster to some people
when they heard the Homeric Hymn to Apollo, but the poet ties her into the
larger family of mythic monsters by mentioning that she had been the
nursemaid of Typhon, a dreadful creature about whom Hesiod had a lot to
say. And the poet makes Apollo himself tell us, a few lines later, that
Python was a pal of the Chimaera—who first appeared in the Iliad and
whom Hesiod said was the child of Typhon (as were Cerberus and the
Hydra).25
The branches of heroic family trees are at least as entangled as those of
the creatures that they kill or subdue: Heracles was the descendant of
Perseus; by killing a Gorgon, Perseus played midwife to the marvelous
horse Pegasus upon which Bellerophon rode to kill the Chimaera; and
Bellerophon was the ancestor of the Trojan ally Glaucus—who was the first
to narrate the story of Bellerophon and the Chimaera.26 Cadmus,
meanwhile, searched for his sister Europa, who had been kidnapped by
Zeus to start a dynasty on Crete—which eventually led to the birth of the
Minotaur, whom Theseus killed.27 Theseus journeyed to the Underworld to
help his friend Pirithous kidnap Persephone—a favor in return for Pirithous
having helped Theseus to kidnap Helen (who was later kidnapped again by
Paris, thus starting the Trojan War, in which Glaucus fought). Failing in this
quest and trapped in the Underworld, Theseus was rescued by Heracles,
who while he was down below also met the ghost of Meleager, whose sister
Heracles later married … by whom he subsequently was murdered, with the

result that Odysseus met Heracles’s ghost in the Underworld.28 To get back
to Cadmus: teeth from a dragon that he had slain while looking for his sister
were carried to Colchis, where they later caused problems for Jason, who
was aided by Medea, who later tried to kill Theseus and subsequently
married Achilles after they both were sent to the paradisiacal White Island,
where the elite of the heroic race hung around after their lives were over.29
Gods were notoriously intertwined with one another, too, not only in the
sense that, during those early epochs of creation, there was no one else with
whom they could dally and reproduce but in other ways as well: Aphrodite
fought with Persephone over Adonis, who had been killed by a boar sent by
Artemis, who was angry with Aphrodite for having contrived the death of
Hippolytus, the son of Theseus.30 Hera wheedled Aphrodite into lending her
a magic charm under the pretense that she wanted to settle a quarrel
between their grandparents, Tethys and Oceanus, and then used that charm
to seduce Zeus, so that Poseidon could cause trouble on the battlefield while
Zeus was taking a postcoital nap. Hermes stole his big brother Apollo’s
cattle and then won forgiveness by singing a song about how the whole
family of gods had come into existence in the first place.31 Gods were
intertwined with monsters and heroes as well—as their parents and lovers,
impeders and helpers. Athena helped Bellerophon put a bridle on Pegasus
so that he could kill the Chimaera. Zeus made love to Io (thus becoming the
great-great-grandfather of Cadmus and Europa, to the latter of whom he
eventually also made love), and as a favor to Zeus, Hermes killed the one-
hundred-eyed herdsman Argus, whom Hera had sent to guard Io.32
Those last three paragraphs were far easier for me to compose than
anything else I have written in my scholarly career. As a child, I read almost
nothing except Greek mythology books, and by the time I was ten, I was an
expert at my own private version of “Six Degrees of Kevin Bacon”: if you
gave me two figures from myth, I could find a way to link them together
(and in fewer than six steps, usually). Such mental gymnastics have a long
and respectable pedigree: starting already in late antiquity, scholiasts and
mythographers reported fully on the loves, hates, and other connections
amongst gods, heroes, and monsters. Earlier still, Ovid sewed his
Metamorphoses together with the threads of such relationships, some of
which he may have invented, but still, they sounded real. He said, for
example, that Arachne was a girlhood friend of Niobe—and if readers are

mythically well informed, they get frissons when they read that line,
knowing that the two girls are also linked by a shared hubris that leads them
to sorry ends.33 The Hellenistic poets played such complex versions of this
game that even my precocious talents were at times confounded. Whom did
the poet Lycophron mean when he referred to the “five-times-married
frenzied descendant of Pleuron?” Helen, of course, as A. W. Mair’s
footnotes to the Loeb edition so kindly informed me: the great-great-
granddaughter of Pleuron, Helen was carried off by a smitten Theseus when
she was a girl, was properly married to Menelaus when she was a bit older,
was kidnapped again by Paris and lived as his wife for more than nine
years, was married to Deiphobus after Paris’s death and then, finally, after
the end of her earthly life, was married to Achilles, on the White Island.
(But wait—didn’t I just say that Achilles married Medea on the White
Island? Well, Lycophron mentions that, eventually, too.)34
But the weaving of the net with which these later poets played their
games began much earlier, and it was not meant exclusively for the erudite.
The Hesiodic Catalogue of Women includes such morsels of information as
the fact that the daughters of Pelops and Hippodamia married the sons of
Perseus and Andromeda. It’s from one of Bacchylides’s epinician odes that
we first learn about Heracles meeting Meleager in the Underworld, and it’s
from the end of Euripides’s Hippolytus that we first learn about the tangled
web of relationships amongst Artemis, Aphrodite, Adonis, and
Hippolytus.35 Those who wished to create new myths, moreover,
understood the importance of tying their threads securely into the existing
tapestry. When the ritual bricoleurs behind the new mysteries of Dionysus
decided, at some point in the late Archaic period, that they needed an aition
for their cult, they created what was essentially a sequel to the well-known
story of Demeter and Persephone, turning the famously kidnapped daughter
into the grieving mother of a kidnapped child herself.36 Theseus, who seems
to have been a fairly low-key hero until the sixth century,37 was groomed for
the big time by being linked to such stars as Medea (who became his evil
stepmother) and Heracles (who became his rescuer), as we have seen.38
Even Orpheus, a character who in most ways stood aloof from the great
mythic families and from the mythological network itself (and properly so;
singers are by nature marginal figures in Greek society), was drawn into
service as an Argonaut.39 Indeed, the voyage of the Argo more generally is a

perfect example of a story that grew through a type of agglomeration
encouraged by the mythic network: over time, almost every hero of the
generation that lived before the Trojan War (including one woman), ended
up on that ship. The Calydonian boar hunt is a similar case (and it recruited
the same woman—but then, there weren’t more than handful of heroic
women to choose from).40
Studying this network for its own sake is pretty boring. Few people read
Timothy Gantz’s Early Greek Myth straight through from page 1 to page
743, valuable though it be as a reference work, and few, I suspect, actually
study those schematized wall charts that trace every mythic character back
to Chaos in exquisite detail. Until we hear or see the stories behind the
relationships and meet the personalities connected with the names, the
relationships themselves just aren’t very interesting. Yet when embedded in
myths, the relationships not only come to life but also help to create a
coherent story world that in turn serves to anchor and validate each
individual myth, in an infinitely reciprocal way. That is, effective narration
of Theseus’s adventure with the Minotaur also lends credibility to Perseus’s
defeat of snaky-haired Medusa or to the birth of Erichthonius as a creature
half snake and half human, because all of these hybrid figures are presented
as inhabiting the same realm—a realm that is thickly crisscrossed by the
relationships that I’ve been talking about. Each story stands as a guarantor
of the existential rules underlying the others and is in turn guaranteed by
them. Each of them contributes to a completely furnished world, from
which audience members may subsequently break off pieces to use as a
situation demands—pieces that still refract the authority and allure of the
whole.41 This is one reason that figures and incidents from myths are so
powerful as symbols: even when we regard them singly, they are never
actually alone.
In a sense, what I am describing is a sort of hyperseriality: that is, an
extended version of the seriality common to Greek myths that I began to
describe in Chapter 3 and will discuss again in Chapter 7. In simple
seriality, the tale of Odysseus and Polyphemus, for instance, is but one
episode in a closely linked set of tales narrating Odysseus’s travels home,
but in hyperseriality, any individual’s cumulative story is also, implicitly, an
episode within a far larger story that stretches “vertically” from the very
first gods (Gaea and her shadowy siblings, if you believe Hesiod; Phanes or
Protogonos, if you believe Orpheus) through to the children of the heroes

who fought at Troy, and stretches “horizontally” to connect each individual
to characters in other mythic families or groups (thus, Odysseus, for
example, is connected vertically to Hermes, through his grandfather
Autolycus,42 and horizontally to other warriors who fought at Troy and to
the strange people he met on his way home). This sort of hyperseriality is
very familiar from soap operas (how many generations of Quartermaines
have marched through the corridors of General Hospital and with how
many generations of Spencers have they fought and made love?), but we
find it in plenty of “higher” cultural products as well: John Galsworthy’s
The Forsyte Chronicles is an example, as is C. P. Snow’s Strangers and
Brothers series. Then there are the hyperserials whose position between
high and low depends on whom you ask: Stephen King’s popular Dark
Tower series of novels and the other novels and stories with which King
intertwines it, for example, and Robert Graves’s I, Claudius and Claudius
the God (the fact that the Julio-Claudians really existed does not exempt
them from being prime fodder for the sort of hyperseriality that I am talking
about). Within such a hyperserial, an individual’s story can be enjoyed on
its own, but it is more resonant, credible and just plain interesting as part of
the bigger picture that is always shimmering behind it. Our knowledge that
Fleur Forsyte’s father had once been married to Jon Forsyte’s mother—
which Fleur and Jon do not know when they fall in love with each other in
the third book of the Forsyte Chronicles—gives their relationship a pathos
it would not otherwise have. Similarly, knowing that Deianira will later
murder Heracles gives the scene at the end of Bacchylides’s fifth ode,
where Meleager’s ghost promises Heracles her hand in marriage, an extra
pathos, too.
Some hyperserials come into existence when someone (it may be the
original author) gloms on to a good thing and expands it by creating
sequels, prequels, midquels (previously unnarrated episodes from the
middles of established stories), and paraquels (previously unnarrated stories
that take place at the same time as an established story but that focus on
different characters, some of whom may have already appeared in more
minor roles in the established story).43 These -quels are not necessarily poor
relations to the “original.” The Horse and His Boy (a paraquel to C. S.
Lewis’s The Lion, the Witch and the Wardrobe) can stand completely on its
own as an adventure story. Each of the eleven novels in Snow’s Strangers
and Brothers (which span five decades in the life of their narrator, Lewis

Eliot) stands as a paraquel to some or all of the ten other books
(confounding the very idea that there is an “original” story amongst them)
and draws from a common cast of characters. And yet The Light and the
Dark (which falls fourth in the series’s internal chronology) immerses us so
completely in Roy Calvert’s development into one of the first scholars of
Manichaeism that we may be surprised, in chapter 4, to be reminded of how
Calvert’s irresponsibly homoerotic behavior (an episode for which he is
brought briefly on stage near the start of the second book in the series) once
wreaked havoc in the lives of a number of Lewis Eliot’s other friends. In
fact, this kind of movement of a character back and forth between
prominence and unimportance (in one novel, Calvert is merely the match
that lights the fuse of a story from which he then departs; in another, he
stays at the story’s heart) is essential to hyperseriality as I understand it
here: characters fade in and out of one another’s stories in a manner that
begins to dissolve any single story into something much larger (“Lewis
Eliot’s recollections of how interactions amongst his friends affected the
course of academia, politics, and science in twentieth-century Britain”).
Sometimes it can be difficult to decide whom a given story is even “about.”
Is To Let, the third book of The Forsyte Chronicles, still about Soames
Forsyte, on whom the first two books centered, but more specifically about
Soames’s problems with his daughter, Fleur? Or is To Let about Fleur and
her problems with Soames? Or to return now to Greek myths, is the tale of
a certain love affair about Jason, who according to the tale’s internal
chronology already had begun his career as a hero, or is it really about
Medea, whose career as an enchantress and all-around “bad woman” would
continue well beyond the demise of her relationship with Jason? Does
Pegasus really “belong” to the story of Medusa, only being “borrowed” by
the story of Bellerophon—or vice versa? Is a certain story set in the
Caucasus about Heracles releasing Prometheus, or is it about Prometheus
being released by Heracles? Who is incidental to whom?
Crossovers
The answers to these questions (which depend in part on who is telling the
stories and who is hearing them) are irrelevant for my purposes, and in the
abstract, they matter very little anyway, except to people such as
Apollodorus, who strove to organize the stories he told according to a
single, coherent plan that avoided repetition. Apollodorus abruptly

truncated the story of Prometheus, for example, at the point where the
fettered god had been bound to a peak in the Caucasus mountains, wrapping
up that part of his mythology with the words, “This was the penalty
Prometheus paid for stealing fire until Heracles later released him—as we
shall show in the chapters concerning Heracles.”
In contrast to Apollodorus, I want to highlight the dense intertwining of
characters and their stories in ancient narratives and the concomitant
difficulty of completely disengaging any single character from the larger
network of which he or she is a part. Although my description of the mythic
network in the previous section was necessarily linear—verbal
communication is always linear—the Greek mythic hyperserial, like all
hyperserials, stretched in many directions at once, entangling each of its
participants with many others. Chart One (shown in Figure 4.2), which
diagrams all the relationships that I mentioned in the three paragraphs with
which the last section began, illustrates what I mean: the lines connecting
one character to another cross thickly. Chart Two (shown in Figure 4.3),
which adds lines that represent relationships amongst those characters that I
could not easily insert into my linear narrative, demonstrates this point even
better. Popular characters, such as Zeus and Heracles, have links to many
other characters, but even Pegasus has several links. A few paragraphs ago,
I proposed that such intertwining lends credibility to the stories in which
these characters participate simply because they all are understood to
inhabit the same expansive and yet bounded story world—each story
guarantees and in turn is guaranteed by the others—but I am further
proposing now that it is precisely this intertwining of characters and their
stories that cumulatively constitutes the story world of Greek myth.

4.2 Chart One.

4.3 Chart Two.
Intertwining brings us to another, more specific, and often deliberately

deployed, technique to which I want to draw attention as well, namely the
“crossover,” that is, the appearance of a character who is familiar from one
context in the middle of another.44 A character may be crossed into a new
context by his or her own creator: twenty-nine years after introducing
Father Donald Callahan as a main character in ’Salem’s Lot, Stephen King
crossed him over into the final three books of The Dark Tower, where the
priest’s deep knowledge of evil and the shakiness of his Catholic faith
played out well against different monsters in a different universe.
Crossovers also occur when a narrator borrows someone else’s character
and plants him or her in a new context: Sherlock Holmes, not surprisingly,
is one of the most frequently borrowed (or, as the Doyle estate insists,
“stolen”) characters of all. And then there is a mode of crossing over that
works on the all-star principle, bringing many established characters
together into a single new venue. Alan Moore and Kevin O’Neill’s League
of Extraordinary Gentlemen is probably the best-known all-star example

from recent years. I am not the first to suggest that the Argonautica is the
grandfather of them all. The Calydonian boar hunt is another.45
The line between crossovers and what I have been calling hyperseriality
can be exceedingly slender, but I would like to pause on one difference that
is important for my purposes. As I emphasized, hyperserials, either by
authorial intention or through evolution, tend to obscure the priority of any
“original” or “dominant” narrative and its characters (and in fact, the last
type of crossover I just described, the all-star gathering, may better be
understood as a special form of hyperseriality for this very reason, given
that all of the borrowed characters typically meet on narrative ground that is
new to some or all of them). The use of crossovers, in contrast, thrives, at
least initially, on the unexpected introduction of someone from outside. The
surprise of meeting a familiar character where we don’t expect to generates
an additional level of interest in the story. If we are right in guessing (along
with Pausanias)46 that it was Sophocles who pioneered the story of
Oedipus’s death and burial at Athens,47 then this is a perfect example; the
ancient audience of the Oedipus in Colonus would have been intrigued to
discover the hero wandering in an Athenian suburb. Crossovers may also
reward audience members with a sense of having special knowledge, which
makes them feel complicit with the narrator and thus further encourages
them to buy into the narrative—somewhat like the “Easter Eggs” that
contemporary viewers spot in movies and television shows. A good
example is Heracles’s meeting Meleager’s ghost in the Underworld. The
unexpected encounter between the two in and of itself rewarded listeners
who were familiar with mythological genealogies: they could pride
themselves on knowing that Deianira, who later became Heracles’s wife,
was Meleager’s sister. But the final lines of the myth as Bacchylides
narrated it—in which Heracles asks Meleager whether he has an unmarried
sister at home and Meleager says that he does—must have sent a chill down
their spines.
Crossovers almost always serve another important purpose as well: by
evoking a story world that is already familiar and accepted, a crossover
character is a powerful way of giving verisimilitude to a new tale and its
characters. The most familiar application of this principle is probably the
television spin-off, in which a character from a well-established show is
sent out to start a new show, transporting parts of the old, familiar story
world into the new one so as to give it instant credibility. The sci-fi story

world of the British series Torchwood (2006–2011), for example, had
already been well limned by the older series Doctor Who before the
character of Jack Harkness crossed from the latter into the former. This
meant that some of the more fantastic aspects of Torchwood (the need to
defend the planet against aliens; the existence of a covert agency in charge
of doing so; rifts in the time-space continuum) were simply accepted by the
new show’s audience as givens. I’ve already mentioned the way in which
the bricoleurs behind the mysteries of Dionysus did something similar by
incorporating Persephone into the story that underpinned their cult; she
carried with her, from her existing role in the Eleusinian mysteries and their
myths, a stamp of eschatological authority. Sometimes, a crossover
validates or revivifies a story by bringing its central character into contact
with one or more characters who are already well established. For example,
a revision of the Athenian myth of Erigone wove characters from the House
of Atreus into her story, giving that Attic heroine panhellenic potential.48
There are subtler uses of the crossover as well. The opening lines of C. S.
Lewis’s The Magician’s Nephew (the sixth entry in The Chronicles of
Narnia) read,
This is a story about something that happened long ago when your grandfather was a child. It is a
very important story because it shows how all the comings and goings between our own world
and that of Narnia first began.
In those days Mr. Sherlock Holmes was still living in Baker Street and the Bastables were
looking for treasure in the Lewisham Road.
Lewis first anchors his story with an explicit reference to the preceding five
Chronicles (the word “Narnia”), but given that this new installment will be
set two generations before the others—meaning two generations before the
lives of the readers for whom he was writing in the 1950s—he additionally
anchors it with crossovers from Victorian fiction with which his readers
were sure to be familiar: Sherlock Holmes (of course) and the Bastable
children, whose adventures, narrated in three books by E. Nesbit at the turn
of the century, were still popular in the 1950s. These crossovers will not
appear again in Lewis’s story, but they create a climate in which it will
unfold. Notably, it is a climate that is decidedly rational (Holmes) and
quotidian: although Nesbit had also written many successful books about
children who have magical adventures,49 Lewis chose to refer to those about
the Bastables, which take place in a London that is firmly realistic. In other
words, Lewis used these opening crossovers to lay down a solidly rational

and believable outer story world in which he could then locate the magical
story world that begins to appear in chapter 2—by extension helping to
make that second story world believable as well. The trick is an old one.
Thucydides similarly predisposed his audience to accept his report on the
earliest days of the Greeks by backing it with what was, for them,
indisputably real and authoritative: the witness of Homer.50 Similarly, by
renaming local Athenian goddesses “Erinyes,” Aeschylus and Sophocles
gave epic authority to the local cults whose foundation stories they were
refashioning in the Eumenides and the Oedipus in Colonus.51
Sometimes a crossover character lingers longer in a story in order to set
the mood. Formally, Pindar brings Chiron into the tale of Apollo’s love
affair with Cyrene as an adviser and a prophet of what will result from the
tryst, should Apollo pursue it. Thematically, Chiron instantiates a balance
between the wild and the civilized that the rest of the poem explores as
well. Compositionally, his speech breaks the linearity of the poem.52 But as
a figure who had been associated with the nurture and training of young
heroes since the earliest stages of mythic narration, Chiron (like Holmes
and the Bastable children, but at greater length) also establishes an
atmosphere within the myth. This is no longer just one of many stories
about Apollo’s pursuit of nubile virgins or just one of Pindar’s many tales
about a colony’s foundation; Chiron’s presence subtly helps to make this a
narrative about maturation, which in turn helps to set the proper ambience
for an epinician ode in honor of a young victor who is poised on the verge
of adulthood himself.53
Crossovers, in sum, can do a number of things very efficiently: establish
the existential, ethical, and operational rules of a new story; lend it
credibility and authority by their mere presence; and establish a particular
climate or mood by gesturing towards other stories. Sometimes, they do one
more thing, as well. By inventing the conversation between Apollo and
Chiron about Cyrene, Pindar not only creates an opportunity to forecast the
entire span of Apollo and Cyrene’s courtship, the birth of their son and the
foundation of her eponymous Libyan colony but also tells us how the god
and the nymph crossed paths in the first place. By inventing the
conversation between Heracles and the ghost of Meleager, Bacchylides tells
us how the seeds for Heracles’s marriage and death were planted. Both
stories, in other words, claim to reveal “how it all came about”—a perennial

object of human curiosity. Filling such holes can do more than just satisfy
curiosity; by borrowing characters, information and episodes from other
established narrations in order to fill those holes, backstories become yet
another way of weaving a particular tale more tightly into the mythic
network. How did it happen that both Cadmus and Jason had to fight armed
men who sprang from dragon’s teeth sown in the soil? Because (Pherecydes
answers) after Cadmus slew the dragon, Athena and Ares collected its teeth,
giving half to Cadmus to sow immediately and half to Aeetes, who used
them to cause trouble for Jason many years later. And thus, events from the
Theban cycle were woven into the story of Jason. How did Apollo manage
to convince the Moirai to trade Admetus’s life for that of another? By
getting them drunk (answers Aeschylus’s chorus in the Eumenides). And
thus, events from a Thessalian saga were woven into a newly emerging
Athenian myth. How did Creusa end up possessing Gorgon’s blood with
which she could try to kill Ion? Because (Euripides tells us) Athena gave it
to Creusa’s ancestor Erichthonius, who passed it down through the royal
line.54 And thus, a parade of kingly Athenian ancestors glimmers behind
Creusa, further legitimating the son she is about to reclaim as an heir to the
throne. Everything can be made to fit together, everything can be
understood as part of a single, bigger picture and thus ratified, if only you
know where to look for the missing pieces—or how to fashion them
yourself. As H. P. Lovecraft put it when responding, once, to the question of
how he felt about fans using materials of their own construction to caulk the
gaps in his Cthulhu universe, even artificial mythology (his term) can be
given an air of verisimilitude if it is widely enough cited—or if it is
provided with enough citations to other works.55
The Story World of Greek Myth

Earlier, I concluded that the story world of Greek myth is relatively short on
the sorts of oddities that typically set a Secondary World apart from the
Primary World, but if I am right that its identity, coherence, and credibility
rests on the thickly crisscrossing network of gods, heroes, and monsters that
I have been describing in the preceding two sections, don’t I need to revise
that conclusion? Aren’t gods and monsters (leaving aside heroes for the
moment) by definition things that don’t belong in the Primary World?
This is where we most clearly begin to see the difference between a story
world created by myths and story worlds created by genres such as fantasy

and science fiction—or, to put it differently, between story worlds inhabited
by characters in which a society encourages its members to believe and
story worlds inhabited by characters who are not intentionally created to be
objects of belief (even if they later become such objects). I suggest it was
precisely myths’ representation of the gods and heroes as existing in a
world that looked much like the Primary World that helped to make it
possible for the Greeks to believe that those gods and heroes still existed at
the time the myths were being told (even if they no longer manifested
themselves so grandly and so frequently as they once had). For the Greeks,
the things, people, and events of what they called the old days and what we
would call the mythic period were understood to have melted into those of
the present age without an abrupt change. They characterized an earlier age
but not a time and place that was disconnected from the present age.56
Greek historians attest to this. Herodotus begins his history of the Persian
Wars by reporting a tradition about a tit-for-tat series of abductions of
Greek women by Eastern men and of Eastern women by Greek men—Io,
Europa, Medea, and Helen, all of whom we now are accustomed to think of
as characters in myths. Eventually, the enmity begun by the abductions
brought about invasions into Greece by very real Easterners, the dire effects
of which Herodotus’s audience knew quite well. Or so said many people—
Herodotus himself declines to weigh in on whether this tradition is accurate
or not.57 Thucydides opens his history of the Peloponnesian War by
suggesting that it was the necessity of attacking Troy that finally welded
small, independent kingdoms together into a unified Hellas and then goes
on to note that such a war would never have been possible without Minos’s
development of the first navy. He posits that Agamemnon was able to
muster the Greek troops more because he was the strongest of the Greek
chieftains than because of the romantic oath that they all had sworn to
defend Helen’s honor, but he does not call the existence of that oath into
question.58
Behind Herodotus’s and Thucydides’s remarks, we glimpse a widespread
assumption that the time in which Io, Europa, Minos, and Agamemnon had
lived was continuous with the time in which these historians were writing,
even if there were some qualitative differences between those two times and
even if one might niggle about the accuracy of certain details (as Herodotus
and Thucydides certainly did in some cases). The Greeks did not assume

that there had been an illud tempus (as Mircea Eliade would have it) that
was categorically different from hoc tempus.59
Later authorities were more explicit about this continuity. The Parian
Chronicle, a lengthy inscription recording Athenian history from 1582 BCE
to 299 BCE, starts with matters such as the rule of Cecrops (an early king
who was half snake and half human), the dispute between Ares and
Poseidon on the Areopagus, the kingship of Minos and the institution of
agriculture by Demeter. Moving forward, it slowly wends its way through
events such as the institution of the Eleusinian gymnastic games and the
tyranny of Pisistratus towards events in what was then quite recent history:
the ascension of Ptolemy I to the Egyptian throne and Philocles’s
archonship in Athens. The Chronicle, as Peter Green nicely formulates it,
“uses chronological specificity as a guarantee of truth.”60 The Lindian
Chronicle, inscribed in 99 BCE, recorded gifts made to Athena by Cadmus,
Minos, Heracles, and Helen right next to those made by Phalaris, Darius,
Alexander, and Ptolemy.61
This continuity of time is evoked by Demosthenes when he praises fallen
Athenians by placing them within a long line of glorious ancestors that goes
back to those who expelled the invading Amazons and Eleusinians and who
saved the sons of Heracles. It is also evoked when he praises Athenians of
his own day for standing up against northern forces as staunchly as their
ancestresses Procne and Philomela once had.62 And it is also evoked by off-
hand remarks such as those made by Socrates and Phaedrus as they strolled
along the Ilissus river: “Isn’t this the place where they say that Boreas
snatched away the princess Oreithyia?” asked Phaedrus. “No, I think it was
about a quarter-mile further along, where you cross to the sanctuary of
Agra,” replied Socrates. “There is, I believe, an altar dedicated to Boreas
close by.”63 Walking through an ancient city, especially if you were a native,
meant being constantly reminded by the monuments and landmarks that
surrounded you of what was said to have happened there before; they were
part of the surround of Noyes’s second and third categories, as discussed in
Chapter 1. If you weren’t a native, there were always natives from whom
you could learn the local lore as you strolled. Pausanias’s many descriptions
of places that evoke myths—here is the well where Demeter sat disguised
as an old woman; there is the chasm through which Heracles dragged

Cerberus up from Hades; this is the city founded by Lycaon, who later
turned into a wolf—are particularly meticulous examples.64
According to such a view, although it might not be reasonable to expect
to encounter more nine-headed snakes in Lerna or three-bodied giants in
Spain (the passage of time was understood to have changed some things,
after all; while Heracles and the other heroes still dwelled on earth, they had
cleared the world of such creatures) or to expect gods to transmogrify
people into new animals and plants (the natural world had been holding
steady for some centuries), it would be reasonable to expect the heroes and
the gods to remain an active part of the contemporary world. And so it was:
Heracles warded off evil from houses that displayed his image and watched
over young men at the gymnasium; Asclepius and Amphiaraus worked
cures in their sanctuaries while their clients slept; Pan spoke to Phidippides
in the mountains; and Apollo fought alongside his worshippers against the
barbarian Gauls in 279 BCE, helped by local heroes and perhaps his sisters
Artemis and Athena.65 Athena manifested herself to her worshippers several
times in Lindus, as her temple’s chronicle recorded; in 490 BCE, she
brought rain to the thirsty people when they were besieged by the Persians.
Other gods and heroes, on many other occasions, were seen, heard or felt by
their worshippers.66
Greek myths provided templates against which such manifestations of
gods and heroes could be shaped, measured, and confirmed. The myths
also, as I have suggested in this chapter, provided a story world that bound
their protagonists—divine, heroic, and human—into a larger network. Each
bond of this network was tight enough to secure the others; each story could
thereby accredit, and be accredited by, the others. Yet the bonds were
supple, as well, allowing the sort of revision that kept Greek myths in step
with changing beliefs about the gods and changing practices in their honor.
They were supple enough, too, that interesting new details and episodes
could emerge, keeping the stories and their characters vigorously alive. The
story world of Greek myth was the ideal companion for a religious system
whose conceptions of divinity were anchored not by sacred texts or canons
of doctrine, but rather by shared beliefs.

5
Characters
CHAPTER 3 was about how narratives, including myths, could be constructed

and delivered in ways that helped to create and sustain belief in Invisible
Others, including the Greek gods and heroes. Chapter 4 was about what
made the world that Greek myths portrayed seem like a single, cohesive
whole and how that, in turn, also helped to sustain belief. This chapter will
look more closely at the characters who populated that world.
I’ll start with a basic question: what is the ontological status of the gods
and heroes whom we hear about in myths? Believers (or at least the
intellectuals amongst them) were interested in one angle of this issue
insofar as they sometimes attempted to categorize gods and heroes, to
establish the sorts of creatures they were. What qualities set a god apart
from a human? Immortality? Omniscience? Omnipotence? Something else?
And what about the heroes? Did they constitute a separate group that lay
between gods and humans, sharing qualities of each? Later on in this
chapter, I’ll return to a variation of these questions, when I look at how the
special qualities of the gods were presented in narratives.
But first I want to pursue some other ontological questions, which are
better asked from our own vantage point, standing outside of ancient Greek
culture. Can we say that the gods and heroes existed independently from the
stories in which they appeared as characters? And if so, what was the
relationship between those characters and the “real” things? Or to put it
otherwise: to what extent, if at all, did the Greeks distinguish between the
gods and heroes as they appeared in narratives and the gods and heroes
whom they worshipped? In recent years, analytical philosophers and
narratologists have argued that some fictional characters (such as Juliet
Capulet, Sherlock Holmes, and James Bond) have achieved existences that
transcend the individual narratives in which they appear, and thus can be
said to exist independently of those narratives.1 Did the Greek gods and

heroes similarly achieve transcendence beyond the words and images that
portrayed them, and if so, how might that have enhanced belief in those
characters?
The Ontology of Fictional Characters

In one way, these issues may not matter much. As I noted in Chapter 3, the
distinction that an individual perceives between reality and fiction and
therefore between real and fictional characters, is conditioned to a large
extent by what sources of authority tell him or her is real and fictional
(religious authorities, civic authorities, scientific authorities, parental
authorities, educational authorities, and so on). As long as the particular
reality offered by one of these authorities does not significantly contradict
an individual’s own experiences, it constitutes a viable alternative, and so
the individual might adopt it as his or her reality. When more than one
viable reality is on offer, the relative importance of the authorities who
promote each one will influence the choice. Thus, to take a contemporary
example: Darwin’s theory of evolution tends to be endorsed by those who
have been raised in a home and an educational system that prioritize
scientific authority, but it tends to be rejected by those raised in a home and
an educational system that prioritize what the Bible is understood to say in a
literal sense. Those who accept either of these endorsements live in a
different reality from those who accept the other, at least as far as their view
of how the physical world and its creatures were generated, which in turn
has ramifications for their assumptions about humanity’s place in the larger
scheme of things, what humanity’s responsibilities are of the natural
environment and what role, if any, god(s) play in such matters.
Not all societies generate and promulgate a range of views, however; in
ancient Greece, what we might call “scientific” models of how the cosmos
came to exist and what made it work were available almost exclusively to
intellectual elites; by default, most non-elites accepted the reality that was
articulated by poets and endorsed by religious and civic officials, which
assumed that the gods existed and wielded a lot of power. In other words,
Zeus, Athena, Heracles, and all the other gods and heroes whom most of us
living in the twenty-first century have been conditioned to call fictional
characters were treated as real by the authorities whose opinions mattered
most to ordinary Greeks, and because they did not encounter significant
evidence to the contrary, most of them probably also took the gods to be

real. This does not mean that every single person accepted every single
thing that every poet told them about the gods, but the existence of the gods
as the poets had cumulatively portrayed them constituted a framework of
belief.
But to return now to philosophers’ and narratologists’ discussions about
the ontology of fictional characters. In spite of what I’ve just said, it’s worth
testing some of the approaches that these scholars have developed for
understanding fictional characters against the Greek gods and heroes, in
order better to understand how the Greeks’ perception of the gods and
heroes differed from our own perception of fictional characters.
Central to all such discussions is recognition of the fact that there are
different types of fictional characters; I am going to propose that there are
five. Characters of the first type exist only in a single instantiation, and
therefore always appear to us exactly as their creator fashioned them. At the
moment that I write this, for example, the characters in Alexander McCall
Smith’s popular 44 Scotland St. series have never appeared outside of
Smith’s narratives, and thus Bertie, Angus, Dominica, and the rest have
been presented to us only as Smith (and his illustrator, Iain Mcintosh)
conceives of them. Of course, we each add our own elements to these
characters while reading the books—we invent details of their physiques
and clothing, the timbre of the voices, and so on, without even realizing that
we are doing so—but we unconsciously rein in our imaginations as we do
so, keeping the characters within the parameters that Smith established.
Bertie and the others therefore retain stable identities and we will instantly
recognize them in the next book of the series that Smith writes—unless he
contrarily decides to change them in significant ways. By the same token,
however, it is difficult for these characters to exist independently of Smith’s
narratives; they live only within the specifically envisioned, semi-fictional
version of Edinburgh that he has created.
Characters of my second type are portrayed by more than one author or
artist and yet can still be recognized as that character in each portrayal, even
if details differ. The philosopher Maria Reicher suggests that we approach
these situations by thinking in terms of “maximal characters” and
“submaximal characters.” The maximal character (who is only a heuristic
convenience and is not actually manifested in any narrative) possesses all
the properties that any instantiation of the character ever exhibits.2 Some of
those properties may be mutually exclusive. The maximal James Bond has

both dark hair (as in Sean Connery’s portrayal of him) and blond hair (as in
Daniel Craig’s portrayal), for instance, even though no single instantiation
of Bond is both naturally dark-haired and naturally blond. Submaximal
characters, which are those we encounter in actual narratives, manifest only
some of the properties possessed by the maximal character, but enough to
be recognized: whatever James Bond’s hair color may be in one of these
submaximal instantiations, he is always a secret agent working for Britain,
always dresses well, and always prefers his martinis shaken, not stirred.3
Characters like these have more fluid identities than the first type I looked
at, but they nonetheless retain a core stability.
The most popular amongst these fluid characters are likely, eventually, to
make a leap into the collective imagination that enables them to stand
completely outside of all the various narratives in which they have
appeared, recognizable as a coherent individual even by those who have
never read or viewed their stories. Thus, for instance, a person who has
never consumed a single narrative in which Sherlock Holmes appears might
see a drawing of a man wearing a deerstalker hat and smoking a
meerschaum pipe in an advertisement, infer that the man is Holmes because
this type of hat and pipe are closely associated with him, and know enough
about Holmes to be able to guess that the business being advertised has
something to do with tracking down information. It is this type of character
who most interests the narratologists and philosophers who have worked on
the ontology of fictional characters, because it is this type that comes
closest to breaking free not only of the original narrative in which he or she
appeared, but also of all narratives altogether, becoming an independently
existing entity.
Characters of my first type, for which I used Smith’s Bertie, Angus and
Dominica as examples, are fully fictional, so closely circumscribed by the
narratives in which they appear as to be unlikely to dissociate themselves
from those narratives. Characters of the second type usually remain
fictional as well: even if many audience members cannot state who it was
that first created Bond and Juliet, most of them realize that Bond and Juliet
were created by someone, at some point, ex nihilo. A few characters of the
second type, however, begin to cross the line into nonfictional status: a
business that installed its headquarters at 221 Baker Street in 1932 soon had
to hire a full-time secretary to handle all the mail addressed to Holmes and
continued to keep a secretary on staff for that purpose until the company

closed in 2005 (all the mail addressed to Holmes is now redirected to the
Sherlock Holmes Museum, a few steps up the street).4 In other words, some
people believed, and still believe, that Holmes is real; he has transcended
the narratives that present him.
A third type of character appears in fictionalizing narratives but is based
on a real person (living or dead). This type develops in ways that are similar
to the second type I discussed, with one important difference. I will take
Queen Elizabeth II as my example. The fictionalized character of Elizabeth
II as portrayed by Helen Mirren in Stephen Frear’s The Queen (2006) and
the fictionalized character of Elizabeth II as portrayed by Claire Foy in
Netflix’s series The Crown (2016–2017), for instance, participate in the
larger character concept “Elizabeth II” in the same way that Ian Fleming’s
novels and Daniel Craig’s portrayal of Bond participate in the larger
character concept “James Bond.” But in contrast to Bond, for Elizabeth II,
we see and hear the real Elizabeth herself, in addition to seeing and hearing
Mirren’s and Foy’s fictionalizing portrayals.
This “Real Elizabeth” provides fodder for the construction of a single,
maximal “Elizabeth” character in more or less the same way as Mirren’s
and Foy’s Elizabeths do. These three Elizabeths (and others) converge in
the minds of those who have experienced them towards a single Elizabeth
concept. The fodder provided by Real Elizabeth, however, tends to carry
extra weight because we have been repeatedly told by numerous
authoritative sources that Real Elizabeth is authentic in a way that Mirren’s
and Foy’s Elizabeths are not. This doesn’t drown out the contributions made
by Mirren’s and Foy’s Elizabeths—after all, they come to us in narratives
that have been carefully orchestrated to hold audience attention and to
imbue their Elizabeths with charisma. Indeed, particularly if we watch
Mirren’s and Foy’s performances before we have seen or heard Real
Elizabeth, our cumulative concept of Elizabeth II might be so significantly
shifted in the direction of those performances as to forever inflect what we
subsequently learn about Real Elizabeth.5 And yet, I suggest, the
overwhelming agreement amongst authorities that Real Elizabeth genuinely
exists means that at least some aspects of Elizabeth’s cumulative character
—her hair color, date of birth, marital history, number of children, favorite
breed of dog, typical mode of dress and moral outlook, for instance—are
inevitably governed by the ways they are expressed by Real Elizabeth.

Notably, it seems that few people are uncomfortable with the coexistence
of Mirren’s and Foy’s Elizabeths alongside Real Elizabeth, even if they
notice that some things that Mirren’s or Foy’s Elizabeths say and do are
inaccurate or unsubstantiated. We are used to the fact that real people and
real events must be altered in the course of being dramatized, either because
the narrator has little idea of what the real words and details were or
because the real words and details are inadequate to express what the
narrator aims to convey. The ancients were also familiar with, and
comfortable with, this situation, as Thucydides makes clear: “my method
has been, while keeping as closely as possible to the general sense of the
words that were actually used, to make the speakers say what, in my
opinion, was called for by each situation.”6 In other words, people have
long understood, and accepted, the idea that their impression of a real
character is gleaned, in part, from sources that are written with the aim of
providing entertainment or instruction as much as or more than providing a
factual portrait. The fact that a representation has been scripted doesn’t
mean that the subject it represents must be understood as an invention.
These scripted versions of real people, moreover, have an advantage over
what we know of the real versions: those who create them can give us as
much access to their characters’ thoughts, desires, fears and motivations as
they wish. Even the most intimate sibling or friend of a real person is
unlikely to possess that. In other words, these scripted versions have a
narrative advantage; they make for characters with whom we can interact
more fully.7
A fourth type of character—Invisible Others who are anchored by
canonical narratives—offers a twist on the issue of combining “real”
characters with fictionalized versions. Just as there is a Real Elizabeth, for
believers there is a “Real God” (or “gods”), for example, and what those
believers think they know about Real God is combined with portrayals of
God in fictionalizing narratives to make up a cumulative God concept. For
most contemporary Western Christians, Real God is the one described in the
Old and New Testaments. Fictionalizing portrayals include the God of
Milton’s Paradise Lost (1667), the God whose voice (provided by Charlton
Heston) is heard talking to Moses in Cecil B. DeMille’s The Ten
Commandments (1956), God as portrayed in a stained-glass window that
one sees every Sunday while at church, God as portrayed by the singer
Alanis Morissette in Kevin Smith’s film Dogma (1999), and so on. The

main difference between Real Elizabeth and Real God (as is typically the
case for Invisible Others) is that information about Real God is less readily
accessible than information about Real Elizabeth. Such things as Real
God’s physical appearance and his favorite breed of dogs remain
mysterious, and God therefore remains what narratologists call a less
“complete” character than Elizabeth—and less complete even than Holmes,
about whose personal history and tastes we learn a great deal by reading
Doyle’s stories—to say nothing of the hundreds of Holmes stories that have
been generated since Doyle’s death.
We can make two observations about the four types of characters I have
sketched. First, that characters of my second type, who are portrayed in
more than one narrative, are likelier than characters of my first type to
eventually develop into characters of the collective imagination and may
even break free of narratives altogether to win an independent existence that
eventually dwarfs those narratives; second, that characters of my third and
fourth types, who are based on real individuals—not only real humans but
also real Invisible Others who are anchored by canonical narratives—
develop in much the same way as characters of the second type, although
the real individual usually exerts a more powerful force over the cumulative
character concept than do other (fictionalized) instantiations of the
character. Of course, the other difference between my second type and my
third and fourth types is that some or all of a society’s authorities agree that
types 3 and 4 are real but not that type 2 is.
Things are different when it comes to Invisible Others in cultures such as
ancient Greece, where there were no dogmas concerning the nature of the
gods and heroes and no canonical texts that wielded authority like that of
the Bible. Because of this, there were no widely agreed-upon sources of
knowledge for “Real Hermes,” for instance, that could be clearly
distinguished from “fictionalizing” portrayals of Hermes; different
portrayals of Hermes simply had to sit alongside each other, none of them
irrefutably able to claim to be true. In epic, Hermes is a helpful, lighthearted
god, who, as Zeus puts it in the Iliad, “greatly enjoys having humans as
companions.”8 And yet the Prometheus Bound presents Hermes as a
ruthless tool of Zeus, willing to inflict cruel punishments on Prometheus,
humanity’s greatest advocate. Each of these portraits of Hermes, and others,
were understood by their recipients to be “true” in the sense of expressing
something about the nature of Hermes, even if they jarred against one

another. If anything in ancient Greece approached canonicity, it was the
narratives attributed to Homer and Hesiod and yet even their information
could be challenged—in fact, the two poets famously contradicted each
other on some points. Homer says that Aphrodite’s parents are Zeus and
Dione, but Hesiod says that she was born from the foam that arose when
Cronus threw Ouranos’s castrated genitals into the sea.9
For centuries, then, Greek authors, artists, and their audiences implicitly
participated in an ongoing conversation about exactly who the gods were
and what they had done during the early epochs of the cosmos’s existence.
The material that fed this conversation came almost completely from what
we in the twenty-first century understand to be fictional sources that the
Greeks themselves understood to be, if not fully fictional in the sense of
having no basis in fact, then at least fictional in the sense of being ficta—in
the sense of being articulated in forms that had been generated by human
individuals.
There are important differences between this type of character and the
other types that I discussed earlier. As long as the work that first created a
fully fictional character continues to be read, viewed or in other ways
present in the audience’s consciousness, it exerts an influence over
subsequent portrayals. James Bond has not yet escaped from Fleming, and
Sherlock Holmes has not yet escaped from Doyle. Similarly, in the case of
fictionalized characters based on real people, such as Queen Elizabeth, or
Invisible Others for whom there are canonical portrayals, such as the
Christian God, our knowledge of the real characters prevents fictional
portrayals from going too far astray. Audiences are unlikely to accept a
portrayal of Elizabeth in which she has an extramarital affair as anything
but a parodic inversion of Real Elizabeth’s well-known adherence to
traditional moral codes,10 and it is unlikely that many people’s concepts of
God were significantly affected by Dogma’s casting of Alanis Morissette in
the role.
In contrast, characters for whom there is no clear original—whose
existences are anchored instead within the drifting overlaps of traits shared
by different portrayals—can be adapted in bolder ways. To be sure, as with
the other types of characters, some elements may never be changed: an
impotent Zeus or an ugly Helen could be understood only as figures of
satire, Heracles must always travel to distant lands and Theseus must
always be an Athenian. But Heracles can be portrayed either as a symbol of

virtuous self-control and abstention or as a gluttonous, drunken lecher,
without either portrayal being understood as satire. Theseus’s faces are
manifold (see later in this chapter). Helen can be either a slut who ran off
with a handsome young visitor, a virtuous wife who was kidnapped by that
visitor, or an unsullied woman who never slept with the visitor at all.
One reason that the existence of such varying portrayals is tolerated is the
very fact that their subjects are assumed to have really existed—and still to
exist, even if on a higher existential level than before in the case of mortals
who have become heroes. Because of this, it is thought possible to discover
new information about them or a new slant on old information. Most
fictional characters—even those who have been instantiated in different
ways a number of times—are not like this. We do not expect that new
information might become available about Madame Bovary that would
change our evaluation of her choices and actions. We do not expect that
new information about Don Quixote’s childhood might someday help us
understand why he undertook his misguided chivalric quests. Indeed, if
anything argues for the mythic status into which Sherlock Holmes is
sometimes said to have moved, it is the constant generation of new
representations of him and his associates by authors such as Michael
Chabon, Caleb Carr, and Karem Abdul-Jabar, which delve into Holmes’s
childhood, his old age, his relationships with women (or with Dr. Watson),
his brother Mycroft’s personality, and so on. In doing so, these narratives
sometimes offer us a Holmes who departs significantly from the character
whom Doyle created without becoming parody.11
Plurimediality and Accretive Characters

I suggest, then, that rather than bemoaning the fact that the Greeks’ ideas
about the gods and heroes come to us mostly through fictionalizing art and
literature, we should acknowledge that those portraits, cumulatively, were
the gods and heroes—that is, that the nature of our own experience of these
gods and heroes at least approximates that of the Greeks themselves. From
there, we can begin better to understand the contributions that the characters
in narratives made to the creation and sustenance of belief.
This brings me to two interrelated concepts, plurimediality and accretive
characters.12 When characters are presented through more than one
narrative (as are the characters of types 2, 3, 4, and 5, described earlier),
each narrative offers a different instantiation of that character from the

others—sometimes a strikingly different instantiation. Such characters are
plurimedial insofar as many different media, and sometimes many different
manifestations within a single medium, represent that character. Sherlock
Holmes, for example, has been represented through printed stories written
by several different authors (accompanied by drawings produced by many
different illustrators), through stage plays, movies, and television shows
starring different actors, through comic books, on advertising logos, and so
on.
But if “plurimedial” describes the nature of the sources from which a
single person’s concept of a particular character may be drawn, “accretive”
better describes the state of the character, as it is taking form in the mind of
a particular individual. As the individual encounters the character in
different instantiations, the individual’s concept of who the character is
slowly accrues traits from some or all of those instantiations. In what
follows, I will use “accretive” and “plurimedial” as appropriate.
When experiencing plurimedial characters, each of us must repeatedly
choose, even if unconsciously, to engage more deeply with some
instantiations than with others, and with some aspects of our preferred
instantiations more deeply than with others. Each of us therefore ends up
creating our own accretive character, no two of which are likely to be
exactly the same. It is a process that may continue over the course of many
years. Because, as a child, I always engaged with instantiations of Sherlock
Holmes that wore the distinctive deerstalker hat, I found it difficult, later
on, to accept one who wore a fedora, as Jeremy Brett sometimes did when
portraying Holmes for the BBC. And yet Brett’s portrayal was so appealing
to me in other ways that my image of Holmes began to incorporate the
fedora. Similarly, once I began, as a teenager, to read Doyle’s own work
rather than the expurgated versions of it that are offered to children, the
element of cocaine addiction became important to my concept of Holmes,
which predisposed me to accept portrayals of Holmes that emphasize his
drug abuse, such as those provided by Jonny Lee Harris in the CBS series
Elementary and by Benedict Cumberbatch in the BBC series Sherlock.
Harris’s and Cumberbatch’s portrayals of Holmes brought his intellectual
arrogance to the fore, and therefore that arrogance also became central for
my concept of Holmes, which made it harder (although not impossible) for
me to accept the older, kinder Holmes that Ian McKellen portrayed in the
film Mr. Holmes (2015).

Plurimedial characters continually compel audience members to make
decisions such as the ones I made for Holmes, even if only unconsciously.
The cognitive and emotional energy that each person invests in doing this
forges an especially close bond between that person and the character. Or in
other words, because my vision of Holmes really is my vision, it opens me
up to being affected by Holmes more strongly than I would be if I had
adopted a single instantiation directly from a single source. And this
situation, in turn, is conducive to my believing that Holmes really exists—if
the authorities in my society endorse that belief.
I want to stress a few more things about plurimediality before I continue.
First, although I have associated plurimediality with the existence of
multiple portrayals of a single character by different authors or artists,
plurimedial characters are not themselves the creation of authors or artists.
Rather, plurimediality is something that accumulates around a character as a
by-product of the work that different authors and artists collectively do.13 A
plurimedial character, then, is a product of its own popularity—popularity
that its plurimedial status in turn continues to promote.
Second, an individual’s experience of a plurimedial character is normally
confined to the imagination of that individual alone. If it moves out onto the
screen, the stage or the page, this is a secondary development, at the hands
of an author or artist who uses his or her perception of the character to
create something new or to make a point. For example, Shane Denson has
demonstrated how a representation of Frankenstein’s monster in a graphic
novel of the 1960s skillfully evokes elements of both Mary Shelley’s
literary portrayal and Boris Karloff’s cinematic portrayal in order to ponder
the nature of the monstrous.14
Third, understanding a character as plurimedial frees us from the
constraints of another approach to which it may seem to bear a
resemblance, namely, intertextuality. By definition, a plurimedial character
shimmers at the nexus of several, perhaps even many, different
instantiations of that character, taken from different texts, none of which has
absolute authority over that character and any one of which may therefore
dominate the others in the mind of the individual who creates it.
Plurimediality, in other words, foregrounds, and indeed thrives upon, the
somewhat messy and constantly metamorphic state in which popular
characters exist. It encourages us to embrace a dynamic view of characters’
development and of personal engagement with characters, according to

which frequent transformation is not only tolerable, but even fruitful. It
releases us from the deadening pursuit of Quellenforschung, given that the
point is not to trace particular traits of a character back to an earlier source
or sources (with which a given audience member may or may not be
familiar), but rather to understand how new figurations emerge and
reemerge against the background of a range of earlier treatments, never
fully obligated to any of them.15
Plurimedial Characters in Antiquity

Although the examples I have given so far focus on the nineteenth and
twentieth centuries, plurimediality and its ramifications have always been
around, as earlier parts of this chapter have already begun to suggest.
Ancient literary and artistic sources provided many portrayals of the gods
and heroes, from which each individual would have had to form his or her
own personal versions. We meet quite a different Odysseus in Homer from
the one we meet in Sophocles’s Philoctetes and yet again a different one in
the final book of Plato’s Republic—and then there is the burlesque
Odysseus whom we see on a black-figure vase from the Theban Kabeirion:
squat and grotesque, his eyes and mouth wide with terror at the approach of
Circe. Or take Theseus as an example. An Athenian man of the late fifth
century might have encountered Theseus as a somewhat foolhardy young
leader-in-the-making, leaping into the sea to settle a bet about his paternity
(in symposiastic re-performances of Bacchylides’s seventeenth ode) or as a
paradigmatic hero conquering paradigmatic monsters (in the same poet’s
eighteenth ode); as a noble, magnanimous defender of the downtrodden (in
Sophocles’s Oedipus in Colonus) or as the cruel betrayer of a woman who
saved him (in the Hesiodic Catalogue of Women); as a democratizing king
embracing civic virtues (in Euripides’s Suppliant Women and Thucydides’s
“archaeology”), or as the careless cause of his own father’s death (in
Simonides fr. 550); as the founder of rituals for young men in Athens and
on Delos (popular traditions that made their way into the Atthidographers
and thence into Plutarch), or as a short-tempered and suspicious father (in
Euripides’s Hippolytus); as a middle-aged man traveling to Hades to back
up his best friend (in recitations of the epic Minyas), or as a dead king
whose bones Cimon had retrieved from the island on which Theseus spent
his last days in exile. And of course, that Athenian man also would have
met Theseus in vase paintings, on coins, on the metopes of the Hephaestion

and in many other works of art, which sometimes showed the hero with his
signature weapon, the primitive club, and at other times with the hoplite’s
sword or spear, sometimes as a youthful slayer of the Minotaur and at other
times at an older age, conquering the centaurs. Like Frankenstein’s monster
or Sherlock Holmes, in other words, Theseus was a dynamic character, of
whom each Athenian would have had a somewhat different conception but
to whom, by the same token, each would have felt an intimate and lively
connection as a result.
But in contrast to Holmes and the monster, Theseus was worshipped in
cult; that is, not only did the Athenians think about him as if he really
existed, but the great majority of Athenians also treated him as if he could
have a significant effect on the contemporary mortal world—indeed,
inscriptions tells us that the Athenian Council decreed that he should so be
treated. At the Theseion near the agora, for example, the Athenians stored
the bones of Theseus that Cimon had retrieved, on the understanding that
those bones would protect Athens against its enemies. The annual Theseia
was celebrated at the Theseion, and all year round the site provided asylum
for runaway slaves and anyone else “who is poor and downtrodden and
fears the strong.”16 In Phaleron, near the western harbor of Athens, the sons
of Theseus (and probably also Theseus himself) were paid cult. The
Oschophoria that was celebrated there, in which two young men of noble
birth dressed up as girls, was said to have been founded by Theseus at the
time of his youthful journey to Crete—an occasion on which, according to
myths, he had disguised Athenian youths as maidens so as to smuggle more
manpower aboard Minos’s ship. At the Pyanepsia, which was celebrated on
the same day as the Oschophoria and which was also said to have been
founded by Theseus, younger boys took center stage (although we don’t
know exactly what they did). The annual Synoikia celebrated Theseus’s
unification of Attica.17
The fact that Theseus was paid cult at these and other festivals leads to
two complementary observations. On the one hand, the cults’ activities—
sacrifices to Theseus, prayers to Theseus, and so on—can be understood as
yet further media through which the Greeks encountered Theseus, and
therefore as yet further factors contributing to his plurimediality and to the
dynamic nature of his reception amongst Athenians. Activities at the
Theseion and the celebration of the Synoikia, for example, seem to evoke
his reputation as good and kindly king, whereas the atmosphere in Phaleron

has a transgressive accent. But on the other hand, and for my immediate
purposes more importantly, the accretive nature of any single person’s
concept of Theseus and the intimate connection to Theseus that this fostered
helped to create and sustain the very assumption that Theseus existed,
which in turn sustained the practice of his cults. The same can be said for
other heroes and gods.
Life in the Margins

The absence of any strict canon governing how the gods and heroes were
supposed to be imagined left room not only for the rise of plurimedial
characters but also for the continual revision of old stories. Frequent
revision was virtually a necessity, in fact, given that poetic narration of
myths served as one of the main forms of public entertainment. Because
there were intrinsic limitations as to whom narrators could use as the
protagonists of their stories (the family of gods didn’t grow much larger
during the historical period, and few new additions to the roster of heroes
were widely enough known to merit big-time poetic attention), narrators
who wished to please their audiences with something new had to find ways
to inject novelty into material that was already familiar. In addition to
offering new slants on established characters, authors could innovate by
tucking new myths inside of old ones (thus, for instance, fifth-century
audiences heard, for the first time, about Heracles acquiring olive trees for
the new Olympic precinct from the Hyperboreans—trees that he had first
seen while completing his well-known third labor—and also heard about
Theseus plunging into the sea to retrieve Minos’s ring).18 Yet another way
was to invite audience members to consider new readings of the truth
behind an old tale. Pindar asserts that Tantalus was punished not for killing
his son Pelops and trying to serve the child’s flesh to the gods, as the
familiar story said, but rather because he stole nectar and ambrosia from the
gods and distributed them amongst his friends. Pindar also offered an
explanation for why the wrong story became popular: he said that Tantalus
was accused of killing and cooking Pelops by envious neighbors after
Pelops had been stolen away by a lovesick Poseidon.19 The poets Acusilaus,
Pherecydes, and Bacchylides challenged Hesiod’s story about the daughters
of Proetus by claiming that it had been Hera, not Dionysus, who drove the
girls mad. Bacchylides challenged all of the other sources by saying that the
girls were cured not by the famous seer Melampus but by Hera herself, after

Artemis interceded with her on their behalf, which Artemis was motivated
to do by the twenty red cows that Proetus sacrificed to her.20 In the religious
system of ancient Greece, famous figures and their myths were malleable
materials, through which new speculations about the nature of the gods and
heroes could be explored.
In a narrative environment filled with such protean characters and
shifting stories, audience members were bound to notice that there were
varying—sometimes strikingly varying—reports about the gods and heroes.
Under these circumstances, even the oldest and most appealing stories had
to be accepted as just single reports about someone whose real nature might
never be fully captured by human words. And yet the very fact that so many
voices had independently arisen to tell tales about a particular character
helped to confirm that someone real stood behind them all, somewhere.
Pindar’s interjection, “surely you all know about Ajax’s blood-stained
valor? … Homer has made him honored amongst humans by setting straight
his entire achievement,”21 implies that Ajax’s story has been transmitted by
particular humans who had particular opinions about him, some of which
were closer to the truth than others, but it also confirms that there is a real
Ajax who can be recovered. Herodotus’s efforts to discover who the real
Heracles was, by visiting Egypt, Tyre, and Thasos and interviewing local
people in each of those places, was, perhaps, an extreme reaction to the
diversity of stories that circulated about Heracles, but it reflects Herodotus’s
confidence that a real man stood behind the various legends.22 Plutarch
articulates exactly the same problem at the start of his biography of
Theseus: “Let us hope, then, that I shall succeed in purifying what is mythic
[mythodes], and make it submit to reason and take on the appearance of
history.”23 Of course, Stesichorus’s palinode to Helen demonstrates that if a
narrative fell egregiously short of truth, the divine figure behind the story
might be expected to intervene and compel a poet to set the record
straight.24
This state of affairs—this general tolerance for a swirling cloud of
opinions about the strengths and peccadillos of the gods and heroes—
resembles that around figures who are undeniably real: was there just one
opinion about the accomplishments of Pisistratus or the morals of
Alcibiades? Was there agreement about whether Pythagoras had really
experienced metempsychosis? It is a character about whose traits and

actions there is unanimity, in fact, who risks looking fictional. Another
reason that the gods and heroes whom the narrators described seemed real,
then, was precisely the fact that they could not be contained within any
single diegetic25 world; they seeped beyond the boundaries of the individual
myths that the poets told about them and flourished in the margins,
nourished by particular narrators but never becoming the property of
particular narrators.
Divine Ontology
I want to switch to a different ontological question from those that I started
with. I want to think about the ways in which—and the degree to which—
Greek gods were different from humans, and why (I’ll discuss the heroes in
more depth in Chapter 7).
In one sense, Greek gods were human: they emerged from a collective
consciousness that couldn’t imagine gods in any other manner—a tendency
shared by many cultures. Most of the Greek gods had arms, legs, eyes, a
nose, a mouth, genitals, and so on that looked exactly like ours—except that
they were more beautiful, more impressive. These gods slept, ate, made
love, became pregnant and gave birth, participated in athletic contests,
fought wars, were wounded, sweated, urinated, farted, shat (at least in
comedies), took baths, and ornamented themselves with clothing, jewelry,
and perfumes.
But if this had been all that there was to them, they wouldn’t have been
gods. What set them apart from humans, first and foremost, was their
immortality, as a common term for them, “those without death” (hoi a-
thanatoi), makes very clear. In this respect, they were as different from
humans, “those who die” (hoi thnētoi), as they could possibly be. They also
had powers far beyond those possessed by humans: they could travel across
great distances rapidly, fly through the air, move heavy objects with ease
and shift their shapes to mimic anything or anyone they chose. They knew
what people were doing even when people thought that they were alone, as
Lycaon and Croesus each discovered.26
This is an impressive list that makes it tempting to add omnipotence and
omniscience to the traits that set the gods apart from humans. And yet,
myths narrate plenty of incidents that suggest that the gods were neither
omnipotent nor omniscient. Gods could be deceived by other gods or
remain ignorant of what other gods (and even humans) had done until those

deeds were called to their attention, for instance. Zeus didn’t realize until
too late that Hera had an ulterior motive for seducing him in Iliad 14.
Poseidon was ignorant of the other gods’ decision to bring Odysseus home
because Poseidon was far away in Ethiopia when they made it. Poseidon
was also ignorant of how Odysseus had injured his son Polyphemus until
Polyphemus told him.27 Zeus knew that he could sleep with a certain
goddess only at his own peril, but Prometheus alone could tell him which
goddess that was; in spite of all Zeus’s power, he could not learn the answer
on his own.28 Hades was able to lure Persephone away to her doom by an
attractive flower. Although Hecate heard Persephone cry out when Hades
snatched her, she did not know who had done the snatching until Helios told
her, and Helios knew this only because he saw the whole thing happen as he
was driving his chariot through the skies, just as we might see something in
our immediate environment. None of the other gods even heard Persephone
call out, and Zeus, who had masterminded the kidnapping, was able to spare
himself the painful sound of his daughter’s cries by removing himself to
one of his temples and listening to people’s prayers. (Apparently, unlike
Superman, the gods couldn’t hear what was happening far away any better
than we can.) Persephone was tricked yet again by Hades when he
convinced her to eat something in the Underworld.29 Eileithyia didn’t know
that Leto was in labor, in spite of Leto’s agonized prayers to Eileithyia,
because Hera had isolated Eileithyia within a cloud. Apollo didn’t know
that Telphousa was lying to him about the best place to build a temple;30 nor
did he know, on another occasion, why his cows had disappeared until he
had put in the same sort of investigative legwork as any victim of a cattle
theft would.31 Demeter, distracted by grief, ate a bite of Pelops’s shoulder
without realizing what it was. The children of Helios forgot to carry fire
with them when they set out to honor the newly born Athena. Eos forgot to
ask that Tithonus be made eternally young, instead of just eternal.32 And so
on, and so on. I could extend this list of the gods’ instances of ignorance
and error to a considerable length.
Gods were not able to prevent their favorite mortals from dying at the
time that the Fates had decreed, as several instances in the Iliad demonstrate
—most spectacularly, the case of Sarpedon, whose father, Zeus, wanted to
snatch Sarpedon off of the battlefield and out of harm’s way but had to
settle, instead, for ensuring his son’s honorable burial and shedding tears of

blood upon the earth.33 Apollo did a little better for his friend Admetus,
convincing the Fates to accept another human life in place of Admetus’s
own when his death came due (although time showed that this was less of a
bargain than Apollo had hoped). Zeus similarly tried to bargain for Castor’s
life, but the Fates relented only on the condition that Castor’s immortal
twin, Polydeuces, agree to take on half of the debt and thereby spend half of
the time in Hades, himself. Although gods were immortal themselves, they
could be physically wounded by mortal warriors: Heracles wounded Hera
and Hades; Diomedes wounded Aphrodite and Ares and made Apollo cry
out in fear.34
And in spite of all the other remarkable abilities of the gods, they
apparently need tools to accomplish at least some of their tasks. Hermes
uses special sandals to travel from place to place.35 Hades owns a cap of
invisibility—which Athena borrows in order to help Diomedes wound Ares,
so that Ares won’t see her.36 Why does a god need such a thing in order to
become invisible? Why does Athena need an aegis to shake fear into mortal
hearts, for that matter? Why does Hermes need to wear special shoes to
cover his tracks after he steals Apollo’s cows—why can’t he just fly over
the ground? And why does he throw those shoes into the Alpheus River to
get rid of them after he’s committed his crime? Why can’t a god who is able
to “twist sideways and slip through the latch-hole” of his mother’s door
simply cause those shoes to disappear?37 Instances like these once again
show that narratives evolved to portray gods who were subject to at least
some of the same physical limitations as humans, at least some of the time.
Of course, fallible gods like these make for better stories. If Apollo, in
the Homeric Hymn to Hermes, had known that Hermes had stolen his cows
as soon as it happened, and if he had been able to recover those cows with
just a snap of his fingers, then the rest of the tale would have disappeared.
We would never hear about Hermes’s charming pretense of being “just a
baby” when Apollo confronted him in his cradle, his comic fart when
Apollo manhandled him, his skillful use of his newly invented lyre to
soothe Apollo and the brothers’ eventual friendship. And if Poseidon had
been omniscient and omnipotent, Odysseus never would have gotten as far
as Phaeacia, much less Ithaca—and we would have no Odyssey to entertain
us now.

Templates and Their Violation
But there is more behind the gods’ fallibility than just the yen for good
stories. The anthropologist Pascal Boyer and the psychologist Justin Barrett
have studied the ways that people tend to think about Invisible Others
differently on different occasions. On some occasions, people engage in
what Barrett calls “theological correctness.”38 That is, they are careful to
talk about Invisible Others—who they are, what they can do, how they
interact with humans—according to what their society officially thinks.39
On other occasions, when people are thinking and speaking more casually,
they reveal their underlying, and probably unconscious, assumption that
Invisible Others are bound by at least some of the same rules of biology and
physics as humans are. Barrett, whose work focuses on American
Christians, ran an experiment in which his subjects read a story about God
saving a man’s life and at the same time helping a woman find her lost
purse. When the subjects were asked to tell the story in their own words
later on, many of them said that God first had helped either the man or the
woman and then later had helped the other one. When thinking “officially,”
Barrett’s subjects accepted the idea that God could be in more than one
place at once, doing more than one thing, but when telling a story about
what God actually did, many slipped into imposing upon God the same
rules of time and space as the narrators themselves experienced every day.40
One reason that there are dual sets of expectations for Invisible Others is
that people think in terms of what Boyer has called “ontological templates.”
These templates are blueprints for the basic ontological categories that exist
in the world.41 Boyer proposes that there are very few of these; in his work,
he mentions only five: person, animal, plant, natural object, tool.
As children mature, Boyer suggests, they intuitively learn which
characteristics define each template and apply that knowledge to new things
that they encounter. Thus, if a child has learned that cats and dogs must eat
and sleep, that they give birth to creatures that grow up to look like
themselves and that they are members of a group called “animals,” then the
next time that the child encounters something that he or she is told is an
“animal”—say, a walrus at the zoo—the child will infer that the walrus
must also eat and sleep and that it will give birth to little walruses.
Eventually, the child will automatically assume that all animals eat, sleep,
give birth to creatures that look like them, and so on.

Supernatural concepts are ontological templates that have been altered by
either the subtraction or the addition of salient characteristics. A ghost, for
instance, is a person from whom has been subtracted physical substance. A
statue that can hear your prayers is a tool to which has been added cognitive
functions. In the majority of religious systems, Boyer observes, gods can
similarly be understood as persons to whom have been added special
cognitive powers (omniscience), physical powers (omnipresence, flight, and
so on) or both, and from whom characteristics such as mortality have been
subtracted.
Barrett and Boyer have shown that such violations are more easily
remembered than what they call “violations of expectations” or “mere
oddities”—things that are possible within the everyday world but highly
uncommon, such as a table made out of chocolate, a man who has six
fingers or a woman who has given birth to thirty-seven children. This is
probably both because the counter-ontological violations constitute more
extreme departures from what we encounter in everyday life than mere
oddities do, but also because many of the counter-ontological violations are
capable of creating situations that are highly significant to our lives. Or in
other words: the chocolate table is something we can consume and the six-
fingered man or the amazingly fecund woman are people whom we can
marvel at, but a person who knows everything and can do anything (i.e., a
god) effectively holds the power of life and death over us.
A delicate balancing act is constantly going on in the imagining of a
counter-ontological entity, however. If too many violations are added to a
template, people stop believing in the resultant entity because too much
cognitive energy is required to override so many of the deeply engrained
inferences that are associated with that template. The entity becomes, quite
literally, “incredible.” If an author described the ghosts in a story as lacking
physical substance and lacking cognitive functions and having the power of
bilocation and sustaining themselves exclusively on the bones of
salamanders and being active only during months with an r in their names,
the ghosts would become too cognitively expensive for anyone to maintain
interest in them even while reading the story, much less as a serious belief.
Successful counter-ontological entities avoid this problem in either of
two ways. The entity may include only a few violations of the template
upon which it is built, leaving people to infer that in all other ways it
operates just like normal examples of that template. Thus, we accept that a

ghost has no physical body and can therefore go through walls, for instance,
but when a ghost manifests itself in a story, we expect it to be able to see,
hear, and understand what is happening in its surroundings because other
members of the person template can do so. Or the entity can be said,
officially, to include more than a few violations, but any given narrative
about that entity evokes only some of those violations—other “official”
violations are ignored or even contradicted in the course of the narrative.
This is effectively the mode of control to which the people interviewed by
Barrett subjected God when they rewrote the story of Him saving a man’s
life and at the same time helping a woman find her lost purse.
This brings us back to the instances of slippage in the stories about the
Greek gods that I collected a few paragraphs earlier—Zeus’s failure to
perceive what Hera was up to when she seduced him, Apollo’s lack of
knowledge about his stolen cows, and so on. These are excellent examples
of the second way of dealing with multiple ontological violations.
Interestingly, almost all of them come from the Iliad, the Odyssey, and the
Homeric Hymns. Other genres were more insistent about the expansiveness
of the gods’ powers. Tragedy, for example, includes statements such as this
one from the end of the Bacchae: “The gods have many shapes. The gods
bring many unexpected things to pass. Things that seem likely are not
accomplished, while a god finds a way to accomplish unlikely things. And
thus has the present matter been concluded as well.”42 Epinicians reject and
correct well-known stories of the gods’ shortcomings. Instead of the
familiar story of a raven informing Apollo of Coronis’s infidelity, for
instance, Pindar gives us this: “But she did not escape the watching god, for
although he was in flock-receiving Pytho as lord of his temple, [Apollo]
perceived it, persuaded by the surest confidant, his all-knowing mind.”43
Greek mythic narratives, then, served two complementary roles in the
construction and maintenance of belief. It was easier to believe that gods
such as those described in the epics and the Homeric Hymns existed
because they aligned with the person template fairly closely. They had
counter-ontological features but not so many of them, or of such severity, as
to require a great deal of cognitive energy to maintain belief. In contrast, the
gods presented in tragedy and in epinicians tended to be significantly
different from humans. As such, they could better be used to explain certain
aspects of the way the world worked and to elicit thought about how, if at
all, humans could hope to improve the circumstances of their existence.

(The distinction I am drawing here is heuristic; I do not mean to imply that
epic and the Hymns never prompted questions about the limits of human
existence, or that tragedy and epinician never showed gods who fail in
knowledge or power. Rather, I want to point out that the Greeks found ways
to think differently about the gods on different occasions.) The strength of a
religion depends, I suggest, on the availability of such a variety of
narratives—on not only having vivid, engaging narratives that create
cognitive and emotional bonds between audience members and the Invisible
Others whose stories are told in those narratives, but also on having
different kinds of engaging narratives.
Names
In arguing that fictional characters had unified existences outside of their
narratives, Maria Reicher suggests that the only thing that was genuinely
vague or nonspecific about them were their names. Thus, Reicher says,
“Faust is not a vague object. If anything is vague at all, then it is the
meaning of the name ‘Faust.’ It is vague in the sense that different users (or
the same users on different occasions) use the name to refer to distinct
characters.”44
Implicitly, this suggests that it is a name, first and foremost, that holds
together the different—perhaps the very many, very different—
instantiations of a character. This is what makes a plurimedial, accretive
character work: the name corrals a large and varied collection of ideas and
opinions about that character, and the continuing use of the name
steadfastly insists that there is a single figure behind them all somewhere,
even if opinions differ as to who exactly that figure is. This suggests that
the continuing use of a name is particularly important for sustaining belief
in an Invisible Other who can never be completely known and about whose
actions and experiences there is no complete agreement. Indeed, the more
varied the opinions about that Invisible Other, the more important the name
is likely to be.
But by the same token, names are extraordinarily rich reserves of
significance. From behind a character’s name shimmer forth not only bits of
the different portrayals of that character that any single person has
consumed, but also those aspects of the character’s history upon which
everyone agrees. The name “Orestes” will always evoke the act of
matricide, however else his story may be changed, and “Antigone” will

always evoke the defiant burial of a brother. The name “Hera” is likely
always to evoke her marriage to Zeus,45 and “Demeter” a mother’s grief,
whatever other roles these gods explicitly play in a narrative. These
associations intone, however subtly, every appearance of that character. A
name also evokes, however subtly, the familial line to which the hero or god
belongs and the other gods and heroes of the larger network to whom he or
she is connected. This power of names to evoke whole histories is the
reason that alternative names for certain gods are sometimes used in cult
worship, as opposed to myths. Both the name Kore (literally, “Girl”) and a
variety of names for Hades (such as Clymenus, “The Famous One,” or
Plouton, “The Wealthy One”) are used in cult much more frequently than
are the names “Persephone” and “Hades,” for example. “Persephone” and
“Hades” were so fraught with meaning, so powerfully evocative of the most
famous story about these two gods—the rape of Persephone by Hades—as
to be ill omened in ritual settings. The names “Kore,” “Clymenus,” and
“Plouton” did not erase the interpersonal history and the characteristics
implied by the names “Persephone” and “Hades,” but they emphasized the
gods’ kinder sides.46
The memories of a character’s history that a name can subtly induce
usually fly below the audience’s conscious radar, but skillful narrators know
how to bring them forward to good effect when they choose. In the first line
of the Iliad, for example, the patronymic “son of Peleus” suffices to
fleetingly evoke, for a mythically embedded listener, the entire story of how
Achilles was sired upon a goddess by a mortal man and the tale of how the
gods’ self-interested machinations led to that liaison—both of which
underlie the problems that drive the rest of the poem. Had the poet chosen
one of the other formulaic epithets that were frequently applied to Achilles
—“swift-footed” or “lion-hearted,” for instance—the effect would have
been quite different. Poets could also play against the resonance that a
mythic name carried. Pindar introduced Medea into his narrative of how
Cyrene had been founded without mentioning the famous episode during
which Medea killed her children and ruined her husband’s life. Yet the
audience’s knowledge of it problematized Pindar’s portrayal of Medea as a
sort of epinician Muse—and thereby lent greater stature to Pindar’s claim,
within his poem, to have conquered her and utilized her gifts.47
Perhaps Roland Barthes was right that there are no objects that are
inevitably and eternally sources of suggestiveness, but within the culture

that possesses them, the names of gods and other Invisible Others are, in
fact, enormously suggestive.48 In contrast, novels, to take up another form
of narrative, must spend a great deal of time developing their characters
because those characters are new to readers; the names mean little or
nothing when the novel begins (unless the novel is one in a series). In
contrast yet again, most folktales and fairy tales rely on generic characters
(the hero, the princess, the stepmother) from whom one expects certain
generic qualities (bravery, beauty, cruelty) but who typically have no real
names, and therefore no individualized personalities or histories.49 If they
have names at all, they are usually descriptive of their most salient feature
or function in the story (“Sleeping Beauty,” “The Little Mermaid,” “Clever
Hans”). In the economy of narrative, then, a myth speaks more efficiently
than a novel (it can convey more ideas more concisely, because it is built
upon an established body of knowledge about who the characters are), and
it can convey more ideas, or ideas of greater complexity, than a folktale or
fairy tale because its characters have more individualized depth, which can
be encapsulated into the names themselves. The closest analogy for what
myth accomplishes through its named actors, and how it does so, probably
comes from contemporary fan fiction. The characters who populate fan
fiction already possess well-developed personalities and histories that they
carry forward into new narratives—Harry Potter, Hermione Granger, Ron
Weasley, and others from J. K. Rowling’s novels are popular cases, as are
characters from the Star Trek oeuvre, from Doctor Who, and from the
Ender’s Game trilogy. So thoroughly versed in the characters’ personalities
and histories are the audiences of these fan narratives that, as the media
scholar Henry Jenkins puts it, “a look, a raised eyebrow, the inflection of a
line” is sufficient to evoke a character’s total life experience.50 By drawing
actors from a pool of characters whose names are already associated with
well-known histories and personalities, both myths and fan narratives are
able to gesture towards a great deal more than they state; they can evoke
ideas or themes that need not (and sometimes should not) be made explicit.
Double Characters and Double Names

No god shares a personal name with anybody else, nor does a hero who
plays a major role in myths. “Theseus,” “Heracles,” “Perseus,”
“Bellerophon,” and “Medea,” for example, clearly denote only one
particular person. But the situation is different with less prominent

characters: we have more than one Leucippus, Atalanta, and Erigone, for
example, and quite a few Agenors. In many cases, the shared name is
transparently descriptive, and the stories about those who bear it reflect its
meaning in different, but parallel, ways. Boys named “Leucippus” (White
Horse) reflect a widespread Greek metaphor that compares young men on
the brink of maturation to unruly horses. All three myths that are told about
a boy named Leucippus articulate the danger of entering into sexual
maturity in the “wrong” way.51 Similarly, both characters named “Atalanta,”
whose name can be understood to mean either “Untiring Girl” or “Equal in
Weight [to men],” are involved in physical pursuits otherwise restricted to
males and prove to be their matches. In the case of both Leucippus and
Atalanta, we can probably assume the independent generation of stories
starring a character on whom narrators bestowed an appropriate name.52
Other shared names are useful stop-gaps for characters who play relatively
small roles in the myths of other people. Like the names found in fairy
tales, they are usually straightforwardly descriptive. “Agenor” (Valiant,
Heroic), for example, is never the star of a myth himself but rather the
father, son or brother of more prominent characters. Occasionally, however,
an author brings such a character into his or her own. “Creusa” (Princess) is
a stop-gap name for the daughter of a king, including, for example, Creusa
the daughter of King Creon of Corinth, whom Medea killed, and Creusa the
daughter of King Erechtheus of Athens, who was mother of the Athenian
dynastic hero Ion. Euripides, in his Ion, crafted that Creusa into a woman of
independent thought who acted in ways that threatened Ion’s survival and
rise to power. In a few other cases of double names, such as that of Ajax,
we seem to deal with two genuinely separate characters who happen to
share a name—a name whose etymology is obscure, perhaps because it is a
survival from a pre-Greek language.
There are also names that send no clear message in and of themselves but
that have come to be associated with a particular type of fate or experience.
“Erigone” is a good example. The name “speaks” only insofar as the “-
gone” syllable implies a noble birth; she is, like most of the characters in
Greek myths, from the upper class. (The “eri-” means “early”—whatever
we want to take that to mean.) Erigone’s story, or at least its climax, always
takes place in Athens, she is always a maiden who hangs herself, and her
ghost always threatens to drive all Athenian maidens to do the same if she
is not propitiated annually during the Anthesteria festival.

Two quite different stories of why Erigone hanged herself became
attached to the name, however, each of which performed its own particular
task. In one case, Erigone hanged herself in grief because her father, Icarius,
had been murdered by other Athenian men. Icarius, an emissary of the god
Dionysus, had introduced the men to the gift of wine and, never having
been drunk before, the men concluded that Icarius had poisoned them and
therefore attacked him. This story works to tie Erigone into the central
concern of the Anthesteria as we know it in historical times, which was the
opening of the new wine casks and more generally the celebration of
Dionysus. We might guess that the story was created in order to pull the
ritual in which Erigone’s ghost was appeased, which probably had once
stood independent from the Anthesteria, more firmly into the Anthesteria’s
orbit. (Icarius’s story was probably always associated with the Anthesteria,
in which case those who invented this myth about Erigone found him a
handy figure upon whom to graft her.)
The second story about Erigone probably emerged only after the ritual
meant to appease her had already been incorporated into the Anthesteria. It
makes Erigone the daughter of Clytemnestra and Aegisthus, who wanted to
see her half brother Orestes punished for her parents’ murders. When an
Athenian jury found Orestes not guilty, Erigone hanged herself in grief.
This story, as well as another one that told about Orestes’s arrival in Athens
while local men were celebrating the Anthesteria, served to tie Athenian
myths and rituals into the prestigious epic cycle in which Aegisthus,
Clytemnestra, and Orestes already played starring roles.
The essence conveyed by the name “Erigone,” then, was that of a girl
who hanged herself and cursed Athenian maidens to do the same if she
were not appeased; any story attached to her name, whatever else it might
do, had to serve as an aition for an important Athenian ritual attached to
Erigone and therefore had to include her suicide. But her name remained
vague enough—malleable enough—that the stories attached to it could be
molded to suit other purposes, as well: to bind together disparate parts or
themes of a festival complex or to graft older Athenian traditions into a
newer, and more glamorous, body of narratives.
Final Words
Ontologically, the Greek gods and heroes are an interesting lot. More fluid
than even the most frequently presented characters of fiction, they

nonetheless held on firmly to core identities through centuries of continual
reinvention. Greater enough than humans to merit worship, they
nonetheless were similar enough to humans in their failings to be credible.
It was through myths that these productive tensions were held together.
What would scarcely be convincing or memorable when stated as a formal
credo played out extremely well as stories, and it was through those stories,
therefore, that the Greeks were inculcated with a sense not only of the gods’
and heroes’ existence, but also of who their gods and heroes were and how
they must be approached.
Personal names were an important part of this process. “Zeus! Whoever
he may be! If it pleases him to be invoked by this name, then by this name I
will call to him! Having considered all my options, I have nothing so good
as the name of ‘Zeus’ to help me cast aside my burden!” cries the chorus of
Aeschylus’s Agamemnon, desperate to find a solution to their problems. As
they continue with their plea, they briefly narrate a story that was surely
familiar to everyone: the first, insolent ruler of the cosmos was overthrown
by his son, and that son was, in turn, overthrown by his own son—Zeus.
The chorus chooses not to mention the names of those first and second
rulers, not because they were ill omened (the names had always been in
widespread, common use), but rather so as to shine an even stronger
spotlight on the glorious name of Zeus, which they immediately repeat
again after they have finished narrating their succession story.53 Zeus is a
god so great and so varied that only his own name can capture all that he is,
folding into a single word everything that poets and artists could ever tell us
about him.

6
Metamorphoses
OVER THE PAST few years, I’ve learned that when my grandson asks me to
tell him a myth, the myth I tell must include at least one of two things: an
episode in which something or someone turns into something else (a
metamorphosis) or a creature made up of parts from other creatures (a
hybrid). If I really want to please him, the myth I tell must include both a
metamorphosis and a hybrid. Having exhausted my supply of existing
stories that meet these criteria (and silently justifying my actions on the
basis of some of the principles that I’ve discussed elsewhere in this book),
I’ve developed a repertoire of tales in which the Chimaera gives advice to
Arachne, for instance, or young Pegasus takes flying lessons from Ceyx and
Alcyone. Sooner or later, when my grandson reads other, more canon-
bound narrations of Greek myths, I’ll have to admit that the Chimaera as we
traditionally know her wasn’t the sort of character who would be helpful to
anyone else (even a spider) and that the circumstances surrounding Ceyx’s
and Alcyone’s transformation into seabirds probably left them in no mood
to tutor a horse. But for the moment, the more metamorphoses and hybrids I
can weave into my stories, the more attractive they are, and therefore I give
my imagination free rein.
Tales of transformation, populated by strange creatures, appeal to most of
us, whatever our age, and over the course of the centuries, humans have
told a lot of them. In the Old Testament, Lot’s wife turned into a pillar of
salt when she looked back at the burning city of Sodom—a transformation
tale with which the New Testament Jesus admonished his apostles, lest they
tarry on Judgment Day to gaze at terrestrial spectacles. The Hindu god
Prajapati lustfully pursued his daughter; she fled from him in the form of a
doe, but Prajapati became a stag and raped her—a scandalous act that drove
the other gods to create Rudra, who hunted down Prajapati and thereby
earned the title “Lord of Animals.” Prajapati’s tale finds echoes in Greek

stories of goddesses turning themselves into mares to evade lecherous gods,
who then turned into stallions—which led, eventually, both to the birth of
wonder horses and to the establishment of new rituals to appease the angry
mothers. We hear more distant echoes in the stories that American Indians
told of women metamorphosing into human-horse hybrids after falling in
love with stallions. The horse-women gave birth to valuable ponies but
became outcasts from their tribes or were even killed—yet another variation
of the endless human thinking about what our relationship to animals could
and should be. So, too, the world has known many stories about a girl who
tries to rescue brothers who have been turned into birds, sometimes by
clothing them in human attire that she has sewn herself. Racing against the
clock, in some of these stories she fails to sew a single sleeve on a single
shirt, and the brother who dons it carries a wing in place of an arm
forevermore.1
We have Ovid to thank for the fact that in the Western world, our appetite
for metamorphosis has always been fed to a significant degree by Greek
myths. Whatever his formal and ideological reasons for taking
metamorphosis as his theme, Ovid reveled in the opportunity to describe the
fantastic, and the many artists and authors whom he inspired—Bernini,
Dante, Titian, Caravaggio, Rembrandt, Moreau, Wharton, Updike, and
Harryhausen (to name only a few)—have reveled in that as well, arousing
and then satisfying their audiences’ appetites. As a result, Greek myths
about metamorphoses and hybrids have remained so widely familiar that
they still can serve as fodder for comic strips (a 2012 Bizarro Comic by
Dan Piraro shows a bald woman sitting across the table from a man dressed
in bishop’s robe and mitre, with the caption “Ill-Fated Blind Date: St.
Patrick and Medusa”).2
Some of the examples that I gave earlier of stories about metamorphoses
and hybrids suggest that they offer more than just thrills or amusement,
however. Frequently, they have been adduced to explain how rituals, gods,
animals or interesting features of the landscape came into existence (late
antique writers identified a pillar of halite near Mount Sodom as what
remained of Lot’s wife; similarly the Weeping Rock on Mount Sipylus in
Turkey was said to be the transmogrified Niobe). The stories can also send
messages about how one should and shouldn’t behave—don’t look at what
a god has told you not to look at, don’t have sex with your daughter—or
with stallions—and do keep your fingers nimble enough to finish your

needlework on time. The ease with which myths could be straitjacketed into
explanatory or admonitory texts was one of the reasons, in fact, that Ovid
and the Greek myths he narrated were able to survive the Christian purge of
pagan detritus; one could transform almost any myth into a moral lesson of
which Jehovah would approve. In the Ovid Moralisé, Callisto, the innocent
victim of rape, became “a hypocrite whose chastity was merely for show,”
and Myrrha, the virgin daughter who secretly seduced her father, could be
compared to the Virgin Mary.3
Hybrids
In later sections of this chapter, I’ll discuss some of the ways that scholars
have tried to understand Greek tales of metamorphosis—why they were
told, what they meant to the tellers and their listeners, the “rules” by which
they seemingly worked. At that point, I’ll leave behind the topic of hybrids,
on which I’ll focus for the next few pages. The two are closely linked,
however: both hybrids and metamorphoses challenge the ontological
boundaries of the world and its creatures as we know them. Hybridity,
moreover, is sometimes the result of arrested metamorphosis, as in the case
of the brother who changes first into a swan and then back again into a
human but retains a swan’s wing, in the American Indian tales of women
who turn partially (but only partially) into horses after falling in love with
stallions, or in one of several stories about the origin of Scylla, whose upper
half remained a beautiful girl but whose lower body became a dreadful
confusion of barking dogs after she waded (but only waist high) into a pool
of poisoned water.4 Whatever may be the particular roles played by such
hybrids in the stories they inhabit, their very presence reminds us that there
are more things in heaven and earth than we have yet encountered.
Or more precisely, they remind us that new things may continue to
emerge in heaven and earth long after the formal creation of those spheres
has come to an end. Implicitly, this is one of the points made by many other
stories of metamorphosis, as well, even if they don’t produce what we
consider hybrids per se. This is most expansively so in Ovid’s chain of tales
borrowed from Greek sources. His epic starts with the self-assembly of a
rudimentary physical cosmos out of raw, disordered matter, and then
introduces an unnamed divinity who fashions the first plant and animal life.
From there, Ovid rolls on through generations of humans that see the
metamorphic emergence of swans, guinea hens, spiders, and other animals;

the celestial emplacement of the Big and Little Bears and the Corona
Borealis; the bursting forth of newly transmogrified youths and maidens
into trees, flowers, and springs; and so on, until he finally reaches the
catasterism of Julius Caesar—a modern metamorphosis that is meant to
crown all the others. As seen through the lens of the Metamorphoses, then
(and through the collective lens of the Greek authors who went before
Ovid), the world has taken many centuries to become what it is and may
still be a work in progress. It is also a place that has been created to a
significant degree out of the raw material of human bodies and by the whim
of the gods. Behind each man-turned-into-animal or girl-turned-into-plant
lay divine anger, envy, pride or desire. It was Pan’s desire that drove
Syrinx’s sister nymphs to turn her into river reeds, and in one version of the
story it was Circe’s jealousy that turned Scylla into a monster. Apollo
pursued Daphne until she was desperate enough to prefer life as a tree; a
spurned Leto changed peasants into frogs and a spurned Dionysus changed
a king’s daughters into bats. Callisto’s double transformation—first into a
bear and then into the Big Bear—was set in motion by a combination of
Zeus’s lust and Artemis’ anger—and, in some stories, by Hera’s anger as
well. It was Venus who orchestrated Caesar’s rise to the stars. The
continuing lability of the world that Ovid and his Greek predecessors
described, in short, was powered by gods’ emotions. Each transformation
was a memento of, and thereby evidence for, the gods’ ongoing, and very
personal, engagement with the world and its mortal inhabitants.
But to return to my main point, in spite of certain similarities and
narrative overlaps between hybrids and metamorphoses, there are also
differences. Most importantly, in Greek myths, hybrids are not usually the
products of sudden metamorphosis but rather are conceived, born, and
reared in more or less the normal biological way.5 The sea goddess Ceto
bore to her sea-god husband, Phorcys, both the Gorgons (women with
snakes for hair and, according to some authors, the bodies of lions and the
wings of eagles) and Echidna (who was half nymph and half snake).
Echidna, in turn, “mingled in love” with Typhon (himself a hybrid with a
hundred fiery snake heads) and bore to him the Chimaera and the two-
headed dog Orthrus. Then, having been raped by Orthrus, Echidna gave
birth to the Sphinx (woman from the breasts up, lion from the stomach
down, and winged).6 The bull-headed Minotaur was conceived during a
sexual encounter that, although certainly odd, unmistakably mimicked the

sort of thing that took place in pastures every day: Pasiphae, desiring the
great white bull that Poseidon had given to her husband, crawled into a
hollow wooden cow and awaited the bull’s attention.7 The centaurs were
born from Centaurus, a man “bearing honor amongst neither mortals nor the
laws of the gods,” and a herd of horses with whom he mated. Centaurus
himself was a child spawned by Ixion upon a cloud—but it was a cloud that
Zeus had shaped to look like Hera.8 Pegasus’s birth was unusual—he sprang
from his mother’s neck after she had been decapitated—yet his conception,
like that of so many other heroes and monsters in Greek myth, occurred
when Medusa and Poseidon “lay down in a soft meadow amongst spring
flowers.”9 Other monstrous creatures—the nine-headed Hydra, three-
headed Cerberus, three-bodied (and three-headed) Geryon, the one-eyed
Cyclopes—were similarly conceived during lovemaking between various
gods—most typically sea gods or their descendants.10
The Greeks were insistent, in fact, that monsters were once somebody’s
babies. Their pedigrees could be announced like those of the heroes. Most
famously, Polyphemus boasted to Odysseus that his father was Poseidon11
—indeed, the Cyclops’s paternity drives the rest of the Odyssey—and even
within our scant fragments of Stesichorus’s lost Geryoneis, Geryon
manages to announce that he is the son of Chrysaor (Medusa’s son by
Poseidon) and Callirhoe (an ocean nymph). Just in case we missed hearing
Geryon say it, Geryon’s friend Menoetes mentions it, too, and the narrator
mentions it a third time. Callirhoe herself pleads with Geryon not to meet
Heracles in battle and opens her robe to display the breasts that once had
suckled him—all three of him, I suppose, however odd a mental picture that
may be.12 In a fragment of Euripides’s lost Cretans, Pasiphae complains
about what it’s like to nurse a Minotaur.13 An Etruscan red-figure vase from
the second half of the fourth century, showing a young Minotaur on the lap
of an upper-class woman who looks disgruntled, evokes Pasiphae’s
complaint (see Figure 6.1). Yet another monstrous baby at the breast is
conjured up by the Homeric Hymn to Apollo, which tells us that after Hera
gave birth to Typhon—a dreadful creature “like unto neither gods nor
humans”14—she gave him over to be nurtured by Python, the baneful
supersnake. Hera herself nurtured the Hydra and the Nemean Lion,
according to Hesiod. The verb that I translate here as “nurture”—trepho—is
frequently used of nursing and rearing children, and the poets’ insertion of

these phrases immediately after they have described the monsters’ births
makes it clear that more than just metaphorical “nurturing” is meant.15
Someone, be it Hera, Python or another extraordinary nursemaid, had to
take these creatures in hand during those difficult years between infancy
and adulthood.
Behind all of these creatures’ monstrous features we glimpse a weird
version of the idea that heredity counts, just as it does for humans: if your
father was the Lord of Horses (Poseidon), you might end up as a winged
horse (Pegasus); if your mother was a snake from the waist down
(Echidna), you might have nine snaky heads (the Hydra) or a snake’s head
growing alongside those of a lion and a goat (the Chimaera). If your brother
was a two-headed dog (Orthrus), you might be a three-headed dog
(Cerberus). Behind the idea that heredity counts, we also glimpse a
determination to bind these hybrids into that network of relationships that I
discussed in Chapter 4. One way of being linked in was to be killed by a
hero, but another way was to be born from the gods. Most hybrids were
both.

6.1 An upper-class woman (probably meant to be Pasiphae) with a baby Minotaur on
her lap. Etruscan red-figure cup from the first half of the fourth century, now in the
Cabinet des Médailles, Paris (detail). Inv. num. 1066. Photo credit: © Bibliothèque
nationale de France / CNRS–Maison Archéologie & Ethnologie René Ginouvès.
No such genetic principle underlies myths of metamorphosis—

narratively, it is the whim of a god that determines the end point of a
change. It may seem as if a principle similar to heredity underlies many of
them: after all, “Spider”-girl (Arachne) becomes a spider, “Wolf”-man
(Lycaon) becomes a wolf, the youth named Narcissus becomes the
narcissus flower and the nymph named Laurel (Daphne) becomes the laurel
tree (daphnē). Doesn’t this amount to a nomen est omen sort of cosmogonic
thinking, a nominative determinism whose rules are just as inescapable as
those of biological heredity? But a moment’s thought suggests that most

such stories developed “backwards.” Someone, at some time, was asked
where the spider had come from, with all of its distinctive talents and
habits, and answered with a story involving a girl who shared the spider’s
talents and habits—and who gave the spider her name. It’s not only
metamorphic myths that work this way; it is, rather, a principle of
aitiological nomenclature that runs through many types of Greek myths.
Otherwise hazy characters with meaningful names hang off the branches of
royal family trees to explain (or, rather, to lay claim to) a kingdom’s
geographic reach, for example. Achaeus, a son of the Athenian princess
Creusa and her husband, Xuthus, furnishes an eponym for the Achaeans—
thus tying the Athenian line firmly to the luster of Greek glory as Homer
presented it—and then he disappears from view.
I will return to the issues raised by nomina that seem to be omina later in
this chapter. But before we leave the topic of hybrids, we should remember
that, as I noted in Chapter 4, many—perhaps most—of the monsters who
terrorize Greek myths are not hybrids or teratomes at all. They are simply
larger or stronger versions of the animals that populated Greek forests and
fields: the Erythmanthian and Calydonian Boars, the Crommyonian Sow,
the Mares of Diomedes, the Cretan and Marathonian Bulls, Python, the
Nemean Lion, the Teumessian Vixen, and Sciron’s gigantic turtle—who
doesn’t even look gigantic in some representations (see Figure 7.5). In
contrast to some other cultures’ myths, where hybrids proliferate, Greek
myths kept them at bay. Genetically speaking, moreover, those that did exist
tended to cluster on one branch of the divine family tree, which grew from
the union of Ceto and Phorcys. This may express the ungovernable,
unpredictable nature of the sea, an element with which the Greeks were
never completely comfortable.16 (In contrast, the main branch of the gods—
culminating in Zeus, his siblings, and his children—is almost completely
anthropomorphic, even if its members chose to disguise that fact on
occasion by turning themselves into all manner of animals.)17 When hybrids
do enter the Greek mythic story world, they enter not as Dunwich Horrors,
evoking existential revulsion as well as fear, but as variations on themes
that were visible in everyday nature—the bird, the snake, the fish, the horse,
the lion, the bull, and, just once, the goat—an odd detail in the Chimaera’s
makeup. In Chapter 4, I suggested that the story world of the Greeks is not a
strongly secondary one, and even here, in its determination to contain the
monstrous within certain bounds, it tethers itself firmly enough to the

Primary World to preserve credibility. What was truly amazing was the fact
that many creatures, plants, and minerals were created by divine fiat—and
those were almost always creatures, plants, and minerals that were familiar
from the everyday world.
But What Does It All Mean?

Is it really only divine whim that determines the end point of a
metamorphosis? Scholars have made an industry out of decoding stories of
transformation on the assumption that much more than that lay behind
them. One of the most enduringly popular approaches focuses on
determining what the animal, plant or mineral into which someone was
changed “meant” to the culture concerned. Sometimes, little work on the
part of the scholar seems to be needed; the answer already seems obvious
from the story. Lycaon, whose name is built on the Greek word for “wolf,”
slaughters a child in order to test the omniscience of the gods: will they
realize that the stew he is serving to them is full of human flesh? They do,
and Lycaon, punitively transformed into the wolf whose name he has
carried and whose outrageous alimentary habits he has adopted, snarls off
into the woods.18 Infanticide and cannibalism are wrong, the story tells us—
but didn’t we already know that? Is that really the reason that this story was
told?19 Similarly, the nymph Iynx, who used magic to seduce Zeus, was
transformed by Hera into a bird called the iynx, whose body was used in
love charms. Anthos, who was killed and partially eaten by horses, became
a bird called the anthos, which had a habit of fleeing from horses.20 Even
when the protagonist’s name does not proleptically signal his or her fate, the
plot of the story often makes the connection between personal behavior and
metamorphosis abundantly clear, or other clues seem to help us build a
reassuringly logical network of associations. It could be said to “make
sense” that Hera turned Io into a cow because Io was the priestess of Hera,
who was particularly fond of cows. The fact that Io later became identified
with Isis, to whom the cow was also sacred, rounds things off nicely.21
Callisto, a nymph who had sworn to emulate Artemis’s eternal virginity,
was raped and impregnated by Zeus; Artemis punished her by turning her
into a bear, a fate that can be said to “make sense” for any or all of several
reasons: bears are creatures of the wild and thus fall prey to the goddess’s
arrows, for example, and / or Callisto became the mother of Arcas (“Bear”),

the eponymous ancestor of the Arcadians.22 The Lycian peasants who
purposefully muddied the water that Leto and her infant twins wished to
drink were turned into frogs, which “makes sense” (as Leto herself
explained, in Ovid’s narration of the story) because frogs are repulsive
creatures who spend their lives wallowing in mud.23 It’s not only the Greeks
and Romans who played these games. As one Jewish tradition has it, Lot’s
wife begrudged her husband’s angelic visitors the salt that was a basic
requisite of hospitality: ordered by her husband to provide it, she
purposefully betrayed the angels’ presence by borrowing extra salt from her
neighbors.24 And so, a quintessentially bad hostess found an appropriate end
as the very stuff she should have been serving.
But there are problems with tying things up so neatly. First of all, some
of the stories I just mentioned have variants in which the narrative logic that
glues them together doesn’t work so well. It’s not always Hera who turns Io
into a cow, in order to punish her; more often, it’s Zeus himself,
transforming the lover whom he is trying to hide from his wife into the very
creature that would be likely to attract her attention.25 It’s not always
Artemis who turns Callisto into a bear, either—sometimes it’s Zeus or
Hera.26 We might still argue that cow-hood and bear-hood make sense
insofar as these heroines become emblematic of the goddesses whom they
first served and then offended or make sense in some other encoded way
(the bear, which was understood in antiquity to be an especially maternal
animal, could be taken to represent the mothers that all young dedicatees of
Artemis must one day become, when they left her care—hence, the ritual
called “playing the bear” that ancient girls performed in honor of Artemis at
Brauron, for example),27 but in championing these views, we’d still have to
admit that as part of a story, the transformational logic we expect to see no
longer works, and we’d also have to concede that some of our own analyses
would have been far less obvious to many members of an ancient audience
than they are to us (was mint, formerly the nymph Mintha, really
understood by the ancients to be diametrically opposed to the cereal grains
associated with Mintha’s persecutor, Demeter, as Marcel Detienne
suggests?).28
Another problem is that there are stories in which a name may
foreshadow a transformation, but the behavior does not. Anyone could
guess that a girl named Arachne would end up as a spider, and we might

even guess that the story of how that happened would involve spinning or
weaving, which are well-known occupations of spiders. But the element of
hubris that motivates the story as Ovid tells it is not otherwise associated
with spiders in any of our ancient texts.29 Ovid’s story of Arachne is
different from the story of Iynx, then, in which the use of sex magic
“logically” leads to the nymph’s new form and function as the magical bird
to which she gives her name. Other, more appropriate motivating behaviors
(appropriate, that is, according to the ancient understanding of spider
behavior) could also be used to explain Arachne’s transformation, as we’ll
see, but it’s Ovid’s tale that remains well-known; it sticks in the mind and
resurfaces over and over again in modern anthologies of ancient Greek
myths. A big part of the reason for this, again, is Ovid’s brilliant narration:
we remember a story because it is intrinsically interesting, not because (or
at least, not primarily because) it encodes information that makes sense in a
strictly logical way according to the precepts of natural history, alimentary
codes, astronomy, ritual practices or anything else. Things usually run
exactly the other way, in fact; if the story is engaging, it is better able to
send messages about natural history, alimentary codes, astronomy, ritual
practices, and so on. Scholars of religion have had trouble accepting this in
their dealings with myths—that is one reason that the myth-and-ritual
approach that I looked at in Chapter 2 of this book took hold so strongly
and for so long, and the structuralist and psychological approaches as well.
Each offered a way of reading myths that made sense to twentieth- (or
nineteenth-, or twenty-first-) century scholars. The assumptions behind
these approaches are not completely wrong—each has the merit of
revealing some of the ways in which myths affect or reflect the cognitive,
emotional, and social worlds of their audiences—but they tend to ignore the
hook embedded in the sheer pleasure of the story itself.
There are also cases in which the kernel of a good story takes off on its
own, leaving behind some or all of the “logical” connections between the
main characters and their transformations as it develops new vehicles in
which to travel. I’ll take the various stories about the origin of the
nightingale (aēdōn) and the swallow (chelidōn) as a case study.30 Each of
these centers on a mother who has killed her child and who, as a result, is
turned into a mournful bird—either as a punishment or out of the gods’ pity.
In the earliest such narration, in Odyssey 19, Penelope calls this mother
Aedon (Nightingale), the daughter of Pandareus, and explains that Aedon

eternally laments her child, Itylos, whom she slew while maddened.31 This
story “makes sense” because the song of the nightingale was perceived by
the ancient Greeks, like many peoples, as having a mournful sound quite
apart from its association with a lamenting mother.32 But two early allusions
to similar stories, from Hesiod and Sappho,33 call the main character
Chelidon (Swallow), the daughter of Pandion, and we presume that if we
had the fuller stories behind these allusions, we’d learn that a swallow is
what she was turned into. “Logically,” this makes little sense, given that the
song of the swallow was characterized by the Greeks as a twittering noise
similar to the speech of barbarians34—far from the melodious, if mournful,
sound associated with a lament. These stories have relinquished, in other
words, any obvious connection between their main character’s behavior and
the behavior of the animal that she becomes. In a fragment from another
Hesiodic work, we learn that both Aedon and Chelidon suffered eternally
because of a crime they had colluded in committing. Aelian, the author who
transmits the fragment to us, goes on to say that Aedon and Chelidon
committed their crime during a feast held in Thrace.35 Judging from what
we hear about that feast in later sources, the crime was again infanticide.
We learn more about these two groups of stories in narrations from the
fifth century onward. In the first group, a daughter of Pandareus, acting
alone, kills her own son, whose name is Itys or Itylus. She turns into a
nightingale who laments her slain child forever. Three narrations of this
story are set in Thebes;36 other versions are set on the island of Dulichium
or in Ephesus.37 In the second group, two daughters of King Pandion of
Athens, one married and one unmarried, collaborate in slaughtering the
child of the married sister in order to avenge the fact that her husband, a
Thracian king named Tereus, has raped the unmarried one. The sisters serve
the dismembered child to his father in a stew. The sisters (and the husband)
are turned into birds—a nightingale, a swallow, and either a hawk or a
hoopoe—and the nightingale laments her slain child forever.38
The name of the main character and the bird into which she is
transformed remain stable in the first group of stories as time goes by—she
is always called Aedon, and she always becomes a mournful nightingale
whose name she proleptically bears. Plot details do change, however:
sometimes Aedon mistakenly kills her son thinking that he is the child of
her sister-in-law, whose fecundity she envies;39 sometimes she kills him

because she thinks he has aided his father’s adultery;40 sometimes no reason
is given other than “madness.”41 In contrast, by at least the fifth century, the
names of the characters in the second story begin to vary. The role played
by Aedon is often filled by Procne, daughter of King Pandion of Athens,
and her sister is called Philomela rather than Chelidon.42 Not only have we
lost any direct aitiological connection between the women’s names and the
birds into which they are eventually transformed in these stories, but
another “logical” connection disappears as well. Behind the name
“Philomela,” the Greeks (and the Romans) would have heard the words
“lover of song” (philo-mela)—an apt name for a nightingale but not for the
twittering swallow that Philomela would become.43 The poor fit between
name and fate is “fixed” in stories told by Sophocles and later authors,
according to which Philomela’s rapist cut out her tongue to prevent her
from telling what she had suffered; she could make only the most guttural
of sounds and was compelled to communicate instead through tapestries
that she wove.44 Ovid, a great fan of nominative determinism, conveniently
sidesteps this problem by not telling us which girl turned into what bird—or
even exactly what the birds were. Later narrators eventually “correct” the
story by reversing the transformations: Philomela turns into the nightingale
and Procne into the swallow. Some of these narrators also omit the episode
of the tongue cutting, so that the nightingale can still sing melodiously.45
But in those cases, we end up with the aunt of the slaughtered child
becoming a mournful nightingale and his mother a twittering swallow,
which upends the familial logic of earlier versions even as it attempts to
restore ornithological propriety. One of the latest narrators of all,
Eustathius, once again tries to impose sense by remodeling the early part of
the story. Philomela becomes the wife of Tereus and Procne becomes the
sister whom he rapes and whose tongue he cuts out. But Eustathius then
reverts to the older stories at the end by having Procne turn into the
melodious nightingale, in spite of her severed tongue, and Philomela into
the twittering swallow.46 And thus, we are back in the soup.
What are we to make of all this? How did the story of the bird who
mourns for the child she slaughtered survive, even thrive, in spite of the fact
that for centuries it didn’t make much sense, aitiologically speaking and
sometimes narratively speaking? Clearly, we need to try a different tack. As
Wendy Doniger has pointed out, myths are mercenaries, serving the needs

of whoever expends the effort to narrate them effectively.47 In the myths
that we have been looking at, what catches the imagination and remains in
the mind—what makes these myths powerful mercenaries for anyone who
cares to hire them—are two sensational elements: a mother kills her child
and the mother is subsequently transmogrified into a bird. Once it has used
these elements to hook us, the myth can expand itself to provoke thought on
various issues (e.g., the debt a woman owes to her natal family versus the
debt she owes to her marital family), can serve as an anchoring backstory
for various ancestral or social groups (e.g., Athenian women, by
Demosthenes’s reckoning),48 and can furnish various points of meditation:
the nightingale story can be evoked by a lonely wife who fears for her son’s
safety (Homer’s Penelope) and by a friendless woman to whom the
infanticidal mother’s fate looks better than what she herself is about to
suffer (Aeschylus’s Cassandra)—neither of whom adheres very closely to
its original “logic.” Formally speaking, on one level these are all still stories
about the origin of the nightingale and the swallow, but the ease with which
they slip free of certain moorings that seem essential to us tells us that
transformations into animals in Greek myths are not, first and foremost,
encoded lessons in etymology or natural history. It also tells us that the
search for a single interpretative strategy that can be applied to all of them
—or even a set of interpretative strategies—is fruitless.
My second case study will be Hecabe—the very poster child of
interpretative difficulties for those who seek tidy correspondences between
a character’s backstory and the nature of the beast into which she is
transmogrified.49 We first get a full tale of her travails from Euripides, in his
play of the same name: when Hecabe learns that King Polymestor has
treacherously murdered her son, she kills his children and blinds the king
himself. Polymestor then prophesies that Hecabe, who is about to set sail
for Greece as part of the Trojan war booty, will be transformed into a dog
with fiery eyes, run up the mast of the ship that bears her, leap overboard,
and perish at a place along the Thracian coast that would thereafter be
known as Cynossema—the Dog’s Tomb.50 Other stories have Hecabe
leaping into the sea in grief as soon as Troy has fallen and then turning into
a dog;51 or being stoned to death by Polymestor’s followers and
simultaneously turning into a dog;52 or being stoned to death by the Greeks
for unspecified reasons and then turning into a dog;53 or being stoned by

Polymestor’s followers, turning into a dog, and yet surviving to wander the
Thracian wilds;54 or turning into a stone dog while still in Troy;55 or simply
turning into a dog, for unspecified reasons, after her son Helenus has taken
her safely to a new home.56 The only consistent element amongst these
remarkably varied stories is her transformation into a dog.57
This metamorphosis is likely to bedevil one, if one looks for consistent
logic behind it. What does the dog mean? Exactly how does it reflect
anything that Hecabe did, or was, in this variety of stories? Is there,
perhaps, an inherent savagery in the ancient Greek view of the dog that
resonates in Hecabe’s deeds as we see them in Euripides’s treatment of her
story, for instance? This would seem to be hinted at in our earliest trace of
Hecabe’s transformation: an unattributed lyric fragment says it was the
Erinyes who turned her into a howling dog with flashing eyes.58 The
Erinyes, after all (who were occasionally portrayed with dog-like features
themselves), were goddesses bent on avenging wrongs, sometimes with
dreadful ferocity. We could choose to understand this story to be telling us
that, like the Erinyes, Hecabe did something that was dog-like in its
savagery. Hecabe’s declaration, in the final book of the Iliad, that she would
like to sink her teeth into Achilles’s liver,59 might have laid the groundwork
for such an association—or might have reflected that association, if the
story of her metamorphosis into a dog was older than the Iliad. But still we
must be cautious: we have no direct statement that Hecabe was turned into a
dog because she was savage (and indeed, in some of the stories about her
transformation, she does nothing savage at all). There is, moreover, no
simple, one-to-one association between dogs and savagery in antiquity:
dogs could also be portrayed as loyal friends, for example, or trustworthy
guardians and, on the negative side (especially for women), as shameless as
well as savage.60
Or perhaps (as at least one scholar suggested) Hecabe’s transformation
into a dog was inspired by the similarity of her name to that of the goddess
Hecate, to whom dogs were sacred.61 Temptingly, a fragment from another
work by Euripides does seem to bring the two together, when someone
proclaims (to Hecabe, scholars presume), “you will become a bitch, the
delight of light-bearing Hecate.”62 This line was well enough known for
Aristophanes to spoof it,63 and Lycophron ran with its apparent connection
between Hecate and Hecabe: in his story, Hecabe becomes part of the

goddess’s uncanny nocturnal kennel, charged with terrifying anyone who
fails to worship her. In return, Lycophron goes on to say, Hecate compels
Odysseus, who had thrown the first rock at Hecabe’s stoning, to erect a
cenotaph in her honor.64 But explaining Hecabe’s metamorphosis by
reference to Hecate (that is, as a sort of one-off application of the nomen est
omen principle) is a ticklish affair, for two reasons. First, the earliest
possible attestation of the connection between Hecate and dogs comes from
the very fragment of Euripides that is presumed to describe Hecabe; the
argument becomes circular, in other words. Second, it’s likely that Hecate’s
own connection to dogs originally sprang from her work as a birth goddess
(to whom dogs were sacred in antiquity more generally), rather than any
uncanniness that the dog and the goddess later came to share.65 Of course,
one might simply switch strategies and start anew from the association
between dogs and birth, arguing that it was Hecabe’s fame as a mother that
led to her transformation into a dog (she did bear nineteen sons to Priam,
after all, as well as an undisclosed number of daughters). A combination of
the two aspects—savage and maternal—might be better still: the ancient
Greeks, like us, used the simile “a bitch protecting her pups” to describe
savage attacks on behalf of something one loved—the very sort of deeds
that Euripides and some later authors described Hecabe as committing.66
But all of this is learnéd speculation—none of these suggestions is
provided by the ancient stories themselves, and none provides the same
satisfying click of a puzzle piece that we get, for example, from Lycaon’s
metamorphosis into a wolf following his cannibalistic feast or Iynx’s
transformation into a tool for love magic after seducing Zeus. This prompts
two observations. The first is a reiteration of something I’ve already said
several times in this book: myths were able to communicate ideas and
emotions first and foremost because myths were entertaining, not because
they were encoded with information that was intended, first and foremost,
to be puzzled out by the audience members. Later in this chapter, I’ll
consider more closely what it was that myths of metamorphosis, as a group,
were meant to communicate. The second observation concerns the rich
diversity of potential meanings that we have been able to read into Hecabe’s
transformation, which spring forth from the diversity of associations that
the dog had in antiquity. What sort of interpretative methodology can we
use to approach such a situation that will enable us to appreciate, rather than
try to eliminate, this diversity, at least in some cases?

Arachne and Her Brother
Like the dog, many animals, plants, and other metamorphic end points in
Greek myths have spectrums of associations, rather than single, simple
meanings. Even the wolf, as Richard Buxton has shown, could be admired
for its community spirit, as well as feared as a ruthless predator.67 Had
Lycaon not tried to serve up his son to the gods in a stew—had he instead
run off into the woods after some other transgression, such as trying to
share the gods’ food with his mortal tablemates (as Tantalus had, according
to one story),68 we might see a very different logic behind his
metamorphosis: we’d probably say that he became a wolf because, like a
wolf, he shared his food, but shared it with the wrong friends.
As a more extended case for investigation and for the testing of a
particular methodological approach, I want to look at a myth that is
unfamiliar to most people—one whose apparent meanings have not already
been well burnished by scholars. It is a myth about Arachne that runs quite
differently from the famous story that we inherit from Ovid. It comes to us
from a scholiast to Nicander’s Theriaca, a poetic treatise on dangerous
animals:69
Theophilus, of the School of Zenodotus,70 relates that there once were

two siblings in Attica: Phalanx, the man and the woman, named
Arachne. While Phalanx learned the art of fighting in arms from
Athena, Arachne learned the art of weaving. They came to be hated by
the goddess, however, because they had sex with each other—and their
fate was to be changed into creeping creatures that were eaten by their
own children.
The scholiast’s reason for mentioning Theophilus’s story is Nicander’s use,

in his poem, of the word phalangia (the plural of phalangion and a cognate
of Phalanx’s name).71 Phalangia, in contrast to arachnai, are spiders with
venom strong enough to kill humans. What we have here, then, at least on
the most obvious level, is an aitiology for two different families of spiders.
Transformation into a spider has a connection with the earlier pursuits of
the sister: she was, after all, a weaver who had learned her skills from the
very goddess of weaving herself. But what about Phalanx? Why did it
“make sense” for him to turn into a spider? And what about the story’s

other sensational elements, the siblings’ incest and the fact that they were
doomed to be eaten by their own children? Do these plot twists somehow
play into the metamorphic logic, or are they simply a spine-chilling
transgression and its equally spine-chilling punishment?
To figure this out, we need to take a more thorough look at ancient ideas
about spiders, but we also need a methodological tool that will help us
make sense of the material we examine. For the tool, I propose that we use
the concept of affordances. The concept was invented in 1979 by the
perceptual psychologist J. J. Gibson to designate a characteristic feature of
an object, to which an individual (human or animal) can react in various
ways, depending on the individual’s own perceptions and capabilities. A
stick, for example, may offer the affordances of straightness, length, and a
tapering tip, but it is understood as “good to dig ants out of a hole with”
only if the individual who picks it up possesses the fine motor skills
necessary for the task and the cognitive sophistication to conceive of the
possibility. A stick that is straight and long with a tapering tip might also be
perceived as a weapon or a scepter, for example. Affordances circumscribe
the potential meanings or uses of phenomena to which they are attached,
but they do not determine those meanings and uses.72
Maurizio Bettini adapted the term “affordance” to the study of cultural
phenomena and particularly to thinking about the ways in which human
observers react to animals’ characteristic appearances and habits. He
suggested, for example, that the weasel’s habit of carrying her pups in her
mouth is an affordance that gave rise to the ancient belief that weasels give
birth through their mouths and that the weasel’s slim, tubular body is an
affordance that led to her ancient reputation as a helper of women in labor
—it was hoped that, like a weasel sliding through a hole, the baby might
slip easily through the narrow space of the birth canal. Bettini emphasizes
that neither characteristic of the weasel compelled ancient thought in a
particular direction; rather, they afforded opportunities for thought that
could lead in any of a number of directions, depending on the backgrounds
of the observers—in another place or another context, the weasel’s carrying
of her pups in her mouth might be interpreted to mean that weasels ate their
young. (Of course, there isn’t necessarily any connection between the
ancient associations of an affordance and its actual function within the life
of the animal as we now perceive it from our lofty perch of twenty-first-
century zoology.)73

“Affordance” is a more useful concept than “symbol” for articulating the
ways in which myths and other cultural products—rituals and art, for
instance—accumulate and convey ideas. A symbol usually has an essential
and nearly static meaning (X symbolizes Y, or perhaps X can symbolize
both Y and Z, but X usually can’t move amongst symbolizing Y, Z, A, B, C,
D, and so on). This essentialism is, in fact, crucial to the success of many of
the symbols that we encounter: unless they can convey ideas clearly to a
fairly wide range of observers, they fail in their task. Had the symbol of the
lily not become associated almost exclusively with purity in Christian
thought, for example, it could not have represented Mary across so many
centuries and such a broad geographic span as it has.74 An affordance, in
contrast, because its meaning arises from interaction between the
observer(s) and the thing in question, allows the development of spectrums
of associations. Some stories draw on some of these, and others draw upon
others. A particular spectrum might even include associations that seem to
clash with each other—for instance, the dog can be associated with loyalty
and shamelessness, savagery and maternal love. Some or all of these
associations might be pondered by someone who watches a bitch protecting
her pups and used by a skillful narrator to produce a subtly complex story
about a dog.
But to get back to Arachne and Phalanx: what then, were the spider’s
most striking affordances in ancient eyes, and how can they help us to
understand the way that Theophilus’s story might have been received by
ancient ears? I’ll look at the three that are most often mentioned by ancient
authors.75
Weaving webs. The affordance most often mentioned is the spider’s
ability to spin fiber and weave it into a web. The associations of this
affordance vary quite a bit, however. It could indicate industriousness;
spiders were almost as highly esteemed, in this respect, as were ants and
bees.76 Yet spiders’ webs could also be used to signify neglect, in the sense
that their presence indicated that an object or place had been abandoned by
humans.77 Sometimes, the spider’s web was lauded as a work of delicacy,
produced by an intelligent creature,78 but at other times, it was viewed as a
repetitively symmetrical product, created by dumb animal instinct rather
than skill and art.79 The ease with which a web could be destroyed
suggested the transitory nature of artifice.80 Finally, a spider’s web could

evoke entrapment and a predatory nature—the best-known case being
Aeschylus’s description of Clytemnestra capturing Agamemnon in a web-
like net.81
Spiders and parricide. In the tale from Theophilus, Arachne and Phalanx
are said to be “changed into creeping creatures that are eaten by their own
children.” This reflects an affordance of spiders that we first hear about
from Aristotle, who says that young phalangia “when they grow to full
size, very often surround their mother and eject and kill her; and not seldom
they kill the male as well, if they can catch him.” This information is
repeated by several later authors, with Pliny adding the detail that the
murderous spiderlings subsequently eat their parents’ corpses. (There is, in
fact, one species of Mediterranean spiders, Stegodyphus lineatus [Latreille,
1817], that eats its mother—it’s possible that Aristotle, Pliny, and others
observed young Stegodyphi lineati taking their meals.)82
Little needs to be said about the significance of this affordance—
parricide can hardly be anything but negative—but it’s worth noting that in
ancient thought, parricide was often paired with incest, another
transgression against the integrity of the family, and that cannibalism was
also paired with incest, interfamilial murder or both to further mark their
gravity. The tangled histories of the House of Atreus and the House of
Laius furnish ready examples of these combinations. It is mythically
“logical,” in other words, for incestuous siblings such as Arachne and
Phalanx to turn into creatures doomed to be killed by their progeny and then
end up as the victims of cannibalism as well.
Spiders and priapism. Perhaps the most horrifying phalangion in
Nicander’s catalogue is the rhōx (also called the rhax). According to
Nicander, its bite causes the victim’s eyes to turn reddish and a shivering to
settle upon his limbs; numbness overcomes his hips and knees. So far, this
is not very different from the effects of a few other phalangia that ancient
authors describe, but a further symptom is quite striking: “[The victim’s]
skin and genitals grow taut, and his penis projects, moistened with ooze.”
Several other authorities describe the same symptom, either echoing
Nicander’s phrases or using their own words; most of them extend this
symptom to the family of phalangia as a whole.83
Working from ancient descriptions of the appearance and behavior of the
rhōx, modern zoologists have identified it as a member of the genus
Latrodectus (Walckenaer, 1805), still alive in Mediterranean countries

today.84 The bite of any member of the Latrodectus genus really does cause
priapism and involuntary ejaculation if antivenin is not administered within
a reasonable amount of time.85 Thus, these ancient reports, incredible
though they may seem, are probably based on something that ancient
observers saw—that is, on an affordance of the phalangion. Nowadays, we
know that the distressing symptoms that Nicander described are produced
by neurotoxins that the spider injects into its victim, but in antiquity, not
surprisingly, the symptoms were understood to mean that the spiders
themselves were filled with an excessive lust with which they could infect
others through their bite.86
These, then, were the three affordances of spiders that the ancient
audience were likeliest to have had in mind when they encountered the
story of Arachne and Phalanx for the first time: an ability to spin and
weave, a habit of parricide that was sometimes followed by cannibalism,
and a lustful nature. The associations of the second and third affordances
are quite limited, which might lead to us reading this story
straightforwardly as a cautionary tale in which sexual transgression within
the family leads to other horrible abnormalities, but two things should
encourage us to think further. First, the spider’s most frequently mentioned
affordance—the ability to spin and weave fibers—is still up for
interpretation. Second, Phalanx is given a characteristic of his own that he
shares neither with Arachne nor with the race of spiders: when the story
opens, he is undergoing military training. The word phalanx, which means
“battle array” as well as “poisonous spider,” epitomizes this (the double
connotation of the word probably helped to inspire the story, in fact).87
Finally, both Arachne and Phalanx are pupils of Athena, within a story set
in Athens, which suggests that what is at issue is not just the fact that
Arachne and Phalanx failed to behave like proper humans, but also the fact
that they failed to behave like proper citizens of their city, even when
nurtured by its leading divinity. How did weaving, warfare, sexual
misconduct and civic duty come together in the Athenian imagination?
Weaving frequently served as a metaphor (in Athens and elsewhere) for
two institutions that underpinned a proper society: marriage and the coming
together into civic groups of the families resulting from those marriages.88
Particularly resonant for these representations was that fact that weaving
began with fibers that could be viewed as opposing one another: some ran

vertically (the warp) and some ran horizontally (the woof). And yet the
proper combining of these fibers produced a textile that was strong, useful,
and beautiful—and so it was also with marriage, which combined the
“opposites” of male and female,89 and with civic coalitions, which
combined groups that might otherwise be at odds with one another.
The metaphors of “weaving a marriage” and “weaving a city” from
disparate fibers are familiar from their extended use in Plato’s Statesman,
Aristophanes’s Lysistrata, and other texts.90 They were important as well at
the Panathenaia, the main festival of the Athenian year, in which all
members of the city, high and low, joined together to celebrate its
accomplishments. A cluster of associated myths emphasized the festival’s
annual rearticulation of Athenian unity: Theseus was said to have founded
the Panathenaia to celebrate the unification of the previously independent
villages of Attica, for example.91 The celebration culminated in the
dedication of a new peplos, woven by girls and women from noble
Athenian families, to Athena, Protector of the City (Athena Polias), at her
main city temple on the Acropolis, in hopes of renewing her affection for
the city that carried her name.
It was not only civic unity that the Panathenaia and its peplos celebrated,
however; the preparation of the peplos brings us back to weaving as a
metaphor for marriage and its extended significance as a task that every
properly raised girl had to master before her wedding. For although the bulk
of the work of weaving Athena’s new peplos was done by older females,
the ritually important inception of the project also involved young girls who
were serving in another cult dedicated to Athena.92 They were called
Arrhephoroi, and their most important duty, performed during the night of a
festival called the Arrhephoria, was to receive a mysterious package from
the priestess of Athena on the Acropolis and carry it down a special
staircase to a temple of Aphrodite. The priestess of Aphrodite gave them
another mysterious package to carry back to the priestess of Athena. The
myth associated with this journey told of how Aglaurus, Pandrosus, and
Herse, daughters of Athens’s first king, Cecrops, had once been charged by
Athena with guarding a basket into which they were not allowed to look.
They looked anyway and caught sight of Athena’s foster child,
Erichthonius, who was part snake and part human—a natural-born hybrid.

Maddened with fright at the sight, the girls jumped off the Acropolis to their
deaths.93
Walter Burkert has noted that whereas the myth narrates a premature,
improper introduction to motherhood and its concomitant sexuality, the
ritual enacts a proper introduction, during which girls leave the realm of
Athena, the virgin goddess, briefly visit the realm of Aphrodite, the goddess
of sexuality, and then travel back again to Athena, whose duties also
included receiving, at her temple on the Acropolis, each and every Athenian
bride on the eve of her marriage.94 All the while, these girls resist the
temptation that the daughters of Cecrops did not: they do not peer into the
mysterious boxes they have been told to protect. Burkert has also stressed
the coherence of the two tasks with which the Arrhephoroi were charged: a
girl’s preparation for marriage properly comprised both an introduction to
sexuality, in preparation for her role as a mother, and the mastery of
spinning and weaving, a good wife’s tasks par excellence. Other myths
bring the two fields of activity together by associating the daughters of
Cecrops not only with their flawed introduction to sexuality but also, and
more successfully, with weaving and the care of textiles (traditional tasks
that every well-brought-up virgin learned before marriage). Aglaurus and
Pandrosus were said to have been the very first wool workers, and all three
sisters were credited with weaving the first clothing for the people of
Athens. Aglaurus also established the Plynteria and Kallynteria (festivals at
which the statue of Athena and its clothing were cleansed).95
And Phalanx? If Arachne represents the failed virgin, then what does he
represent? Our best clue comes straight from Theophilus’s story itself:
Athena was teaching him hoplomacheia, “fighting in arms,” a type of war
craft in which young Athenian men were trained.96 One of the occasions on
which young men’s preparation for military service was highlighted was the
Panathenaia, during which they competed in a variety of contests designed
to display their military might, including a hoplite race—that is, a race
amongst men dressed in full armor.97 At the Panathenaia, in other words,
young men exhibited their prowess in the skills that defined maturity as an
Athenian male just as women exhibited their accomplishment in one of the
skills that most centrally defined Athenian femininity.98 Phalanx served as a
(failed) representative of all Athenian youths and their potential just as
Arachne served as a (failed) representative of all Athenian girls and their

potential—and both Phalanx and Arachne failed not because of any lack of
skill on their parts (they had been trained by Athena, after all) but because
they could not channel their sexuality properly.
Theophilus’s story of Arachne and Phalanx, then, explored some of the
same ideas as were articulated in Athens’s most important festival: weaving
and military skills were the proper pursuits of young people, on which both
strong marriages and strong cities could be built, but properly controlled
sexuality was essential as well. If a strong city is built upon the union of
diversified families and a strong family is built upon the union of
diversified spouses, then the union of siblings, by definition, weakens the
fabric of both. This leads to a final observation about the affordance of a
spiderweb. Spiders, as the ancients already observed, are isolated creatures
that share their webs only when they mate.99 Indeed, our only ancient
description of spiders sharing a web reads like a parody of marital
unification: the female sits in the middle of her web, and the male sits on
the periphery. She pulls on a strand to move him a bit closer, and then he
pulls on a strand to bring her a bit closer. They repeat this until their hind
parts finally meet, and it is in this awkward position that they clumsily
engender the offspring that will eventually kill and consume them.100 The
spider’s web, then, is very different from the cloth under which the new
bride lay with her husband, which she had woven while still a virgin, and
very different as well from the metaphorical textiles that Plato’s Statesmen-
Weavers produce: the spider’s web is a clumsy marital bed from which only
catastrophe can arise. Athena’s erstwhile wards, Arachne and Phalanx, fail
in their transitions to both sexual and civic maturity and doom themselves
to lives as lonely creatures that lack any community at all.
If we have properly unpacked Theophilus’s story of Arachne, then we
have to assume that it had a lot of meaning for the ancient Athenians who
were its original audience and also for later ancient readers, for whom the
metaphors of weaving and marriage and weaving and the city-state
continued to resonate. Why, then, did this story languish, ignored by
everyone except the scholiast to Nicander? Why doesn’t it show up in Ovid
or anywhere else? Certainly, it was lurid enough to be attractive: what more
could an audience want than incest, cannibalism, parricide, and
metamorphosis?
A first, easy answer is that Ovid chose to narrate another story about
Arachne—a story that he may have picked up from a Lydian tradition about

the spider or may have made up himself. His enormous success as a
narrator guaranteed that any other version would drift into the shadows. It is
Ovid’s Arachne whom Dante met in Purgatory, amongst the proud and
Ovid’s Arachne whom Velázquez and Rubens painted.101 But this only
pushes the question back a step: why didn’t Ovid choose to narrate the
Athenian story?
Programmatically and thematically, the story set in Lydia serves Ovid’s
purposes very well. Arachne gets in trouble for her skills as a weaver of
textiles, an undertaking that already in antiquity was used metaphorically to
represent the construction of texts—of stories. A storyteller is exactly what
Ovid presents himself to be in the Metamorphoses: a meta-commentary on
his own art seems to be embedded in his narration of the contest between
Arachne and Minerva, then. The stories that Arachne and Minerva weave
into their tapestries, moreover, gave Ovid ample opportunity to narrate
numerous further myths of transformation. If myths are mercenaries, then
Ovid’s band of warriors beat out any army that may still have been fighting
on behalf of Theophilus.
The Power of the Gods

So far, we have seen that Greek myths sometimes seem to have clear
messages or at least make clear connections: Apollo’s ardor causes Daphne
to be turned into a daphnē tree; therefore, the daphnē becomes sacred to
Apollo. Lycaon is willing to serve up human flesh as stew meat; therefore,
he becomes a wolf—and therefore, dear listeners, be sure you don’t serve
human flesh. At other times, messages and connections seem to get lost in
transmission as a great story is passed from narrator to narrator, each of
whom crafts it to suit his or her own desires and purposes: Aedon becomes
a tunefully lamenting aēdōn, and her sister Chelidon becomes a twittering
chelidōn; but then the Athenians push Aedon aside in favor of someone
named Procne, and her sister takes the new name of Philomela—the tuneful
one—but has her tongue severed. Then suddenly, it’s the sister with a
severed tongue who turns into the nightingale—a nonsensical
transformation, zoologically speaking, yet in the meantime, the luridly
attractive story has managed to do a lot of important work on behalf of its
many narrators. Adopting the concept of affordances helps us to see that
myths can send more complex messages than we have generally credited
them with and to better understand a particular myth against the background

of other myths, rituals, and social institutions with which it is in dialogue.
Perhaps affordances would help us better understand what was going on
with Hecabe, as well.
If anything is completely clear so far, it’s that there is no standard key or
even set of keys that’s guaranteed to unlock the meanings of all
metamorphic myths. The one unassailable observation that we circle back
to over and over again, instead, is that whatever other aitiological,
ideological or cautionary work was done by stories of transformation, one
of their primary obligations was to be great stories—ripping good yarns
about extraordinary events, about “wonders” (thaumata), as Richard
Buxton has reminded us.102 Indeed, as if to make sure we don’t miss that
point, characters in the narrations sometimes express wonder—or horror—
at the events themselves. The most extended instances of this, before Ovid,
come from Aeschylus’s two descriptions of Io. In his Suppliant Maidens,
the chorus describes the Egyptians who received her as “trembling at her
strangeness, with pale fear at their hearts,” as they beheld “a mixed-breed
creature, half-cow, half-human—a monster to be marveled at.” The verb
“marveled at”—ethamboun—is a cognate of thaumata. The chorus of his
Prometheus Bound declares that they have never seen something “so
offensive to their eyes, so shameful and frightening, so chilling to the
soul”—and this in spite of the fact that the chorus members themselves are
ocean nymphs.103
With this point in mind, we should step back and ponder the sheer
number of stories of metamorphosis that we inherit from Greek myths and,
at the same time, the particularity of each. Already from Hesiod we hear
about Actaeon, Battus, Lycaon, Callisto, Ceyx, and Alcyone, Atalanta, Io,
and Hyacinthus—if we had more than fragments of his Catalogue of
Women, we would undoubtedly hear about others. Our extant tragedies
narrate or allude to twenty-two metamorphoses; titles of lost tragedies
promise quite a few more. Other genres are less lush in their offerings, but
nonetheless, we glean some examples from lyric,104 and the local historians
and early prose mythographers supply quite a few.105 Then there is the huge
flowering of metamorphic narrations during the Hellenistic period, when
poets such as Nicander and Boeus assembled them from earlier sources,
preserving for us some stories that we would not otherwise have. Finally, let
us not forget visual representations: on vases, we find Actaeon as a human

with stag’s horns (even in the Underworld!), Io as a cow with the face of a
maiden, Niobe turning into stone from the feet up, pirates turning into
dolphins, and so on (see Figures 6.2–6.4).106
6.2 Actaeon turns into a stag while his dogs devour him. Artemis and Zeus look on;
Lysa (Rabid Madness), who has a dog’s head emerging from her own head, urges
Actaeon’s dogs onward. Above Actaeon’s head is the word “Euaion,” the name of a
famous tragic actor. Attic red-figure bell krater by the Lykaon Painter, dated to about
440 BCE (detail). Now in the Boston Museum of Fine Arts, inv. num. 00.346. Henry
Lillie Pierce Fund. Photograph © 2018 The Museum of Fine Arts, Boston.

6.3 Hermes prepares to kill Argus, who is guarding Io, here shown with the body of a
cow but the face of a woman, to indicate her transformation. South Italian red-figure
oinochoe by the Pisticci Painter, dated to between 445 and 430 BCE (detail). Now in the
Boston Museum of Fine Arts, inv. num. 00.366. Henry Lillie Pierce Fund. Photograph
© 2018 The Museum of Fine Arts, Boston.

6.4 Niobe, standing in a funeral naiskos, turning into stone from the feet upwards.
Apulian red-figure loutrophoros by the Painter of Louvre MNB 1148, dated to third
quarter of the fourth century (detail). Now in the J. Paul Getty Museum, inv. num.
82.AE.16. Digital image courtesy of the Getty’s Open Content Program.
Amongst all of these materials, remarkably, seldom is there much room

for confusion. Usually, each name is linked to only one story, or cluster of
similar stories, and there is generally only one backstory, or cluster of
similar backstories, for each animal, plant or mineral that started out as a
human. It’s unlikely that an ancient listener, hearing the name Io, would
think of any metamorphosis other than that of a woman into a cow or,
hearing about a hunter who was turned into a stag and then torn apart by his
own dogs, would think of anyone other than Actaeon. Of course, there were

some local stories that complicate this picture—the stories of Aedon and
Chelidon seem to have coexisted alongside those of Procne and Philomela,
each leading to transformations into a nightingale and a swallow. There
were two different Scyllas—one turned into the monster with dogs below
her waist and the other into a seabird.107 In the overwhelming majority of
cases, however, it must have been very clear which metamorphosis went
with which name. This specificity is one of the reasons that Greek
metamorphic myths survived through the centuries as allegories, metaphors,
and significant allusions: each is so vividly distinct from the others that it
can reliably convey a whole set of images and ideas through little more than
a name and perhaps a telling detail. Vergil, in listing creatures that thwart
the beekeeper’s work, can refer to “the spider, cursed by Minerva,” and
expect us to know at least one of the stories of Arachne; Francis Bacon can
describe traitorous servants as plotting to treat their master like Actaeon and
expect us to know that the servants will tear that master apart, literally or
figuratively.108 It is one of the lures of modern works such as John Updike’s
The Centaur. Once we have realized who Iris Osgood really is, with her
bovine eyes and milky arms, we are eager to see how her transformation
into a cow will play out within the confines of the 1940s rural high school
in which Updike’s novel is set.109
Many metamorphosed characters, moreover, are tied into one of the great
ancestral families of Greek myth: amongst other things, Actaeon’s story is a
story about the royal family of Thebes, and Io’s story is about the royal
family of Argos (and how one branch of it ended up in Egypt). This gives
the characters, and their stories, a concrete link to the lives of their
audiences: these are the great-grandfathers and great-grandmothers from
whom their city (or their neighbors’ city) sprang and to whom noble
families still trace their lineage; these are the heroes and heroines who are
worshipped at local shrines. So persistent is this idea that even a royal
maiden transmogrified into a bear (Callisto) may nonetheless give birth to a
son (Arcas) who becomes the eponymous king of a land (Arcadia) and the
progenitor of a dynasty.110 The story of the nightingale and the swallow,
once it had been firmly tied into the Athenian line of descent by its
association with Procne and Philomela, daughters of the primordial king
Pandion, could be used by Demosthenes to exemplify the patriotic heroism
of Athenian women: the sisters’ infanticide was their own way of rebelling

against a northern barbarian.111 Nicander, in his Metamorphoses, carefully
ties each story to a locale: it was in Calydon that the Meleagrides turned
into guinea hens, in Orchomenus that the daughters of Minyas became bats,
on the farthest border of Thessaly that Aegypius and Neophron became
vultures. Sometimes one of his stories has aitiological implications for a
local cult: Leucippus turned from a girl into a boy on Crete, giving rise to a
new festival there in honor of Leto, who had worked the trick.112 In part,
Nicander’s focus on places bespeaks the same learnéd love of obscure local
histories that we see in Callimachus and other Hellenistic poets, but the
information had to be there for these poets to use: they inherited older
stories that had shaped local self-perceptions.
We are a long way, in other words, from the nameless swan brothers and
their sister whom I mentioned in the first section of this chapter and the
anonymous women who married stallions; Greek myths are most often
about people who are entrenched in the larger history of a particular place.
They are not “once upon a time, in a kingdom far, far way.” In fact, Greek
myths more generally seem determined to tell stories about specific
individuals who lived in specific places. This does not mean that the Greeks
did not tell what we would call fairy tales, in which the characters are
usually given generic names. It would be remarkable if they did not, given
that fairy tales are found in almost every culture. What it means, rather, is
that myths and fairy tales serve different purposes, even if they draw on a
shared pool of motifs such as the wicked stepmother and a pool of plot
themes such as the little man overcoming a giant or ogre through
cleverness. It is always horrifying to hear about a man being turned into a
predatory beast, but a man who is one of the ancestral kings of Arcadia
turning into a predatory beast (Lycaon) is horrifying in a more resonant
way.
It is also the particularity of these stories that keeps each of the marvels
marvelous—that allows each to seize the imagination without sating the
palate. Indeed, the lack of reliable “rules” or “logic” that govern
metamorphoses enhances that particularity; the wonders offered by these
stories cannot be circumscribed within any grammar of correspondences
and therefore cannot become rote and predictable. They remain wondrous.
But it is the sheer number of metamorphoses, cumulatively, that carries
another kind of weight: as consumers of Greek myths, we are constantly
reminded of something that I said earlier in this chapter: the gods can do

almost anything they please. What we are accustomed to think of as reality
remains fluid under their touch and is at the mercy of their emotions. In this
sense, it becomes even clearer that the metamorphosis of humans into other
creatures and things is just one, although an especially spectacular, type of a
larger category of wondrous events that we hear about both in myths and in
other types of narrative. Gods can enhance the beauty of an individual’s true
form, as Athena does for the wave-swept Odysseus before he meets
Nausicaa on the Phaeacian shore.113 They can cover great distances in the
blink of an eye, as they do when darting on and off the Trojan battlefield, or
send a winged golden ram to rescue their children from the sacrificial altar,
as Nephele does in our earliest stories of the Golden Fleece.114 They can
work miraculous cures, as we hear from the Epidaurian iamata, even
performing surgery while the patient is asleep.115 They can fight alongside
their worshippers against the barbarian Gauls, as Apollo and two “white
maidens” (Artemis and Athena?) did in Delphi in 279 BCE, and as Demeter
and Persephone did in a naval battle.116 And of course, they can change their
own forms as easily as we change clothing—not only in myths (in which
they disguised themselves as humans as well as animals) but also in real
life: St. Elmo’s fire was understood as an epiphany of the Dioscuri;
Asclepius appeared to the son of his worshipper Isyllus as a man gleaming
in golden armor, but sometimes appeared to other worshippers as a snake. A
priestess of Demeter named Alexandra asked Apollo, at his Didymean
oracle, why the gods had recently manifested themselves so often in the
forms of maidens, women, men, and children.117
Of course, manifestations such as those that Alexandra talks about were
usually understood to have happened more recently than the mythic ones,
perhaps in the generation of their audiences’ parents or grandparents,
perhaps even in the audiences’ own. These wonders had the advantage,
therefore, of temporal proximity and the credibility that comes with it, as
well as the credibility that comes from a reasonable degree of experiential
verisimilitude: “real-life” sources do not mention humans turning into
plants and animals, but they do mention gods taking on human form (and
occasionally snake form) to appear to their worshippers. Who is to say that
the gods aren’t lurking amongst right us now, disguised as sausage sellers,
nursemaids, or the boa constrictor that’s for sale in the local pet store? But
the wonders described by myths had advantages as well. Most importantly,

they included—indeed, they fervently embraced—the negative side of the
gods’ power and how it affected humans. Many mortals were
transmogrified because of a god’s anger or lust, and many of those who
were transmogrified because of divine pity landed in their pitiable
circumstances because of divine anger or lust—perhaps even the anger or
lust of the very same god who later took pity. Myths, then, were a place
where the Greeks could talk about an aspect of the gods’ nature, and its
potential to explode at any moment into mortal life that they did not care to
confront in venues such as public inscriptions and prayers. So has it always
been: narratives that stretch the imagination, narratives about the wondrous
(whether they are strictly regarded as “fictional” or not), provide space to
entertain all sorts of issues that are hard to address elsewhere. Indeed, the
further they stretch the imagination, the safer they may seem, detached as
they are from everyday experiences. As Victoria Nelson has observed, for
example, representation of the supernatural in contemporary films
predominantly takes the form of the grotesque and demonic, rather than the
benign and angelic. Nelson attributes this to the secular, Aristotelian
worldview that has held sway in the West for the past four centuries, which
she suggests has repressed all religious thought outside of that prescribed by
doctrine (which, in Western Judaism and Christianity tends to emphasize
the benign aspects of divinity). The repressed thoughts—the negative side
of the transcendent—bubble to the surface in media that can be neatly
compartmentalized away from normal life.118
This is not exactly the case for antiquity, of course; practices and beliefs
that we might describe as “demonic” in nature, or as directed against the
demonic, were widely practiced—binding spells, amulets, monthly suppers
left at the crossroads to avert bad luck. Yet the parallel holds insofar as
almost all formal declarations about the gods in antiquity emphasized their
benignity towards mortals. As Diogenes the Cynic said, if everyone who
foundered and was lost at sea had been able to make dedications to the gods
at Samothrace, they would have outnumbered by far the existing
dedications, which had been thankfully set up by those who had survived
the rigors of the waves.119 Are those who suffer at the hands of the gods in
any position to advertise the fact that they have suffered, and if they are,
would they dare to do so? Myths were the proper places to explore the
grimmer side of the mortal-divine relationship, whether it ended in
metamorphosis or some other ill. These tales were closely tied to their

audiences through ancestry, through the local landscape and through cultic
aitiology, but they were conveniently kept at a distance through the
perceived gap of time and through the very fact that audiences knew that
fallible humans had had a hand in their creation. Who could be sure exactly
how much of a tragedy, epicinician ode or other composition came from a
poet’s own imagination and how much had been inspired by the Muses
(who in any case had boasted to Hesiod of being accomplished liars, when
they chose to be)?
Many of these narratives—performed in the theater, danced at a victory
celebration or sung from the rhapsode’s dais—were delivered during
festivals that were meant to honor the gods. The dark side of divine power
was nonetheless power, and there is no question but that some myths of
divine thaumata (metamorphic or not) were meant to remind listeners of the
damage that the gods could do when they were not happy with their
treatment at the hands of mortals. Here, we drift into the category of the
aretalogy—a narrative meant to exalt a god’s works. Typically, aretalogies
focused on the good things that a god had done, but goodness always lies in
the eyes of the beholder. We may sympathize with Dionysus’s decision to
turn the pirates who have kidnapped him into dolphins, and we may
therefore describe that part of his Homeric Hymn as an aretalogy, but we
may also sympathize with Metanira’s outcries when she sees Demeter
placing her infant son in the hearth fire, in the Homeric Hymn to Demeter.
What are we to call the famine that follows this incident, signaling
Demeter’s rage?120 That it establishes Demeter’s absolute power over the
growth of grain is indisputable, but it surely brings no benefit to any human.
Or to take a real-life example, in the third century BCE, a man on Delos set
up an inscription praising Isis and Sarapis for striking his opponents dumb
in court—the lawsuit concerned his right to establish a temple for the two
gods in the agora. This was a demonstration of the gods’ powers, indeed,
and we might understand it to have been justified by their desire for a
permanent home on the island; but then, we don’t know the other side of the
story—or to which gods the other side might have been praying.121
Transformation and Identity

Wonders, whether they happened in narratives or real life, changed the lives
of the mortals whom they touched. Those that involved metamorphoses
altered their victims’ exterior forms. But did metamorphosis alter the

essential self? It seems not: narrators could present the person as being
aware of what was happening during the change and as reflecting on his or
her new life in a new body after it had happened. Ovid is the master of this.
For his Callisto, “human feelings remained, although she was now a bear;
with constant moanings she shows her grief and stretches up such hands as
are left to her to the heavens, and, though she cannot speak, still feels the
ingratitude of Jove.” And his Actaeon “marvels to find himself so swift of
foot, and when he sees his features and his horns in a clear pool, tries to say,
‘O woe is me,’ yet no words come.”122 We find authors earlier than Ovid
exploiting the possibilities as well: as Achilles tells Niobe’s story, Niobe
“although being stone,” still “broods over the sorrows the gods gave her”;
Euripides’s Cadmus cries out, in mid-transition, that his lower half has
already become a serpent and Aeschylus’s Io remembers the horrible
moment when her “form and her wits were distorted.”123 Moreover, even as
a cow, Io is able to carry on a detailed conversation with Prometheus about
what her future holds.124 In other cases, preservation of identity is
poignantly suggested by the fact that the newly transmogrified mortals are
compelled not only to continue doing whatever it was that led to their
transformations in the first place but also to continue feeling the
accompanying emotions: Phaethon’s mourning sisters cry forever, even
after they become poplar trees;125 Alcyone and Aedon mourn for their lost
loved ones even after they become birds;126 Anthos flees in terror from
horses, even after he has become a bird; Arachne spins and weaves
forevermore.
One easy conclusion that can be drawn from this is that the Greeks
understood humans to be dual in nature: there was the outer body, which
might be changed, and then there was something else, representing the
essential person, which remained stable. The Odyssey, describing the
transformation of Odysseus’s men into pigs, makes that stable element the
mind (nous), but most narrators leave the details vague. Ovid occasionally
also specifies that it is the mind (mens),127 but then very late in his narrative,
he seems to imply that it was the soul (anima) that survived transformation
instead of, or as well as, the mind: he introduces Pythagoras, who claims
that after the deaths of our current bodies, our souls will enter into new
bodies, perhaps even those of animals or plants—metempsychosis.128 From
here, Pythagoras turns to discussing all sorts of transformations, including

some that are not very different from those that Ovid has been narrating all
along: Hyperborean men who grow feathers after bathing in a pool
belonging to Minerva, hyenas that can change their gender as needed.
Metamorphosis and metempsychosis are two ways in which a soul might
experience different forms of embodiment—the first being a premature, and
divinely imposed, version of the second.
Once having entered the Greek world with Pythagoras in the sixth
century, the idea of metempsychosis never departed, persisting as a familiar
alternative to the more traditional belief that the soul, after parting from the
body at death, would spend eternity in the Underworld—in pleasure, in pain
or simply in boredom as the corpse rotted away in the soil of the upper
world. Under either scenario, the soul—implicitly the location of the self,
eschatologically speaking—was separable from any body that it had
inhabited or would inhabit in the future. From the concept of the separable
soul there also grew tales of extraordinary men who could temporarily send
their souls out of their bodies while they were still alive, to travel the world
and gather information before returning home: Abaris, Aristeas,
Hermotimus, Apollonius of Tyana and, earliest of all, Pythagoras himself.129
From the concept of the separable soul, too, grew the Greek fear of ghosts
—the souls of those whom death had not been able to restrain inside
Hades’s walls.130 In other words, the duality of self that is explored in
stories of metamorphosis underlies other Greek ideas as well.
Yet, although peoples other than the Greeks similarly understood humans
to be dual in nature, none of them developed myths of metamorphosis of
the same type, or with the same vigor, as the Greeks did. In spite of the
deeply rooted Hindu belief in metempsychosis, for example, the Hindus
have no stories of humans changing into other creatures or things, as far as I
can discover. In many other cultures where stories of human metamorphosis
do develop, the metamorphoses are only temporary, in contrast to Greek
metamorphoses, which are almost always permanent.131 Temporary
metamorphoses can be reversed when a spell is broken, as in many fairy
tales (the swan brothers return to human form when they don the shirts their
sister has sewn; the Beast returns to human form when Beauty’s tears fall
upon his prostrate form; the frog turns back into a prince when the princess
kisses him—or, as in the earlier version of the story, when she throws him
against a wall in frustration). Or temporary metamorphoses might be
deliberately undertaken by those who have the ability to shape-shift back

and forth of their own volition: the Norse Berserker who shifts between
man and bear, the Welsh Selkie who shifts between seal and woman, the
American Indian skin-walker who shifts into whatever form he desires, the
European werewolf and his various relatives—the Chinese were-fox and
were-dog, the African were-hyena, and so on. The Greeks had notably few
stories about human shape-shifters: Periclymenus and Mestra are the only
two names that come down to us, each of whom were given their special
ability by an affectionate Poseidon, who himself was one of the
polymorphic gods of the sea.132 The were-animal enters the Greek
imagination only as something that outsiders believe in: the distant
Scythians said that their neighbors, the Neuri, turned into wolves and then
back again; the Arcadians (paradigmatic primitives of the Greek world)
claimed that boys undergoing rites of social maturation occasionally
changed into wolves for nine years and then returned to human form if they
had avoided eating human flesh.133 In Greece, in other words, shape-shifting
was a talent reserved almost exclusively for the gods—not only sea gods
such as Thetis, Nereus and Proteus, who had a habit of changing from one
form to another as rapidly as water runs between your fingers, but also
Olympian gods such as Apollo, who changed form three times within fifty
lines of his Homeric Hymn (into a dolphin, a star and a young man).134
All of this reemphasizes, by contrast, certain implications of mortal
metamorphoses as we find them in the Greek world: not only are the
changes almost always permanent (no tears, no kiss, no sisterly shirts will
reverse them), but they also occur at a god’s behest, rather than that of the
individual who undergoes them. And they are presented as dire—
descending either upon an unwilling human or upon one who is in such
desperate straits that metamorphosis seems desirable.
This is not the case in every culture. In other places where stories of
metamorphosis from human to animal are common, transformation is often
presented as something to be celebrated, or at least as something that, at the
time of the world’s awakening, was normal. Particularly amongst African
and American Indian peoples, myths tell of humans spontaneously
metamorphosing into animals because it is simply the way things are: a
woman who loved to swim became the first beaver, for example.135 A
Cherokee story tells of a man who began dwelling together with bears,
living as they lived and growing thick fur. When he tried to return to his old

life, with his human wife, he longed to be amongst bears again.136 These
same cultures also tell stories of humans marrying animals and producing
animal children, of humans and animals adopting and raising one another’s
children, and of humans, as a species, developing from animals. There is an
implicit assumption of what two scholars of Andean culture have dubbed
“interpenetrability” amongst humans and animals, made possible by a core
of sameness that underlies what are understood to be superficial
differences.137 The issue of whether the self is dual—of whether something
such a soul or a mind “stays the same” even as the external body changes—
is irrelevant, because the inner selves of humans and animals are not
significantly different, even if they are clothed in different exteriors.
Many Greek metamorphoses seem to be aimed at tidying up the world,
by moving a self that doesn’t belong in the human realm out into a realm
where it is more at home. Cannibalism, incest, infanticide and lesser
transgressions such as rape and theft—all fairly common events in
metamorphic backstories—must not be tolerated in our world, but in the
natural world, the world of animals, they can be tolerated. Indeed, they
might even be considered characteristic of animals, as the story of Arachne
and Phalanx demonstrated: parricide, lust, and cannibalism are par for the
course amongst spiders. What is bad behavior for a human, then, can be
naturalized or normalized through metamorphosis138—a positive change for
the world at large, which now has everything in its proper place, even if not
for the transformed, who find themselves forever exiled from the
communities into which they were born. This may, indeed, be the most
significant loss that the self experiences during metamorphosis: the
essential self remains Actaeon or Io, but he or she is no longer welcome
amongst (and usually no longer recognized by) those whom he or she used
to hold dear: Ovid’s Io, to take an especially poignant case, finally has to
write her name in the sand with her hoof, to make her family understand
what has happened. The new community into which the transformed mortal
must move, moreover, may be no community at all: Ovid’s Callisto hides at
the sight of other bears, forgetting what she has herself become, and flees
from wolves, forgetting that her father (Lycaon) now runs with the pack.139
Aeschylus’s Io wanders the outer wastes of the world with only a gadfly for
company. Procne, Philomela, and Tereus are doomed to chase one another
through the sky in an endless loop.

The stark divide between human and animal communities that is implied
by metamorphic myths surfaces in other Greek stories, as well—most
importantly, in the story of the great Flood. In the Akkadian, Babylonian,
Sumerian, and biblical tales of the Flood, the human in charge of building
an ark (Atrahasis, Utnapishtim, Ziusudra, Noah) takes animals on board,
thus carefully ensuring that their species will continue after the waters
recede—and so also in a Hindu story of the Flood.140 In all of these stories,
in fact, where enough text remains for us to hear the beginning of the tale, it
is by divine command that the animals are included. In the Greek story,
however (which may have been borrowed from the Near East and adapted
to Greek tastes), Deucalion and Pyrrha, having been forewarned of the
storm by Deucalion’s father, Prometheus, ride it out alone in a larnax—a
small chest usually used for storing household goods and far too cramped to
hold anyone other than Deucalion and Pyrrha themselves. Upon drifting
ashore at Mount Parnassus, the couple is told by the goddess Themis how to
re-create the human race.141 Not a single author talks about how the animals
were regenerated after the Flood until we reach Ovid, who tells us that they
spontaneously burst forth from the postdiluvian soil without any help from
either gods or humans.142 The Greeks, then, in contrast both to their
Mediterranean neighbors (from whom they inherited a number of
mythological themes) and to one of their Indo-European neighbors, did not
imagine humans and animals making common cause against the gods’
destructive rage. For Greeks who told the story of the Flood, humans and
animals were categorically different; if humans had natural collaborators
during this great disaster, they were, rather, two of the more kindly gods
(Prometheus and Themis).
Striking, too, is the lack of any Greek myth justifying the sacrifice of
animals. The tale of how Prometheus performed the first sacrifice explains
why the victim’s meat was divided up between mortals and immortals in the
way that it was, but offers no justification for the act itself.143 Stories about
specific sacrifices trace their origins to bad behavior on the part of specific
animals, thus placing the blame firmly on the victims. The aition for the
Buphonia (literally the “Murder of a Cow”) that was performed each year in
Athens, for example, said that a cow had once wandered onto the Acropolis
and eaten the grain off of Zeus’s altar. A pious bystander reacted angrily,
killing the animal. After a kangaroo court exonerated the cow’s murderer,

an oracle ordered the Athenians to repeat the act every year thereafter.144
Similarly, pigs deserved to be sacrificed to Demeter because a herd of pigs
that chanced to be in the vicinity when Persephone was dragged beneath the
earth by Hades happened to be swallowed up as well—guilt through
association.145 It is also animal guilt that explains the prohibition of pig
sacrifice in cults of Aphrodite: she developed an aversion to the entire
porcine race after a wild boar killed her lover Adonis. (The same story
could be used, however, to justify the sacrifice of pigs in a few of the
goddess’s cults—once again, we see how flexible a mercenary myth could
be).146
Quite different from these Greek stories is a Hindu myth about the origin
of sacrifice. In the beginning of time, there were five types of animals
deemed appropriate for sacrifice: the human, the horse, the cow, the ram,
and the goat.147 The gods first sacrificed humans, until their sacrificial
qualities had left them. At that point, they moved on to horses, until their
sacrificial qualities had left them, too, then to cows, and so on, until the
elusive sacrificial quality finally lodged itself in rice and barley—et voilà,
vegetarianism. Once upon a time, then, humans were not only a sacrificial
animal but the sacrificial animal par excellence. When they escaped from
this role, other animals escaped with them. No Greek would ever imagine
such a thing—the few stories they told of human sacrifice associated it with
barbarians in distant lands such as Libya, with the maniacal rage of an
Achilles, or with an angry goddess who at the last moment changed her
mind, sending a deer to fill in for the virgin she had demanded. It could
never have been the norm, even at the beginning of time.
Animals were, at best, servants: Greek versions of the worldwide topos
of the abandoned child protected by a friendly animal (Telephus, Peleus and
Neleus, Atalanta, Paris, and so on)148 end the idyll while the child is still an
infant: herdsmen or hunters discover it being nursed by a mare or doe or
whatever other animal has adopted it and take it home to be raised amongst
humans. For the Greeks, there is no Enkidu, Mowgli or Tarzan who grows
to young adulthood in intimate acquaintance with the ways of the animals,
learning their skills and forming enduring bonds.149 The few mythic figures
who obtain an animal talent do so without really interacting with the
animal. Some seers obtain their prophetic abilities when snakes lick their
ears, for example, and others have honey dropped on their lips by bees, but

the recipients are asleep when these things happen.150 The healer-seer
Polyidus observes one snake curing another and adopts the technique for his
own use, but the snake is not aware that he is instructing Polyidus: there is
none of the teacher-student relationship that normally accompanies the
learning of a skill.151 In contrast to many other cultures, there are no Greek
heroes who are part animal—the closest we come is Heracles wearing a lion
skin—and the Greeks knew of no paradisiacal time or place where humans
and animals peacefully could live side by side.152
The dire tone of Greek myths about metamorphoses into animals, and the
Greeks’ fascination with such myths, then, may express both the perception
of a greater ontological division between humans and animals than other
cultures seem to have had and a greater anxiety about crossing it. This
sharpens the formulation that I offered earlier about myths of
metamorphosis: they were a place where the Greeks could talk about an
aspect of the gods’ nature, and its potential to explode at any moment into
mortal life, that was central to their belief system but that they did not want
to confront in other venues. The entertainment value that the myths offered
heightened their usefulness as such not only because the audience could
become more deeply immersed in, and thus more deeply engaged with,
their subject matter but also because these myths were adamantly set in an
earlier time. They were about ancestors: members of one’s own group and
yet distant, people like oneself and yet not oneself—models on which one
might hope to have improved.

7
Heroes
IT WAS OVID WHO ensured that Greek stories of metamorphosis would

survive down through the centuries, but no such patron saint of stories
about heroes ever existed—nor does one seem to have been needed. From
our earliest to our latest narrations of Greek myths, heroes are omnipresent.
Perhaps this point seems facile: after all (one might say), didn’t the plots of
the Iliad and the Odyssey guarantee that heroes would be front and center at
the dawn of Greek literature? And wasn’t it simply the immense popularity
of these two works, throughout antiquity and beyond, that guaranteed the
renarration and elaboration of their protagonists’ stories?
But some thinking about the types of characters that inhabit stories and a
survey of Greek mythic narratives will show that matters are more complex
than that. I’ll start with the first. Stories can involve four types of
characters, ontologically speaking: gods and other supernatural creatures
(angels, demons, and so on), ordinary people (that is, humans who are more
or less like us), animals (either real animals or the anthropomorphic animals
of fables, who arguably might be put in the second category), and finally,
heroes (whom I will define as humans who either are born with or acquire
status and abilities beyond that of other humans, which they retain after
death and can use to benefit the living humans who worship them).
Greek myths have very little to say about the second group, ordinary
people. Once in a while, ordinary people play minor roles in a myth: local
people help to build Apollo’s Delphic Oracle in the Homeric Hymn to
Apollo and Cretan sailors then staff it; an old man in Hermes’s Homeric
Hymn tells Apollo where Hermes has hidden Apollo’s cattle; Phaedra’s
nurse in Euripides’s Hippolytus fatally intervenes in her mistress’s affairs;
the choristers in Greek tragedies, who are often identified as servants or
other ordinary sorts of men or women, stand by to give advice that is

usually ignored by the main characters—who come from the class of
heroes.1
Greek myths have very little to say about the third group, either, in
contrast to African and Native American myths, for example, in which
animals are often the protagonists. In Greek myths, cattle are raided and a
rabbit’s pregnancy can serve as an omen, but as narrative agents, animals do
almost nothing. Notable exceptions are the eagle who conveys Ganymede
to Zeus’s waiting arms and the eagle who consumes Prometheus’s liver
each day, Argus, the faithful dog who recognizes the disguised Odysseus,
and the dolphin who rescues Arion.2 Only the latter two might be said to act
of their own volition; the eagles are merely tools serving Zeus’s will. People
sometimes become animals in myths, as we saw in Chapter 6, and we
sometimes get brief, poignant glimpses into their new lives before they
disappear into the underbrush, but stories of metamorphosis are more
interested in exploring how human behavior or misadventure leads to
transformation into an animal than in the animal qua animal. Other animals
we meet in myths are abnormally large or fierce or swift, or are born
hybrids—but almost any animal who is worth noticing in a myth is in one
way or another connected with a god, a hero or both, either because it is the
child of a god (Pegasus, Cerberus, the Lernaean Hydra), because it is the
quarry of heroes (Cerberus, the Lernaean Hydra, the Nemean Lion),
because it is a hero’s steed (Pegasus, Xanthus, Balius) or, quite often,
because of some combination of these reasons, as is the case with all of the
examples I just gave. The few animals who are worthy of an individual
name or a geographic tag are well knit into the mythic network of gods and
heroes that I discussed in Chapter 4.
In other words, it was on the first and fourth groups of characters, the
gods and the heroes, that myths focused in Greece. Before we can say more
about that, however, we need to give more thought to the term “hero” itself.
Provisionally, I defined heroes as humans who either are born with or
acquire status and abilities beyond that of other humans, which they retain
after death and can use to benefit the living humans who worship them. In
Greece, this phrase covers a large and varied group of individuals. At one
end of the spectrum would be someone such as Perseus, a son of Zeus who
traveled to the end of the world, encountered strange creatures such as the
Graeae and the Stygian nymphs, and, having been lent magical tools by the
gods, managed to behead the Gorgon Medusa, the merest glimpse of whom

would have meant death, had he not looked at her only indirectly, in the
shining surface of his shield (as Athena seems to be instructing him to do
on the bell krater shown in Figure 7.1). Perseus then went on to rescue a
princess from a sea monster and to use Medusa’s head to petrify the wicked
king who was trying to force Perseus’s mother to marry him.3
7.1 Perseus gazes into a shield at the reflection of the head of Medusa, held aloft by
Athena. Hermes, to the right, is barefoot because he has lent Perseus his winged
sandals. The elaborate helmet worn by Perseus is presumably the one that Hades was
said to have lent him. Apulian red-figure bell krater by the Tarporley Painter, dated to
between 400 and 385 BCE (detail). Now in the Boston Museum of Fine Arts, inv. num.
1970.237. Gift of Robert E. Hecht, Jr. Photograph © 2018 The Museum of Fine Arts,
Boston.
Perseus’s story is a fantastic adventure tale, set in exotic climes populated

by strange creatures; portions of the stories of Jason and Odysseus share
these characteristics as well, as do episodes in Heracles’s career. Towards
the other end of the narrative spectrum would be heroes such as Thymoetes,
the eponym of the Athenian deme Thymoetidae. Thymoetes, who as a
bastard had no right to inherit his father’s kingdom, killed his brother and

seized the throne, thereby also winning for himself the dubious honor of
being the last of Theseus’s descendants to rule Athens. He later lost his life
in battle. Although Thymoetes’s life had some exciting moments, his story
lacks any of the fantastic details that made Perseus’s story so narratively
appealing. Nonetheless, Thymoetes was treated as a hero by the Athenians
—as a superhuman entity who could benefit their lives if he were properly
worshipped.4 Even further towards the end of the spectrum would be Iops, a
Spartan hero who received worship at a shrine outside the local marketplace
but about whom we know only, as Pausanias says, that he was born “in the
time of Lelex or Myles.”5 Iops’s story doesn’t seem to have much interested
even the local folks whom Pausanias interviewed. Also at the very end of
the spectrum would be Thoricus, the eponym of the Athenian deme
Thoricus, who received sacrifice twice a year according to inscriptions but
who is otherwise unknown.6 Similarly, many of the oikistai (founding
heroes) of colonies lacked stories that were deemed worthy of passing down
by the poets and historians.
Somewhere in the middle of the spectrum would be heroes such as
Oedipus. Like Perseus, he killed a terrible monster, the Sphinx (or rather, he
drove the Sphinx to suicide when he solved her riddle), but everything else
that happened to Oedipus was completely possible within the everyday
world of the audiences who heard his story, even if highly unusual.7 Ion
belongs in the middle of the spectrum as well: as the son of Apollo, his
genetics should have predisposed him to perform extraordinary deeds of
bravery, like a Perseus or Heracles, but his claims to fame ended up being
those of a wise ruler (he introduced synoecism to Athens, amongst other
things) and a productive father (he sired four sons who became eponyms of
Athenian tribes and colonized Asia).8 Menelaus belongs in the middle of the
spectrum, too. The son of a mortal man, Menelaus was a good warrior, but
certainly not the best of those who fought at Troy. He won immortality at
the end of his life, but not because he had performed any wondrous deeds
himself; rather, it was granted to him because he was the husband of Zeus’s
daughter Helen.9
All of these men, and many more, were called “heroes” by the Greeks
themselves, and modern scholars have followed suit: each of them has an
entry in the catalogue at the end of L. R. Farnell’s Greek Hero Cults and
Ideas of Immortality (1921), and those who are Athenian also have entries

in Emily Kearns’s The Heroes of Attica (1989). This forces us back onto the
broad definition of heroes that I started with, if we want to be inclusive, for
there are no traits that all of the men in this group share: not all kill
monsters, not all fight battles, not all are sons of gods, not all serve as kings,
not all found cities or establish dynasties.
This vagueness tells us something in itself: as a class of humans, heroes
were so important to the Greeks that the definitional boundaries were
allowed to remain fluid; the Greeks preferred to be able to add to the ranks,
now and then, someone who would help them if he received proper cult.
The Delphic Oracle encouraged this practice: over and over, Apollo advised
cities to establish cults to heroes whom their citizens had not previously
worshipped—heroes who the citizens had not even known were heroes
before Apollo told them so in some cases.10 Indeed, the category was so
flexible that it may seem to overflow, at one end of the spectrum, into that
of ordinary humans. After all, even ordinary humans, once they had died,
were understood to have special powers in their new guise as disembodied
souls—as “ghosts”—that they could use to affect the living, for better or for
worse. The most salient distinction between heroes and ghosts is that of
time: ghosts fade away once no one amongst the living any longer
remembers the person they had been, whereas heroes remain powerful
down through many generations. Scope of worship is also an important
distinction: ghosts are primarily the concern of the families they leave
behind, helping or harming them as they deserve, whereas heroes attract
worshippers from a larger range: a clan, a city-state or even many city-
states.11
Modern attempts to divide heroes up into categories have largely
foundered. For one thing, their basic premises have failed. It is hard to
prove, in most cases, that a given hero belongs in whichever of seven
categories Farnell placed him; few scholars would now agree with his
proposal that Linus began as a god, for instance, simply because, like the
infant Dionysus, he was torn into pieces.12 Any such attempt at
categorization, moreover, runs roughshod over the very inclusivity that I
just emphasized. Although the Greeks probably would have conceded that
some heroes, such as Heracles, were greater than others in terms of
accomplishments, renown, and perhaps also postmortem power, no Greek
author ever suggests that there was any functional or ontological difference
amongst the heroes (indeed, not even the Neoplatonists of later antiquity,

who had a penchant for precisely stratifying all the inhabitants of the
cosmos, tried to subdivide heroes into smaller groups).
The class of heroes, in other words, exhibited variety within what was
viewed as essential sameness. This inclusivity made the heroes, as a group,
very useful to think with. Many heroes (such as Thymoetes) had lived lives
that differed little from those of at least some men of the historical period,
which implied that the difference between these heroes, while they were
still alive, and the people who later paid them cult was no greater than that
between, say, a king and a commoner. Those in the next tier up (the heroes
who had fought in the Trojan War and their sons, for example) differed a
little more from the average worshipper—they lived in a world where gods,
and occasionally monsters, might still be encountered face-to-face—but
these were chronologically amongst the youngest of the heroes and thus
were closest in generational age to the audiences who listened to their
stories, which helped to close the gap. Many of those heroes, moreover,
shared adventures, ancestry or networks with older heroes and, by doing so,
attached the heroes as a group to their audiences, implying that the distance
between the ordinary man in the street and even Heracles himself was not
impossibly vast. The Attic hero Antiochus, for example, who became one
of the ten heroes after whom Cleisthenes named the tribal units of Athens in
the late sixth century BCE, seems to have done nothing notable, himself,
but Demosthenes and several later authors mention that he had been sired
by Heracles upon an Athenian girl and that his own sons had spearheaded
the return of the Heracleidae.13 Antiochus ran in the right circles, in other
words, and shone in the reflected light of his relatives. Perseus was the
great-great-grandfather of Heracles, who in turn was the great-great-
grandfather of Aristodemus, the ancestor of Spartan kings of the historical
period.14 Heracles was also the forefather of any number of other peoples
with whose primordial princesses he had slept. Cumulatively, this suggested
a continuity of descent that ran from Perseus, a remarkable slayer of
monsters, all the way down through lesser heroes and finally to the man on
the Spartan street. The father of both Perseus and Heracles was Zeus, which
could be taken to hint, if one chose to read it that way, that ordinary
Spartans weren’t utterly different from the very gods themselves. Hesiod’s
myth of the generations intimates this, too, even as it insists on a grim
evaluation of Hesiod’s own fellow humans: by inserting the age of the
heroes immediately before the Iron Age in which he himself lives, Hesiod

implies that, although contemporary people are dreadfully immoral and
overburdened with work, they are descendants of the gods.
In this chapter, I will focus on heroes who fall within that segment of the
spectrum that runs between Perseus and Menelaus—that is to say, the
heroes whose stories were fantastic enough, or in some other way striking
and important enough, to become well-known both within and outside of
the immediate geographic area where they were set (which typically was
also the center of the hero’s worship in cult). This is because I am writing a
book on myths; heroes who had no myths, or whose myths are scarcely
known to us, provide no grist for my mill, even if inscriptions or remarks by
ancient authors assure us that they were paid cult. We don’t want to forget
these other heroes as we make our way through this chapter, however,
precisely for the reason I have just sketched: as part of a larger group, they
helped to bridge the divide between heroes and ordinary humans and thus
helped to lend credibility to many of the other ideas that heroes’ stories
cumulatively conveyed. I will also focus mostly on male heroes, both
because we have more stories about them than about females of that class
and also because the nature of the female’s stories is different enough to
require separate treatment, which other scholars have provided.15
Gods versus Heroes

Greek myths are not populated by heroes alone, however. There are also
gods. Which group takes pride of place? To answer this, I’ll survey our
mythic narratives chronologically, working through the main genres of
literature from the Archaic and Classical periods, starting with epics.
In addition to the Iliad, the Odyssey and Hesiod’s Theogony, we have the
fragmentary remains or at least the titles and descriptions of quite a few
other epics from the Archaic period: the Little Iliad, the Returns, and
various other poems about the Trojan War and its aftermath; the Thebaid
and the Epigonoi, both of which are about the heroes who fought around
Thebes; and assorted poems about Heracles, Theseus, and Oedipus. We also
have poems about the building and voyage of the Argo, a poem about the
early rulers of Corinth, and some other poems about humans of an earlier
age such as Phoroneus, the first inhabitant of Argos, who sprang from its
very soil.
Although gods play significant roles in these stories, the focus in almost
all of them is squarely on remarkable humans—on heroes. In contrast, for

poems that focus squarely on the gods we have, in addition to Hesiod’s
Theogony, just the fragments of a few theogonies and narratives about the
gods’ early days that are attributed to Orpheus. Ancient authors tell us about
some other poems that are now lost—the Titanomacheia and Epimenides’s
theogony, for instance—but not very many.16
Heroes continue to hold center stage in the myths narrated by poems that
we collect under the terms “melic” or “lyric.” Stesichorus, for example,
focuses closely on heroes—most notably in his Geryoneis, which tells of
the battle between Heracles and Geryon, and in his own version of the
Oresteia.17 Simonides gives us a song that was sung to the infant Perseus by
his mother as they were swept over the waves in the chest that imprisoned
them, and he also narrates the stories of Jason, Theseus, and other heroes.18
Bacchylides’s dithyrambs, when enough of them remains for us to identify
their subjects, are always about heroes—Helen and Menelaus, Heracles,
Theseus (twice), Io, Idas, Cassandra, and perhaps Meleager, Pasiphae,
Achilles, and Orpheus.19 Although Pindar was credited with returning the
dithyramb to its Dionysiac roots, his contributions to the genre also
concentrate on heroes, when we can identify a subject.20 The poets also told
stories of lesser-known, local heroes: Myrtis narrated the tale of Eunostus of
Tanagra, and Corinna narrated that of Ogygus of Thebes, for instance.21
Two of the three remaining genres of Archaic and Classical poetry—
epinician and tragedy—focus on heroes, as well (the third genre, comedy,
uses both heroes and gods as characters, but outside of Aristophanes’s
Frogs, which stars Dionysus, most of its protagonists are ordinary people).22
Of the tragedies that remain to us, only three might be said to take the
problems of the gods as their central topics: the Prometheus Bound, the
Eumenides and the Bacchae (and even these spend a great deal of time on
human dilemmas). Twenty-eight of the twenty-nine other surviving
tragedies focus on heroes (the remaining one, the Persians, focuses on an
event in what was then recent history). These proportions seem to reflect
the norm—of the more than 200 other tragedies that we know about
through fragments or titles, there are only six that may have focused on
gods.23 I don’t mean to minimize the important roles that gods play in
tragedies, but I do want to stress the overwhelming interest that tragedies
showed in humans of the heroic age.

The third remaining genre, epinician, was naturally drawn to stories of
the heroes because their deeds, which typically required great strength,
nimbleness, and endurance, mirrored the athletic feats performed by the
victors whom epinicians celebrated. Heroes, moreover—like the athletes,
but in contrast to gods—were human. Two of the three epinician poets
whose works remain to us, Bacchylides and Simonides, follow this trend
closely. Three of Bacchylides’s five longer epinicians tell stories about
heroes (Heracles and Meleager, the daughters of Proetus, the Trojan War).
His story of Croesus, although historically based, is also told in the manner
of a myth and gives Croesus the aura of a hero. The fifth epinician long
enough to include a story is about the Telchines and their daughters, who
are arguably gods.24 Of other gods who play significant roles in these
epinicians, we glimpse only Apollo, who plays an important part in the
story of Croesus. Our fragments of Simonides’s epinicians are frustratingly
few and small, but he seems to have treated the stories of Castor and
Polydeuces and of Heracles, at least.25
With the third epinician poet, Pindar, things get more complicated. Eight
of his forty-six surviving epinicians take the affairs of the gods as their
main narrative theme—Pythian 9, for example, includes a long dialogue
between Chiron and Apollo concerning the latter’s lust for the nymph
Cyrene, and Olympian 7 is about the apportionment of the earth’s lands
amongst the gods and the birth of Athena.26 In many other cases, however,
stories about the gods are juxtaposed with stories about heroes to which
they have little or no narrative connection. Pythian 8 squeezes the battle
between the Olympians and the Titans in amongst stories of Amphiaraus
and the descendants of Aeacus. Nemean 1 opens with the story of Zeus
giving the island of Sicily to Persephone as a bridal gift before it moves on
to describe the childhood of Heracles. Some of what we now consider well-
known stories about the gods come down to us first through Pindar’s
glancing treatments (four lines of Olympian 7 tell us about Hephaestus
splitting Zeus’s head open with an axe to enable the birth of Athena, for
example).27 Suffice it here to say that, as a poet who struck some listeners as
enigmatic already in antiquity, Pindar did things in his own way.
Many of the poets whom I have just mentioned are also credited with
composing hymns and paeans, which are typically (although not
exclusively) directed to the gods, but these poems usually concentrated on

invoking, praising and entreating the gods, rather than on narrating their
exploits—the hymns of Sappho and Alcaeus being characteristic in this
respect and the seven longer Homeric Hymns28 being the major
exceptions.29 Hymns and paeans of either type, moreover, are notably
underrepresented in surviving literature for reasons that are hard to fathom.
It’s tempting to blame Christianity. We know that the Alexandrian scholars
included two books of Pindar’s paeans and hymns in their editions of his
work but that centuries later, when Christian monks produced manuscript
copies of Pindar, they included the epinicians but left out the hymns and
paeans. We might guess that the monks could justify preserving stories
about pagan heroes more easily than praise of pagan gods.
But Christianity can’t take all the blame for the paucity of hymns and
paeans. Had there been a lot of noteworthy hymns and paeans by authors
other than Pindar, we would expect a few of them to survive and also expect
our ancient sources to mention more of them—yet this is not the case. This
brings us to a second tempting assumption. Perhaps a lot of hymns and
paeans are lost to us because they simply were never recorded at all? We
might hypothesize that in some cases, this happened because the cults for
which the hymns and paeans had been composed considered them to be
secret. This would seem to be confirmed by the fact that we don’t get any
hymns claiming to originate in a mystery cult until we reach the Orphic
Hymns of late antiquity. And yet logic dictates that this can’t be the whole
explanation: we scarcely have any hymns or paeans from Athens, a city that
was otherwise obsessive about preserving its literature.
It’s hard to avoid the conclusion that the reason few formal, polished
narratives focusing on the gods survive is that there simply weren’t many.
Stories about the gods must have circulated in other, informal ways. We
know that nursemaids and mothers told stories about the gods to children.30
We also know that local historians and, in Athens at least, exegetes
associated with cults recorded stories about the gods.31 We hear about
women telling each other the story of Demeter and Persephone at the time
of the annual Thesmophoria; other festivals to the gods must have prompted
the exchange of tales, too.32 And finally, many stories about the gods were
embedded within compositions that focused on heroes. I’ve already noted
that Pindar’s epinicians include quite a few of these, but there are plenty of
other examples. In Iliad 6, the hero Diomedes narrates the story about how

Dionysus punished the human king Lycurgus, just as he is about to engage
the hero Glaucus in combat. In Iliad 24, the hero Achilles narrates the deeds
of Artemis and Apollo to King Priam. The chorus of Aeschylus’s
Agamemnon gives us a version of the divine succession myth. The chorus
of Euripides’s Helen tells the story of how a goddess called “The Mother”
searched for her missing daughter. In Euripides’s Ion, a conversation
between Creusa and an old man provides the story of how Athena acquired
her aegis while fighting in the Gigantomachy.33 All of these instances, and
many more like them, not only narrate divine stories but also—if we can
assume that literary art is imitating life—confirm that informal, ad hoc
narrations were a significant mode through which tales of the gods were
transmitted. The visual arts narrated stories about the gods (as well as the
heroes), too.
Greek Heroes and Their Neighbors

Nevertheless, when it came to formal narration of myths, myths that
focused on heroes were far more popular in Greece than myths that focused
on the gods—a rough estimate gives a three-to-one ratio. This may not
strike us as particularly odd. After all, we might reason, people would have
liked hearing about heroes more than gods because they could identify more
closely with heroes. Moreover, an emphasis on heroes seems normal to
most of us because, in the Western world, we have long grown up with
Greek myths as our “default” myths: whatever characteristics they happen
to possess have inevitably, implicitly, come to define what a myth is for us.
Stories from the Old and New Testaments, which similarly permeate
Western cultures, privilege exceptional humans as well. Jehovah
commands, judges, and smites, but it’s characters such as Eve, Abraham,
Moses, David, Salome, and Jesus who stick in our minds. Some scholars
have suggested that some of the Testaments’ portrayals of notable humans,
as we now have them, were influenced by Greek myths, in fact.34 Perhaps
this borrowing was motivated by the biblical authors’ perception that their
religion—which fought to establish itself within unwelcoming territory—
needed the additional appeal that thrilling human narratives could provide.
Be that as it may, Greek myths and the two Testaments are actually the
odd ones out. The narratives that we inherit from other ancient
Mediterranean cultures are markedly more interested in gods than they are
in heroes. From Mesopotamia, we have two versions of a theogony called

the Enuma Elish; two versions of the Anzu (another epic about the gods’
battle for cosmic supremacy);35 yet another theogonic poem centered on the
god Ninurta called the Lugul-e and its companion poem the Angim, which
praises Ninurta;36 the Theogony of Dunnu;37 the Descent of Ishtar (or in its
Sumerian version, the Descent of Inanna);38 Inanna and Enki;39 Enki and
Ninhursag;40 Enki and Ninmah;41 Nergal and Ereshkigal;42 and Erra and
Ishum43—all of which concern the gods.
From Ugarit, we get the lengthy Baal Cycle and a number of shorter
stories about the gods, such the tale of El’s drunken feast and the hunt of the
goddess Astarte. From Anatolia, we get the equally lengthy Hittite
compositions known as the Kumarbi Cycle, three versions of The
Disappearance of Telipinu, two versions of the Story of Illuyanka, the Song
of Hedammu, the Song of Ullikumi, and numerous shorter divine tales such
as Telipinu and the Daughter of the Sea God, the stories of the goddesses
Inara and Kamrusepa, The Disappearance of Hannahanna, and others.44
Egyptian myths, as has been often observed, are almost entirely about the
gods.
The best-known Near Eastern hero story (both now and in antiquity) is
the Mesopotamian Epic of Gilgamesh, which is preserved in several
versions and languages. Gilgamesh is described, like many Greek heroes, as
part human and part god, and like many Greek heroes, he defeats
monsters.45 His friend Enkidu, who accompanies him on his adventures,
should probably also be described as a hero, given that he helps Gilgamesh
defeat monsters. We also have two short stories about the adventures of
Lugalbanda, the second king of Uruk and father of Gilgamesh. These
constitute the latter half of a four-part tale of conflicts between the kings of
Uruk and Arata that includes interaction between these kings and Inanna,
Shamash, and other gods. In one of these stories, Lugalbanda gains the
power of flying through the air at extraordinary speeds in return for his
kindness to the chick of a lion-headed bird. Like his son, Gilgamesh, then,
he has abilities beyond those of the ordinary human and therefore is similar
to the hero as we know him in Greece.46 In addition, there are three versions
of a Mesopotamian Flood story with a hero who goes by different names,
depending on the language in which the story is told,47 and there is a story
about Etana, a postdiluvian king who flies to heaven on the back of an
eagle, seeking the means of siring a child.48 There is a Sumerian story about

the relationship between Inanna (called Ishtar in Akkadian) and Dumuzi
(Tammuz), the mortal lover whom Inanna uses to ransom herself out of the
Underworld. Later, Dumuzi is granted the privilege of returning to life
periodically—in some versions thanks to the intervention of his faithful
sister, Geshtinanna.49
Finally, the res gestae of two historical Mesopotamian kings began to
move them onto the spectrum of heroes, insofar as the kings interacted
personally with gods and also because these kings’ stories attracted motifs
that we find associated with heroes both in Greece and in other cultures.
The first figure is Sargon, who reigned from 2340 to 2284 BCE. According
to an Assyrian tale of the seventh century BCE, the infant Sargon, like
Moses, was hidden in a basket and cast into a river, whence he was rescued
and raised by a humble man to be a gardener. While still a gardener, he had
an amorous encounter with Ishtar, which enabled him to become king.50 In a
Sumerian story, he cures the king of Kish of a urinary disease with advice
that he receives from Inanna in a dream (echoing the biblical tale of Joseph
and Pharaoh). This story also includes an adventure similar to
Bellerophon’s, in which Sargon unknowingly delivers a letter that instructs
its recipient to kill him; again, Inanna saves him.51 In other parts of his
biography, Sargon is portrayed as a strong warrior who leads campaigns
against neighboring kingdoms and into distant lands. The god Marduk is
said to have eventually grown angry with Sargon’s excessive reach and
plunged his kingdom into famine.52 The second figure is Naram-Sin, the
grandson of Sargon, who reigned from 2254 to 2218 BCE. Naram-Sin is
said to have joined forces with Erra, the god of war, to fight enemies backed
by the god Enlil. Eventually, Naram-Sin went too far by plundering Ekur,
Enlil’s temple in Nippur. Enlil retaliated by plunging all of Mesopotamia
into plague, famine, and death until a delegation of other gods intervened,
decreeing that the city of Akkad alone should be destroyed but that the rest
of the country would be preserved.53
From Ugarit we get two hero tales. In one, King Danel prays to the gods
for a son. They duly give him Aqhat, who loses his life in a fight with the
goddess Anat. In the other Ugaritic tale, King Keret similarly asks the gods
for help in siring children and eventually finds himself pitted against the
angry goddess Athirat, whom he had failed to thank after his sons were
born.54

From Anatolia we have six Hittite stories, or fragments from stories, in
which mortals who might qualify as heroes play prominent roles. One
concerns a queen of Kanesh who bears thirty sons, abandons them to be
reared by the gods and then bears thirty daughters. Another is about Appu, a
man who seeks help from the Sun god in siring children. A third story,
about the Sun god, a cow, and a fisherman, may be a continuation of Appu’s
tale. The fourth is about a hunter named Kessi, who neglects to honor the
gods of the forest after he marries a beautiful woman; the fragments suggest
that their anger manifests itself, at least initially, in bad dreams.55 The Song
of Release tells of the destruction of the town of Ebla by warriors overseen
by the Storm god because the people of Ebla have refused to release
hostages from the town of Ikinkalis. Although most of the interaction in our
fragments of this poem takes place amongst the gods, there are also some
encounters between gods and prominent humans.56 Finally, in one version
of the Hittite Song of Illuyanka,57 the goddess Inara, wishing to help her
brother the Storm god Teshub defeat the dragon Illuyanka, seduces a human
named Hupasiya, and then, having gotten Illuyanka drunk, commands
Hupasiya to tie him up, to make him an easy target for the Storm god, who
finishes him off. Inara weds Hupasiya, but the marriage between goddess
and mortal is predictably short-lived.58
From Middle Kingdom Egypt (2065–1650 BCE) we have two tales that
share some traits with hero myths. One is the “Tale of Sinuhe,” which tells
of a courtier who flees from the prince he is serving, wanders the world,
marries well and sires children, fights valiantly in a war, and then, growing
old, goes home to die in peace. Sinuhe’s wandering, his military glory, and
his eventual return may remind us of some Greek heroes’ stories. What is
missing, however, is any encounter with “the marvelous”—with monsters
or gods. The other is “The Tale of the Shipwrecked Sailor,” which tells of a
man who drifts to a mysterious island where he is befriended by a gigantic
serpent. Certainly, we have the element of the marvelous here, but the
story’s protagonist remains unnamed—he is a sort of “everyman”—and (as
Geraldine Pinch has observed) he fails to do anything that is particularly
brave or strong.59 We also have a tale about five divinities helping a human
woman give birth to triplets who are the sons of Ra and are destined to
become kings (in which sense they are no different from all other Egyptian
kings, who formally traced their paternity to Ra) and traces of another tale

about a goddess trying to seduce a human herdsman. These two are closer,
at least in plot, to some of the myths about heroes that we know from
Greece.60 From New Kingdom Egypt (1550–1069 BCE), we have the “Tale
of the Two Brothers,” in which, much as in myths about Hippolytus and
Bellerophon, a man is falsely accused of raping a woman—in this case, his
brother’s wife. He is subsequently betrayed by his own wife, as are several
Greek heroes in various circumstances. The problem with calling this story
a hero myth is that the names of the two brothers, Inpu and Bata, are those
of gods.61 From Roman Egypt (31–311 CE), we get stories about magicians
who have adventures. Several of these are about a magician named Setna
who, like many heroes, journeys to the Underworld and sees the horrors of
the souls in torment. Given its date, it is likely to have been influenced by
Greek myths about several heroes’ journeys to Hades.62
My survey has turned up approximately twenty-five ancient Near Eastern
stories about heroes63 and more than thirty about the gods, giving us a ratio
of five-to-six to set beside the roughly three-to-one ratio that I calculated
for Greek myths. My figures are approximate in the sense that some people
might quibble with exactly where I drew the lines between hero stories and
stories of average people in the Near Eastern material or where I drew the
lines between what counted as a story about a god and a story about a hero
in Greece. My survey is also approximate insofar as I may have missed a
story or two from the Near East, in spite of my attempt to be complete.
Nonetheless, my survey suggests that the Greeks preferred stories about
heroes to stories about gods significantly more than did people in Near
Eastern cultures. Most of the Near Eastern characters whom I have
characterized as heroes, moreover, cluster together on one half of the
spectrum that I sketched for Greek heroes in the first section of this chapter
—the half that runs from Menelaus to Thymoetes and Iops. That is to say,
many of the Near Eastern characters interacted with gods and were great
warriors, but unlike Perseus or Heracles, for example (who sit at the other
end of my spectrum), relatively few of them conquered any monsters or
performed other superhuman feats. The heroes whom we meet in Near
Eastern sources differ from Greek heroes in another way, as well: there are
not enough of them in any single culture to form a strong and extensive
network, as the Greek heroes do. Some of them are generationally linked in
a short-term manner—Lugalbanda is the father of Gilgamesh, Sargon is the

grandfather of Naram-Sin and Danel is the father of Aqhat, for instance—
but they have no complex web of interrelationships with one another or
with a larger cadre of heroes.
We might try to rationalize the relative dearth of hero myths that we find
in the ancient Near East by positing that Near Eastern stories reflect what
scribes were expected to record, and by hypothesizing that additional stories
focusing on heroes circulated less formally, through channels that are now
lost to us. But such a hypothesis only brings us to the question of why those
scribes chose to focus on divine stories in the first place, in contrast to
Greek poets. Or to put it otherwise, why did stories about heroes prove to
be such an idiosyncratically attractive topic for the Greek poets and their
audiences?64
The prominence of the Greek hero in myths is matched by the
prominence of hero cults in Greece, which were established as early as the
ninth century and were flourishing by the mid-Archaic period.65 In contrast,
there is almost a complete absence of anything like hero cults in Near
Eastern cultures. Although many ancient Near Eastern peoples paid some
type of cult to their dead ancestors at least in the years immediately after a
death, there is no trace of cults in honor of specific, named individuals that
endured through generations, after their immediate descendants were gone,
and that spread throughout a wider group.66 The distinction I am making is
exemplified by the cult paid to the Ugaritic rephaim, a collective group of
dead warriors and kings who were understood to serve Baal and Anat and to
have some ties to the fertility of the harvest. Although they were imagined
to have a leader, who was referred to as rp’u (the singular of rephaim), and
to have anthropomorphic features insofar as they were imagined to feast
and drink at the time that they were celebrated during rituals, the group of
rephaim was just that—a group—without any individual names or
individual personalities and stories as far as we can tell.67
The presence of hero cult in Greece cannot be used to explain the
prominence of hero myths in Greece, however, for it is impossible to say
whether the habit of paying cult to heroes is older than the flourishing of
hero myths, the myths are older than the cults, or—as I think likely—the
two developed concomitantly, nudged along by other things. One of the
most important of these was probably the negotiation of Greek identity
during the Dark Age and the early Archaic Age (approximately 900–700

BCE). The development of a shared pantheon of gods during this period
and shared sanctuaries to those gods that drew worshippers from far and
wide was one of the factors, as has often been pointed out, that helped to
bind independent groups (Spartans, Thebans, Argives, and so on) into a
collectivity that could understand itself, when it chose to, as being “Greeks
together.” But we can imagine that in response to this amalgamation of
originally separate local divinities, each independent group would have
clung all the more firmly to its own, separate traditions about exceptional
men who had emerged from its local soil, and expanded upon those men’s
reputations as powerful entities by further developing both narratives about
them and rituals performed in their honor.68 In contrast, as Jan Assmann has
argued,69 ancient Near Eastern cultures that were ruled by a single
individual, such as Mesopotamia and Egypt, had no room for a plurality of
other humans who achieved divine or semi-divine status—this was reserved
for royalty alone. This does not resolve the quandary completely, however;
the Greeks wove many figures such as Theseus into the early parts of cities’
king lists or embroidered the deeds of early historic kings until they looked
like heroes. Other cultures could have found ways to embellish more of
their kings or cultural leaders into heroes in the Greek sense as well—but
few of them chose to do so. Once Israel had embarked upon monotheism,
for example, it refused to grant superhuman power to even its most
important founding figures.70 Christianity, in contrast, embraced saints’
cults early in its development. That is, while formally remaining just as
monotheistic as Judaism, Christianity developed a penchant for honoring
exceptional humans of the past who reputedly could help those who paid
them cult—heroes, in a word. The popularity of Greek hero myths and hero
cults may have had some influence here.
Another factor that may have contributed to the rise of the hero in Greece
is the Greeks’ perception that they were wanderers by nature—that they had
come to the Greek mainland from somewhere else and that they would
continue to move onwards, colonizing new lands.71 With the exception of
the Israelites (whose stories, as I noted earlier, also focused on exceptional
humans), no other ancient Mediterranean culture chose to understand itself
in this way; most peoples, in fact, firmly claimed that they had always been
in the same place, from time immemorial. Groups that perceive themselves
as relatively mobile may have a greater need to assert their attachment and

entitlement to a place they currently inhabit. One way of doing that is to
develop myths about ancestors who were connected to that place—heroes
who were born there, who performed their great deeds there, or who died
there. It’s harder to make similar claims about one’s gods. The Eleusinians
could brag that their ancestors had welcomed Demeter during her search for
Persephone and the Delians could claim that their island had been the
birthplace of Apollo, for instance, but these and other gods were
panhellenically at large to a much greater extent than any hero except
Heracles ever was; because of this, ties to the gods could never feel as
intimate and exclusive as could the ties to heroes.
In the end, one can only make conjectures about why the hero became so
important in both narratives and cults in Greece. Perhaps it is more fruitful
to contextualize the phenomenon of the hero by noting that other
institutions and practices that highlight the potential of the human and the
value of the individual were first manifested in Greece, amongst ancient
Mediterranean cultures, as well. Athens developed the first democracy. It
was in Greece that there first arose religious cults promising individual
members an enhanced afterlife, separate from the crowd of other listless
souls trapped in Hades (some inscriptions attesting to these ideas even call
these privileged individuals “heroes”). Greek artists were the first to pursue
anatomically and proportionally accurate portrayals of the human body,
exalting its form. Hero myths, which suggested that there had once been
exceptional human individuals who came close to the status of gods in their
deeds, and hero cults, which suggested that those same individuals, after
death, could come close to the gods in their power to affect mortal lives, fit
well within this general trend of celebrating the human potential.
The Canonical Greek Hero

Over time, as Greek poets were hired to elaborate upon each hero’s
accomplishments and then carried those stories abroad to other cities, each
hero’s luster would have been burnished both by his wider fame and by his
slow accrual of characteristics from other heroes’ stories that similarly
traveled throughout Greece. Inevitably, an informal, if unarticulated, canon
of heroic traits developed. Few Greek heroes partook of all its elements, but
collectively their stories presented the ideal hero as someone who (1) had a
divine parent; (2) received help or advice from the gods; (3) relied upon
help from a sibling or friend to accomplish at least some of his tasks; (4)

used intellectual, as well as physical, skills to meet challenges; (5) was a
good warrior; (6) founded cities; (7) established dynasties; (8) killed or
conquered monsters; (9) journeyed to distant places, including Hades; and
(10) remained active after his death.
Some of these traits will be discussed in detail later in this chapter, but
here, before we finish comparing Greek heroes with heroes from other
cultures, I want to touch on four of them that Greek heroes exhibit to a
higher degree than do most heroes in other cultures:
(1) Relatively few heroes from other cultures have a divine parent. In the
ancient Mediterranean, only Gilgamesh is an exception (if we leave aside
the Egyptian habit of formally declaring that all pharaohs were the sons of
Ra). Gilgamesh’s mother was the goddess Ninsun, and his father was the
mortal king Lugalbanda. The Epic of Gilgamesh tells us that this made
Gilgamesh two-thirds divine, suggesting that at least in Mesopotamia, a
divinity’s genetic contribution to a child was understood to be twice as
strong as a mortal’s.72
If we go further afield geographically and chronologically, we find some
other exceptions, such as the Norse hero Sigi, who was sired by Odin upon
a mortal woman and who became the progenitor of the line that led to
Sigurd, the greatest of Norse heroes.73 Other exceptions are the five
Pandava brothers of the Mahabharata, who were sired by five different
gods upon the two human wives of King Pandu.74 Of course, Jesus Christ is
another example.
Having a divine parent more clearly marks a hero as someone who is
situated between humans and gods. By his very existence, the hero implies
that the two groups are not impossibly different. Divine parentage, and the
divine siblings that come with it, also provide a narrative motivation for the
many instances of divine help that heroes typically receive. In other words,
in Greek myths, gods help the hero not only because he has demonstrated
special abilities or traits that make him more deserving of help than other
humans but also because he is as close to being one of them as any human
ever can be. A black-figure lip cup dating to about 560 BCE that shows
Athena excitedly tugging her half brother Heracles by the hand towards
their father Zeus, eager for Heracles’s apotheosis to be completed, is a
touching expression of this familial feeling (see Figure 7.2).
This trope plays out in some other variations, too. The hero Pelops, for
instance, was born of two mortal parents but became the lover of Poseidon,

who later gave him a golden chariot and winged horses in order to ensure
that Pelops would win a chariot race. Cadmus, Menelaus, and Peleus
married goddesses, which guaranteed them places in an idyllic afterlife
retreat. Having familial or romantic connections to the gods, then, sets the
hero apart not only genetically in some cases but more importantly socially
—the hero runs with the right crowd. We see this in association with the
heroes of some other cultures, too: Sargon is the favorite of Ishtar / Inanna,
for example, and Odin helps several Norse heroes at crucial moments. The
gods also helped Greek heroes who were not their relatives or lovers,
Odysseus being the most notable example. In contrast, some heroes made
careers without any apparent help from the gods at all (Meleager, for
instance). As a rule, however, the Greek hero was distinguished from other
humans by having close ties to the gods.
7.2 Athena eagerly tugs her half brother Heracles towards their father, Zeus, following
Heracles’s apotheosis. Black-figure lip cup from Vulci by the Phrynos Painter, dated to
about 540 BCE (detail). Now in the British Museum, inv. num. 1867.5-8.962. Photo ©
Trustees of the British Museum (CC BY-NC-SA 4.0).
(2) Many Greek heroes kill or conquer monsters. Some, such as Heracles
and Theseus, kill or conquer several.75 From approximately the same time
and place as our Greek examples, we can cite only Gilgamesh and his
friend Enkidu, who kill Humbaba and the Bull of Heaven.76 For additional
examples of monster killers, we have to move forwards in time and further
afield geographically (although my sample will stay largely within the
Mediterranean and Europe, venturing no further east than India). The

Avestan hero Thraetaona defeats the dragon Azi Dahaka.77 Rustam, the
main hero of the tenth-century Persian epic the Shahnameh, kills a dragon,
as do several other characters in this lengthy poem, which scholars have
shown to be indebted to Greek stories.78 Rama, the semi-divine protagonist
of the Indian Ramayana, dating perhaps to the fifth century CE, defeats
Ravana, a demon who has kidnapped his wife, Sita.79 In poems dating to the
seventh century, the Irish hero Cuchulain kills a dog that had “the strength
of 100 dogs,” three monsters who manifested themselves as cats, twenty-
seven sea wraiths, a great slimy worm, and a giant who emerged from the
sea.80 In a tale at least as old as its tenth- or eleventh-century manuscript, we
hear about the Scandinavian hero Beowulf killing Grendel, Grendel’s
mother, and a dragon who simultaneously kills Beowulf himself.81 In texts
and representations as early as the eleventh century, the Norse hero Sigurd
kills a dragon variously named Mimir or Fafnir.82 Saint George and several
other medieval knights also slay dragons, and several Arthurian knights
pursue the Questing Beast (a hybrid animal made from parts of a snake,
leopard, lion, and hart). King Arthur himself, according to early sources,
defeats a cat monster, a divine boar, dog-headed monsters, dragons, and
giants.83 European folktales and fairy tales take up the idea of monster
killing, as well—there are many variations of the story in which a small
man fools and subsequently kills a giant and some instances of dragon
killing as well. These are not hero myths in the sense that our Greek
examples are, however; the protagonist either is nameless (“a tailor,” “the
youngest brother,” “the prince”) or carries an all-purpose name such as
Jack. He has no other history or connections to other characters in other
stories.
I will discuss Greek monster killing in more detail later in this chapter;
suffice it here to say that whatever else such episodes accomplished in
myths, they made for thrilling narratives and also that, as far as our sources
show, the Greeks told stories about monster killing in greater abundance
than other ancient peoples did; probably, the Greek stories provided models
for many such stories that were told in later centuries.
(3) Heroes from most cultures travel to distant places, but few of them,
outside of Greece, journey to the realm of the dead and return. Gilgamesh
travels through the Underworld on his way to visit Utnapishtim, from
whom he hopes to learn the secret of immortality. Previously, he had sent

his friend Enkidu to the Underworld to regain some lost possessions, but
Enkidu could not return; the two friends have their last conversation either
through a crack in the earth or during a dream (the text is unclear).84 In the
course of this conversation, Enkidu tells Gilgamesh what the Underworld is
like. In the sixth book of The Aeneid, Vergil sends the Roman hero Aeneas
to the Underworld to learn about the future of the kingdom he will found.
Christ harrows Hell before ascending to Heaven, and late antique
apocryphal works claim that various other biblical figures tour Hell (and
sometimes) Heaven as well, bringing back reports of what kind of person
ends up in either of these abodes and what sorts of treatment they receive.85
The unnamed mother of the Finnish hero Lemminkäinen descends into Hell
to recover the body of her son, which she then revives.86
The Greek heroes who visit the Underworld, like Gilgamesh, Enkidu,
Aeneas, and Lemminkäinen’s mother, have personal reasons for making the
trip, which distinguishes them from the heroes of the apocryphal works,
who, like Christ, act out of an interest in the welfare of the entire human
race; many of the latter, moreover, do not make the journey of their own
initiative but instead are snatched away by God or an angel. The Greek
heroes, then, fall into line with most Mediterranean travelers to the
Underworld in demonstrating a pronounced determination to obtain or
regain whatever it is that they seek there for themselves—a lost wife in the
case of Orpheus, a new wife in the case of Pirithous (and therefore also his
traveling companion, Theseus), the dog of Hades in the case of Heracles,
and knowledge of what awaits him at home in the case of Odysseus. It is
the Christian travelers who stand apart.
If Greek hero myths are about pushing the boundaries between human
and divine, then the ability to visit the land of the dead certainly fits this
model. The problem is that, of the four major heroes who visit Hades—
Heracles, Theseus, Orpheus, and Odysseus—one (Theseus) cannot return
without help from another one (Heracles), and in doing so he must leave his
friend Pirithous behind forever in the land of the dead; a second hero
(Orpheus) returns, having failed in his quest and soon thereafter dies as an
indirect result; and a third (Odysseus) is never truly in the Underworld at all
but rather lingers on its border, calling the souls up to where he waits to
interview them. Only Heracles can be said to successfully journey to the
Underworld and back with his goal accomplished, thereby truly conquering
death—a feat that is echoed in the tale of his winning back Alcestis’s soul

in a wrestling match against Death itself. Even Heracles needed help from
the gods to complete this journey, as a number of vases demonstrate by
adding Athena, Hermes or both to the scene of his victory over Cerberus (as
we see on the black-figure hydria from about 510 BCE that is shown in
Figure 7.3, for example).
Not coincidentally, Heracles, the only hero to successfully return from
the Underworld, is also the only Greek hero to become a full-fledged god,
ascending to Olympus even as his mortal body burns on a pyre. Most Greek
myths about journeys to the Underworld underscore the idea that, although
heroes are more resourceful and daring than other humans, in the final
result death wins out.
7.3 Heracles leads away a chained Cerberus while Athena and Hermes look on. Attic
black-figure hydria by Painter S, dated to about 510 BCE (detail). Now in the Toledo
Museum of Art (Toledo, Ohio). Purchased with funds from the Libbey Endowment,
Gift of Edward Drummond Libbey, 1969.371. Photo credit: Richard Goodbody Inc.
(4) In contrast to the lack of hero cults elsewhere, hero cults were a
distinctive characteristic of Greek religion—which brings us back, again, to
the point that heroes are defined by their relationship to death; they must
eventually die if they are to become the powerful figures who could help

their worshippers of later ages. This leads us to the next section of this
chapter.
The Hero as Narrative Character

The time during which the heroes walked the earth amongst other living
humans was understood by the historical Greeks to have preceded the time
when they themselves were living. By the time that the heroes’ stories were
being told, therefore, the heroes were technically dead, but as recipients of
cult, they were understood to wield significant power: they could protect
cities at times of war, fend off plagues and perform various other sorts of
feats. Part of the heroes’ appeal lay exactly in their ambiguous nature, in
fact: as godlike entities who had once been mortal, they could be assumed
to have more empathy for humans than the gods would have, and as former
mortals who now had godlike powers, they could put that empathy to work.
Yet their ambiguous nature presented storytellers with an interesting
dilemma: how do you juggle making a hero human enough that listeners
will feel that he had once been one of us and demonstrating that he was
extraordinary enough for listeners to accept the idea that he rose to a higher
level after death?
It’s a dilemma that brings us back to the broader question of how myths
help to create and sustain beliefs, which I began to explore in Chapter 3.
For another way of articulating the issue I just sketched is to say that, by
straddling two categories that were otherwise mutually exclusive (mortals
and immortals), heroes should have provoked a cognitive dissonance. This
dissonance would have been amplified by the fact that whereas cult practice
emphasized the heroes’ superhuman status (after all, why make offerings to
someone who wasn’t more powerful than you?), myths emphasized the
heroes’ lives as humans and their very human deaths: there are almost no
myths that talk about what a hero did after he ceased to be human and
moved on to higher things.87 Particularly when myths about heroes were
performed in concert with their cults (that is, when stories about the human
adventures of a hero were told in order to honor and please him in his now
more-than-human manifestation)—and perhaps even more particularly
when myths about heroes were performed in concert with the cult of a god
(that is, in concert with myths that further emphasized, through contrast, the
erstwhile mortality of heroes, most of whom had suffered at the hands of
the gods while they were still alive)—myths about heroes had special work

to do, if the ambiguous nature of the heroes was to be made not only
credible but also fruitful.
Greek storytelling developed ways of doing this, but to think about them,
we need to return briefly to two topics that were introduced in Chapter 5:
plurimediality and accretive characters. I suggested that plurimedial
portrayals of mythic characters laid the groundwork for belief in their
existence by compelling each individual in the audience to develop his or
her own conceptualization of a character, which led to strong cognitive and
emotional bonds between that individual and the character. Each person had
his or her own “Theseus,” so to speak, made from pieces accrued from the
different representations of Theseus to which he or she had been exposed
since childhood. In a culture where these were not just “characters” in our
usual sense of that word but entities whom most authorities—religious,
political, poetic—said were real and able to affect the lives of humans for
better or for worse, plurimediality made a significant contribution to the
creation and sustenance of belief.
But plurimediality supports belief in the characters of myths in another
way as well. In a plurimedial environment, as I noted in Chapter 5, it is
impossible for audience members not to notice that varying—sometimes
strikingly varying—portrayals of a character coexist. Under these
circumstances, every story, however old and however popular, must be
accepted as just a single report about a character who can never be fully
known by the people who listen to his story. Such tolerance for varied
opinions made it easier to continue developing new stories about a
character.
Serials and Series

The topic of new stories brings us back to something else that I discussed,
in Chapter 3. I noted there that narratologists have shown that telling a story
episodically prompts people to engage with the story and its characters in
between episodes, and I suggested that this also helps to sustain a belief that
those characters exist and are poised to intervene in the lives of the
audience—assuming, again, that authorities within the given culture support
the idea that the characters exist.
I want to return now to the observation that there are two different types
of episodic narration, each of which has its own characteristics. In episodic
narration that constitutes a serial,88 there is a chronologically determined

sequence of episodes, which ideally are provided to an audience in a
particular order. The modern examples that I used as illustrations in Chapter
3 included books such as George Eliot’s Middlemarch and Charles
Dickens’s The Old Curiosity Shop, which were initially published in
installments, and TV shows such as Homeland and Downton Abbey. As I
also discussed there, some audiences will consume the episodes of a serial
out of their proper order, but the awareness that there is an order—that there
is a narrative arc that stretches across the episodes—nonetheless affects
how the narrator crafts the story and therefore how the audience reacts to
each episode. The longer arc more easily allows a narrator to portray
characters undergoing changes—changes in status, in states of existence, or
in moral and ethical outlooks. Traces of these ongoing changes are found in
each episode, even if it is consumed independently of those that come
before or after it.
Other episodic narratives are series. That is, there is no narrative arc that
stretches across all of them. Instead, the episodes are constructed in such a
way that the audience’s attention focuses closely on one self-contained
story and then on the next and then the next again after that. This makes it
difficult to show significant development within a character. And in fact,
significant development is usually undesirable in this sort of episodic
narration: the coherence of a body of stories in a series typically relies on
the continuing presence of one or two well-known characters, such as
Sherlock Holmes and Dr. Watson. Were these characters to change
significantly from one episode to another, some of that coherence would be
lost. Many daily comic strips are series, too. In the course of the ten years
during which Bill Watterson narrated the adventures of Calvin and Hobbes
(1985–1995), neither boy nor tiger matured one whit.
Greek stories about heroes partake of both types of episodic narration—
serial and series. Heroes’ stories were virtually always delivered in the way
that series are delivered: in discrete doses with little regard for whether the
audience members had already heard formal narrations of chronologically
earlier parts of the heroes’ adventures, which were able to stand on their
own or with only brief references to earlier events. Celebrants of
Hippocleas’s victory in the boys’ double-course race at Delphi in 498 BCE
heard Pindar’s story about Perseus’s trip to Hyperborea, for example, but
not about the rest of Perseus’s adventures.89 Anyone present when a
rhapsode recited what we now call Odyssey 12 would have heard about

three of Odysseus’s greatest adventures—his encounters with the Sirens,
Scylla and Charybdis and the cattle of Helios—but that would still be only
three out of many. Although tragedies sometimes offered “prequels” and
“sequels” to their main plot through the mouths of gods and ghosts who
appeared briefly at the beginning or end of the action,90 each tragedy
focused closely on just one episode, which tradition decreed must be
dramatically contained within a single day. Thus, for example, those
attending the Dionysia of 409 BCE would have heard Sophocles’s story
about Philoctetes’s encounter with Neoptolemus (in the Philoctetes) but
would have gotten only the briefest résumé of what happened to these two
heroes next, when Heracles made an epiphany at the end of the play.
There were exceptions; there were occasions on which a hero’s entire
story was narrated, albeit succinctly. Some of them, interestingly, occurred
when a character within a myth was narrating another myth (or at least
that’s how it looks from our perspective). In some of these cases, the
internal narrator used the myth to try to persuade someone else of
something: Diomedes told the whole story of Lycurgus in Iliad 6 when he
wanted to impress upon Glaucus that the only opponent he would fear
would be a god. And then Glaucus narrated the entire story of Bellerophon,
at even greater length, because he wanted to impress Diomedes with his
ancestral lineage. Phoenix told almost the whole story of Meleager in Iliad
9, when he wanted to convince Achilles to return to battle.91 In some such
cases (Lycurgus, Meleager), it was necessary to tell the full story in order to
get the moral across (the audience needed to see the hero brought low at the
end of his life). In other cases (Bellerophon), the full story enabled the
narrator to squeeze in all the glory he could (although it must be said that
this hero, too, was brought low at the end of his life).
Narrations such as these prove that people were familiar with heroes’
complete biographies, and some of the cases also suggest that people could
informally narrate those biographies themselves when they wanted to.
Because of this—because at least the general outline and the high points of
each hero’s life were widely known—no single episode taken from a hero’s
career was ever completely disembedded from the longer tale of the hero’s
life.92 Each was understood to be part of the longer narrative arc that
necessarily stretches over every human life—audience members knew that
every hero, like them, had been born a mortal and that every hero had
experienced a mortal death, as they would. Or in other words, each episode

implicitly participated in a serial, even if it was narrated in the manner of a
series.
We should pause to take note of how much hero myths differed from
divine myths in this regard. Gods escaped the chains of temporality almost
completely. Gods were born (the Greek poets took particular delight in
narrating their births and childhoods)93 but they did not die, and very few
other events in which they participated brought about any significant
change—at least for them. Aphrodite’s machinations led to the death of
Hippolytus, Artemis’s favorite acolyte, and in retaliation, Artemis contrived
the death of Aphrodite’s lover, Adonis, yet there is no indication that either
of these human deaths had a lasting effect upon the two goddesses
themselves.94 The argument between Zeus and Poseidon over who would
take Thetis’s virginity led eventually to the Trojan War, in which every god
took sides and sometimes even fought on the battlefield. But aside from a
minor scratch here and there, it was the mortals who suffered, not the
gods.95 The lives of the gods truly ran as series, insofar as they continued
forever, and few episodes within them had links to other episodes or carried
lasting repercussions for the gods themselves. Trying to arrange a god’s
adventures along a timeline is nearly impossible. Even those of us well
versed in Greek myths would struggle to say, for example, whether we
should understand Aphrodite to have adopted the orphan daughters of
Pandareus before or after she abducted Phaethon, the son of Eos and
Cephalus. If we are determined to have an answer, the only way to construct
one, notably, is from what we know about the humans involved—their
adventures, familial lines, and relationships with one another. For the
record, having performed these scholarly gymnastics with the help of the
family trees appended to Timothy Gantz’s Early Greek Mythography, I
conclude that Aphrodite’s adoption of the daughters of Pandareus preceded
her abduction of Phaethon by one human generation. I base this on two
points: (1) Pandareus was a companion of Tantalus, which makes him two
generations older than Tantalus’s grandson Pittheus; (2) Pittheus’s grandson
Theseus, according to the Athenian family tree, was two generations
younger than Phaethon, whose mother was the goddess Eos and whose
father was Cephalus, who was married to the Athenian princess Procris
during the time he dallied with Eos and sired Phaethon. Thus, Phaethon was

of the same generation as Pittheus and thereby younger than Pandareus’s
daughters by a generation.96
Yet not even the late antique mythographers tried to impose upon the
gods such detailed chronologies as I have just compiled for the families of
Pandareus and Phaethon. Once we are past the early days of cosmic warfare
and Zeus has settled onto the throne, the gods’ experiences are seldom
perceived sequentially in the same way that humans’ are, and any
relationships that the gods have with one another—of love, of envy, of
enmity—are fleetingly transitory, dwarfed by the unity of the gods as a
whole. The scene in the Homeric Hymn to Apollo in which a group of the
gods—Aphrodite, Artemis, and even Ares, amongst others—merrily dance
together to the sound of Apollo’s lyre is emblematic not only of their dolce
far niente attitude but also of their cliquish solidarity.97 They may squabble
over such matters as whether Odysseus should be allowed to reach home,
but neither the affairs of mortals nor their own affairs cause serious,
enduring changes to their living conditions. One of the few exceptions to
this rule brings us to a story that scholars have found odd for a number of
other reasons: when Hades kidnapped Persephone, it changed Persephone’s
mode of existence forever. However hard her mother fought it, Persephone
was compelled forevermore to spend part of her time in the Underworld.
For Persephone, there was a narrative arc, and it did bend downward.
(Although, strangely enough, it then bent up and down again every year.
Was this truly an arc or rather just a particularly weird instance of the
episodic nature of gods’ existences?)
The implicit serial within which each hero’s story was told—the sense
that there was a beginning and an end, however many episodes lay between
—reflects the greatest difference between mortals and immortals, then. It is
not simply that mortals die and gods do not; more generally, mortals exist in
a world where the passage of time inevitably brings significant change and
immortals live in a world where it does not. In contrast to other heroes of
series—say, Sherlock Holmes—Greek heroes are meant to die.98 They must
die, in fact, if they are to become the powerful entities who are worshipped
in cult, and narratives must confirm that mortality by representing it. Heroic
biographies take this even further: the disasters that befall heroes typically
come at the end of a chain of events that slowly but surely concatenate over
the course of many years that lie between. After killing the Hydra, Heracles
dipped his arrows in its poison and later used one of them to kill Nessus, the

centaur who was trying to rape his new bride. With his dying words, Nessus
contrived to guarantee that, many years later, the bride herself would
unwittingly murder Heracles with that very same poison.99 Laius, in fear of
an oracle that his own child would kill him, exposed his newborn son,
Oedipus, who nonetheless grew up and, having received an oracle that he
would murder his father, ran blindly into the very act of patricide that both
he and Laius had sought to evade.100 And so on. This was the stuff of
tragedy, quite literally, and part of tragedy’s brilliance was its ability to
collapse all the freight of a heroic past that had been narrated as a series
into a precise moment, a single episode in which its debts came due and the
serial was clearly revealed as what it had really been all along.
As a genre, then, tragedy thrived on the tension that could be created
between the series nature of Greek mythic narration (the fact that it focused
on single adventures and circumscribed events) and the inevitable mortality
of its protagonists (the fact that they lived within a narrative arc that had a
beginning and an end). But to return to the topic of series narration more
generally, when a character is presented in a series, the self-contained
episodes between his or her implicit birth and death can be multiplied
almost infinitely, as either need or desire demands. Who is to say how many
opponents either Sherlock Holmes or Heracles defeated? And who will
object if another is introduced, as long as the story is engagingly narrated?
This flexibility enabled poets to graft new events into heroes’ lives to suit
the occasions for which their poems were commissioned (and so Pindar, in
his third Olympian ode, could contrive to have Heracles travel to
Hyperborea to fetch olive trees to beautify the Olympic precinct) and
enabled local populations to pull famous heroes into their orbits (and so the
people of Megalopolis could tell of how a maddened Orestes bit off his own
finger while visiting their neighborhood, hoping to assuage local goddesses
called the Crazy Ones).101 But this flexibility also granted the heroes
something similar to the eternity in which the gods existed: the golden mid-
stages of their careers could be expanded indefinitely.
Multiplication could not be allowed to become rote, however; a hero who
killed the same type of opponent in the same way too many times risked
boring his audience. This brings us back to the Greeks’ tolerance for—
indeed, their apparent desire for—hearing different versions of a well-
known story; such variation was yet another way of keeping a hero fresh, of
lingering on a favorite tale without becoming sated. Poets might do this by

offering new insights into heroes’ morals and motivations: Euripides’s
Orestes is certainly different from that of Sophocles, who is again different
from that of Aeschylus. Or rather, I should say, Euripides’s four Orestai are
different—not only from those of Aeschylus and Sophocles but also from
one another: they range from a young hero off on a quest that restores his
sister to home, in the Iphigeneia in Tauris, to a treacherous murderer in the
Andromache. This is more than a matter of giving a hero a behavioral
facelift just for the sake of variety or politics; each of these representations
provided audiences with another opportunity to engage with the hero and
the range of possible responses he could make to his circumstances. In
contrast to characters such as Elizabeth II, whom I discussed in Chapter 5,
there was little to constrain the narrative development of heroes.
Iteration
Some types of multiplication generate what Umberto Eco calls “iteration.”
Eco suggests that characters whose adventures are narrated serially—his
prime examples are Superman, Hercule Poirot, Nero Wolfe, and Lieutenant
Columbo—please their audiences by repetitively engaging in the same sorts
of adventures and displaying the same behavioral (and sartorial) tics over
and over again. Their predictability rewards loyal audiences with feelings of
knowledge and intimacy, which in turn heightens the audience’s desire to
hear further stories about the character.102
Greek hero myths engage in iteration in two ways. First, certain heroes
repeatedly perform certain types of tasks or apply certain types of skills to
overcome challenges. Odysseus uses his cunning to outwit young Achilles
on Scyros, Polyphemus in his cave and the suitors on Ithaca, amongst
others. Heracles repeatedly conquers monstrous animals (the Hydra, the
Erythmanthian Boar, Cerberus, and so on) and barbarians who engage in
aberrant customs (Antaeus, Busiris, Cycnus), and typically he wears the
same lion-skin “outfit” to do it—a sartorial tic.
Second, Greek heroes collectively iterate certain behaviors. Although no
one had as many notches in his club as Heracles did, Theseus and Odysseus
each overcame several monsters, Perseus two, and Bellerophon, Cadmus,
Oedipus, and some lesser-known, local heroes overcame one monster
apiece. In doing so, they each contributed to an expectation that conquering
monsters is something that heroes do, and thereby stoked audiences’

appetite for hearing about it again. Artists also fueled this fire, as shown in
Figures 7.1, 7.6, 7.7, and 7.9, and in metopes 2, 9, and 11 of Figure 7.4.
Eco also argues that iteration makes series ideal mechanisms for sending
a simple message because the message is repeated over and over (whether
the stories’ narrators are conscious that they are doing so or not). What
message, or messages, might we understand Greek hero myths to be
sending? The most obvious answer is that there had once been a time when
at least some humans were bigger, better and nearer in nature to the gods—
the same message sent by the mythic genealogies that I discussed in the first
section of this chapter.
But when we look more closely at other iterated features, we see
something different. Frequently, these features emphasize the hero’s
humanness and thus his distance from the gods. The presence of one or
more companions at his side as he performs his labors, for example—with
Theseus, it’s Pirithous; with Jason, it’s the Argonauts, for instance—sets
him firmly in the human realm. Gods form alliances to accomplish specific
tasks—Hera, Hypnos, and Poseidon collude against Zeus in Iliad 14, for
instance103—but these alliances are as shifting as the sands; the gods have
nothing as enduring as friendship. The hero’s frequent need to seek, and
sometimes then interpret, information from a higher authority—an oracle, a
mantis, or a god in disguise—is human as well. In contrast, I can think of
only three or four instances in which gods seek information: Zeus attempts
to learn from Prometheus upon which female he should not sire a son;
Demeter attempts to learn what has become of Persephone; and in the
Homeric Hymn to Hermes, Apollo attempts to learn, from an old man,
where his stolen cattle have gone—an episode that I presume is meant to
parody the many instances in which human heroes sought information from
Apollo himself, a role of Apollo’s that the poet of the Hymn later
emphasizes, when he portrays Hermes and Apollo negotiating over which
of them will control which kinds of prophetic knowledge.104 Finally (a
dubious case), Pindar presents us with an Apollo who dissembles,
pretending not to know something so that he can ask Chiron (another
famous prophet) about the identity and lineage of a lovely maiden whom he
has glimpsed, and whether it is right for him to make love to her. Before
answering, Chiron chides the god, asking how it is that someone who
knows how many leaves the earth puts forth in spring, how many grains of
sand lie in the sea and rivers, and all that will happen in its appointed time

cannot discern for himself who the girl is and exactly what his lovemaking
with her will bring about.105
More human yet are the labors in which heroes engage—and, in
particular, the types of physical exertion required by those labors. Greek
heroes are able to use a sword, spear or bow as well as any hero from
another culture, but in many cases, the work they do evokes not that of a
warrior or hunter (pursuits suitable for upper-class men in historical Greece)
but rather that of a craftsman or even an agricultural worker. Homer tells us
in detail about how Odysseus constructs the stake that he used to blind
Polyphemus, the raft on which he leaves Calypso’s island and his marital
bed, using the tools and knowledge of a carpenter.106 Heracles digs a ditch
to clean manure out of stables; Heracles’s traditional weapon, the club,
points towards a humble station in life as well (Theseus often carries a club,
too).107 Jason and his Argonauts also have to dig a ditch, in order to launch
the Argo on the journey that will win them glory, and later they have to
carry the Argo on their shoulders across the Libyan desert.108 Both Cadmus
and Jason yoke animals, plow fields and plant things (even if the crops that
spring up are unusual ones).109 Odysseus knows how to plow, as well, as we
learn from the episode in which he feigns madness by yoking an ox and a
horse to a plow and sowing the seashore with salt.110 At least half of the
monsters that heroes overcome also pull them towards the human side of
things, insofar as these monsters are simply bigger or more vicious versions
of the same animals that the average person might encounter: the
Marathonian Bull, the Crommyonian Sow, the Teumessian Vixen, the
Nemean Lion, and so on. Half of the metopes from the Temple of Zeus at
Olympia that show Heracles at his labors make this clear (metopes 1, 3, 4,
5, 7, and 8, as labeled in Figure 7.4), as do numerous vase paintings of the
labors performed by Heracles and other heroes, such as Theseus.111 One of
the iterated messages in Greece, then, was that the heroes not only had to
undertake remarkable tasks that other humans would not even attempt—
killing Gorgons and Minotaurs, gathering Hesperian apples and Golden
Fleeces—but also had to know how to do humbler, more ordinary tasks.
Indeed, completing the ordinary tasks was often a perquisite for
accomplishing the remarkable ones. Without the skills of a carpenter,
Odysseus would not have escaped from Polyphemus’s cave or from
Calypso’s island. Without the skills of a farmer, Cadmus and Jason would

not have moved forwards in their careers. Without the sweat of the
Argonauts, the Argo would not have been launched.
Perseus is furthest away from this model. He does nothing that mimics
everyday labor, the monsters whom he kills are not found amongst the
familiar fauna, and a large part of his adventures takes place in a realm that
lies completely beyond the known world. But Perseus’s foster father is a
fisherman with the speaking name Dictys (Fishing Net), and it is in Dictys’s
house that Perseus grows to manhood—presumably being taught how to
catch fish as well. It’s tempting to shrug off this detail as a joke—a clever
twist on the fact that Perseus and his mother, set adrift in a chest, were
fished out of the sea by a net—but from early on, Dictys is given noble
genealogies by ancient mythographers; he was regarded as more than just a
clever play on words—and we have no way of knowing which idea
appeared first and precipitated the other one, anyway: Perseus’s rescue by a
fisherman named Dictys or Perseus’s rescue by means of a dictys?112
Some narrators emphasize the human side of heroes even further. The
Iliad and the Odyssey recount their sweat and bodily pain, in battle and on
seemingly endless journeys home. Their essential humanness is further
accentuated in the Iliad and the Odyssey by the choices that the epics’
protagonists make. Achilles, having come to terms with human mortality
after losing his friend Patroclus, returns to battle in spite of the fact that he
knows that this decision will lead to his own death. Odysseus refuses
Calypso’s offer of immortality in order to return to a mortal life with a
mortal wife in spite of having learned from Achilles’s ghost that what
awaits all mortals after death is a yawning emptiness.113

7.4 Drawing of the metopes from the temple of Zeus at Olympia, as printed in
Olympia: Die Ergebnisse der von dem deutschen Reich veranstalteten Ausgrabung, by

Ernst Curtius and Friedrich Adler, vol. 3, plate 45 (Berlin, 1894–1897). The metopes,
which date to between 470 and 457 BCE, show the twelve labors of Heracles.
Tragedies show us heroes functioning in a world that is scarcely

distinguishable from that of their Athenian audiences, populated by
pedagogues, nurses, guards, and other ordinary people. Early in Euripides’s
Alcestis, a group of townspeople gather before the palace, gossiping about
what will happen to their king and queen; a maid from the royal household
brings them news of the events that will drive the rest of the play.114
Euripides’s Medea begins with a conversation between an old nurse and a
pedagogue.115 The action of Aeschylus’s Oresteia is kicked off by a night
watchman stationed on the palace roof, who complains about the “unrestful
couch … drenched by dews” that he has occupied for many a year.116 We
first meet Euripides’s Ion—the hero of his eponymous play—as he is
cleaning the floor of a temple dedicated to his father, Apollo, and chasing
away the birds that would soil it.117 This is a pronouncedly humble start in
life for a divinely sired child who would become king of Athens. Sophocles
shows us the struggles of Philoctetes’s life on Lemnos as he drags himself
groaningly across the stage.118
And tragedy often dwells on the repercussions that heroes’ actions have
for the people who are close to them. In Euripides’s Heracles, Heracles’s
first wife, Megara, her children, and Heracles’s foster father await death at
the hands of an enemy because Heracles has once again abandoned his
family in order to dash off upon a labor.119 In Sophocles’s Trachiniae,
Heracles’s second wife, Deianira, describes her daily life as one of
“nourishing fear upon fear” during his frequent absences and bemoans the
fact that he scarcely knows his own children.120 In Sophocles’s Ajax,
Tecmessa reminds the hero not only of the dire fate that will befall her and
their child if he kills himself out of wounded pride but also of how harsh an
old age his parents will endure if he does so.121 I need scarcely mention the
effect that Jason’s choices have on his wife and children. However poorly
they lived up to them, Greek heroes were shown to have human
responsibilities—for wives, children, and parents—from which the gods
were completely free.
We might expect that epinician odes, at least, would embroider the
glorious side of the heroes’ lives; after all, it was to the heroes that the

young victors celebrated in epinicians were being compared. Bacchylides
meets this expectation in two of his longer epinicians—his description of
Heracles meeting the ghost of Meleager in the Underworld and that of
Achilles and Ajax on the Trojan battlefield are colorful and rousing, for
example.122 In his dithyrambs, too, he paints vivid pictures of heroes’
exploits, in Theseus’s confrontation with Minos and visit to the undersea
palace of his father, for instance, and in the description of Theseus’s
approach to Athens, ridding the Isthmus of monsters and villains as he goes
—a popular topic for vase painters as well, as the images on the early fifth-
century cup in Figure 7.5 show.
7.5 In the center of this image, Theseus upends Sciron, while Athena looks on and
Sciron’s turtle waits below. To the right, Theseus wrestles with Cercyon and at the far
right is the back end of the Crommyonian Sow. Theseus’s adventures on the Isthmus
continue around the outside of the vase, and the image inside shows him fighting the
Minotaur. Attic red-figure cup by Douris, dated to around 480 BCE (detail). Now in the
British Museum, inv. num. 1843,1103.13 (BM E48). Photo © Trustees of the British
Museum (CC BY-NC-SA 4.0).
But Pindar’s treatments of the heroes are often clipped or focus attention
away from their greatest deeds. Pythian 4 spends only seven lines (out of
299) on how Jason yokes Aeetes’s fire-breathing bulls and forces them to

plow straight furrows; we hear nothing of the teeth that Jason must plant in
those furrows and his single-handed battle against the men that spring from
them (a tale that the earlier, epic poet Eumelus had narrated). And although
Pindar goes on to describe the serpent that Jason must confront in order to
snatch the fleece, he truncates this climax of the story, giving it just a single
line.123 In other words, Pindar lingers longest on those of Jason’s deeds that
can most easily be aligned with labors familiar to his audience—even
adding the quotidian detail that Jason used a goad to urge the bulls
forwards.124 Olympian 13 was composed for a Corinthian victor and
therefore Pindar sensibly chose the Corinthian hero Bellerophon as his
theme. Pindar lingers over Bellerophon’s reaction to meeting Athena in a
dream and then summarizes his three greatest deeds—slaying the Amazons,
the Chimaera, and the Solymoi—in just four lines.125 In Pythian 10, Pindar
compresses the highpoint of Perseus’s career into just three and a half of the
seventy-two lines that constitute the poem: “he slew the Gorgon and,
bearing her head adorned with locks of serpents, came to the islanders,
bringing them stony death.”126
It’s not that Pindar is unimpressed by what the heroes do; on the contrary,
he follows the lines about Perseus and the Gorgon with the comment, “but
to me, no marvel, if the gods bring it about, ever seems beyond belief,” and
the lines about Bellerophon with the comment, “The gods’ power easily
brings into being even what one would swear impossible and beyond
hope,”127 phrases that firmly slide Perseus’s and Bellerophon’s
accomplishments over to the divine side of the ledger. Yet Pindar chooses to
emphasize the human nature of the heroes when he can. We can rationalize
such choices by noting that they align well with Pindar’s larger program of
reminding athletes that they should not attempt to be something they are
not, but it was the Greek conceptualization of the hero as suspended
between mortal and immortal and yet essentially human that gave Pindar
the leeway to do this. Bacchylides interrupts one of his most lavish
descriptions of a hero’s accomplishments with a similar comment. In the
middle of his narration of Theseus’s visit to the undersea palace of
Amphitrite, the poet declares, “Nothing that the gods wish is beyond the
belief of sane mortals.”128
I suggested in the preceding section that treating heroes’ stories as series
enabled poets to extend the “golden mid-stages” of the heroes’ careers, thus

giving the heroes something similar to the eternity in which the gods
existed, but that the implicitly serial treatment of heroes’ stories
emphasized the mortal arc of their lives, which always ended in death. I’ve
now suggested, in addition, that the iterative potential of series narration
could underscore either or both of these messages. The ways in which
Greek hero myths were narrated, then, reflected and confirmed the heroes’
suspension between the two otherwise mutually exclusive groups that they
mediated. Yet the preference amongst Greek narrators was to locate heroes
more firmly within everyday life and its challenges. Greek heroes as we
meet them in narratives are unquestionably human.
Monsters and Gods

In the ancient Near East, there were quite a few stories about gods battling
monsters. For instance, in the Enuma Elish, Marduk destroys an army of
dreadful creatures that were called into existence by the primordial seawater
goddess Tiamat—as well as destroying Tiamat herself, who is described as
a dragon-like monster.129 In the Hittite Song of Illuyanka, Teshub defeats the
great sea serpent Illuyanka (whose name means “serpent”) with the help of
his daughter Inara.130 In the Hittite Song of Hedammu, Tarhun and his sister
Anzili battle the sea serpent Hedammu,131 and in the Song of Ullikummi,
Tarhun battles the rock monster Ullikummi.132 Egyptian Ra battles the
serpent Apophis every night to ensure that the sun will rise the next
morning.133 Yahweh battles the sea serpent Leviathan.134
All of these Near Eastern battles have cosmogonic significance. Not only
do they determine which gods will be in charge of the cosmos, but they
often also provide the raw material from which the physical cosmos is
constructed or protect the physical order once it has been established:
Marduk creates the physical world from Tiamat’s corpse, for example, and
Ra’s repeated battle against Apophis ensures that the celestial order of
things will continue, day by day.135 It has also often been noted that most of
the monsters who are defeated in these battles are snaky in form.136 This
strengthens the stories’ cosmogonic messages, for in the ancient
Mediterranean (as in many other parts of the world), the snake was viewed
as a particularly dangerous creature, sometimes even an evil creature, that
by its very nature was pitted against the forces of order.137

In Greek myths, in contrast, battles between gods and monsters are
relatively rare, considering how many more myths in general survive from
Greece compared with those that survive from the ancient Near East. I can
name only five examples. (1) Cronus and his allies defeat a monster called
Ophion—the name literally means “Snaky”—and Ophion’s children, in
order to establish Cronus’s rule at an early stage of the cosmos’ existence
(thus, the story, like the Near Eastern stories, has cosmogonic implications).
(2) Zeus defeats Typhon, who in most versions of this story is a direct threat
to young Zeus’s rule and who has one hundred snakes sprouting from his
waist in place of legs (again, the snakes underscore the cosmogonic
implications of this story).138 (3) Apollo kills Python, a giant she-snake, in
order to establish his Delphic Oracle (more on that shortly). (4) Hermes
kills the herdsman Argus.139 Argus is only a borderline monster, however;
aside from the fact that he has extra eyes, there is nothing unusual about
him. Some sources make him the son of Gaea, like some other monsters,
but others give him human parents.
And finally, (5) all of the gods battle monsters called gigantes in the early
days of the cosmos, in order to defend Zeus’s right to rule the cosmos. We
can’t be certain what these gigantes were. The word is typically translated
as “giants,” but the only possible evidence that gigantes were particularly
large is a passage from the tenth book of the Odyssey that says the
Laestrygonians were more similar to gigantes than they were to humans.140
Because the Laestrygonians themselves are described as large shortly
afterward, we tend to assume that the gigantes were large, too, but Homer’s
comparison may have been meant to suggest, instead, that the
Laestrygonians (and therefore also the gigantes) were non-anthropomorphic
creatures of some sort. That would make them align nicely with three of the
other foes whom Odysseus confronted in books 9 through 13: the Cyclopes
(who had only one eye apiece), the Sirens (who were not described by
Homer but who in texts and art from the later Archaic age are portrayed as
part bird and part human), and Scylla (a maiden who had six dogs’ heads
springing from her waist). The earliest artistic representations show the
gigantes as fully anthropomorphic—sometimes they even wear hoplites’
armor—but beginning in about 380 BCE, visual representations show them
with snakes for legs, as well, as we see in Figure 7.6. Poets seem to have
described them with snaky legs, as well, by the late fourth century onwards,
and certainly some (but not all) of the gigantes have snaky legs when they

make their most spectacular appearance, around 170 BCE, on the Great
Altar from Pergamum.141 Euripides offers us one more tantalizing bit of
information: it was during the battle with the giants, he says, that Athena
killed a Gorgon, whose head she mounted upon her breastplate.142 Euripides
may have invented this story (all of our other sources say that Athena
received her Gorgon’s head from Perseus), but even so, the detail must have
more or less fit with what was generally known about the battle, and it
therefore helps to confirm that the gods confronted creatures who were
“monsters” in our usual sense of that word (that is, they did not look like
normal human beings) and probably monsters with snaky aspects.

7.6 Heracles, wielding his characteristic club, attacks a snaky-legged giant. To the left
is the front portion of a two-headed griffin ridden by Dionysus. Attic red-figure wide-
bellied lekythos dating to the first quarter of the fourth century (detail). Now in the
Antikensammlung, Berlin, inv. num. V. I. 3375. Photo credit: bpk Bildagentur /
Antikensammlung, Staatliche Museen, Berlin / Johannes Laurentius / Art Resource, NY.
I kept the gods’ battle against the gigantes until the end of my list of
divine fights against monsters because it includes an important twist that
sets it apart from almost all Near Eastern monster-killing stories. Already in
Hesiod, we are told that the battle was won by the gods with the help of
Heracles, who personally slew many gigantes and (as Euripides later says)
“raised a victory cry alongside the gods.” Artistic sources sometimes show
Heracles fighting alongside the gods, too, as do the two images from vases
included here (see Figures 7.6 and 7.7).143 All of our textual sources insist
that Heracles was still a living mortal when he participated in the battle; this

gives us a singular Greek instance of gods needing human help to overcome
monsters and a nearly singular instance within the ancient Mediterranean
more broadly. The only comparable story concerns the help that Hupasiya
gave Teshub to overcome Illuyanka, but Hupasiya’s contribution was far
smaller than Heracles’s: rather than doing battle directly with Illuyanka
himself, Hupasiya merely bound the sleeping serpent after the goddess
Inara had enticed it to a feast and gotten it drunk. After the beast had been
tied up, Teshub arrived and delivered the coup de grâce.144
More than a century of scholarship has established that Greek
cosmogonic myths drew significantly on stories from other Mediterranean
cultures. This, combined with the peculiarity of Heracles’s fight against the
gigantes and with the fact that two of the remaining four divine battles
against monsters in Greece (Cronus against Ophion and Zeus against
Typhon) had clear cosmogonic settings, raises the strong possibility that
there was an earlier version of the Greek gods’ battle against the gigantes in
which Heracles did not appear, set at the dawn of the cosmos. This hunch is
strengthened by the fact that the gigantes were children of Gaea and thus
were both genealogically and conceptually parallel to the two other
challengers of Zeus’s rule, Typhon and the Titans, whom the gods overcame
without human help. The snaky legs of the gigantes in at least later
representations also align them with the Mediterranean cohort of
cosmogonically threatening serpentine creatures.

7.7 Heracles, wearing his lion skin and with his club lying nearby, joins the gods in
their battle against the giants (detail of a larger scene). Attic red-figure amphora from
between 410 and 400 BCE, attributed to the Suessula Painter (detail). Now in the
Musée de Louvre, inv. num. S1677; MNB810. Photo credit: Chuzeville / Musée de
Louvre / © RMN–Grand Palais / Art Resource NY.
If Heracles was stitched into what was once a gods-only cosmogonic

battle, this would seem to betoken the rising prominence of heroes in
Greece during the Archaic Age—and traces of similar stitchery can, in fact,
be glimpsed elsewhere in Archaic sources. During his narration of divine
generations, Hesiod pauses on the marriage of two primeval sea gods,
Phorcys and Ceto, from whom springs a family of monstrous creatures,
many of whom have snaky attributes. Ceto bears the three Graeae, Medusa
and the two other Gorgons, Echidna, and the snake that guards the apple
tree that grows at the ends of the earth. Medusa and Echidna, in turn, have

monstrous children and grandchildren of their own: Geryon and his dog
Orthrus, Cerberus, the Lernaean Hydra, the Chimaera, the Sphinx, and the
Nemean Lion.145
Five times in the sixty-seven-line section treating Phorcys and Ceto’s
dreadful offspring, Hesiod pauses to interject comments about how a
particular monster eventually fell at the hands of a human hero. Perseus cut
off Medusa’s head. Geryon, Orthrus, the Hydra, and the Nemean Lion were
killed by Heracles. The Chimaera was slain by Bellerophon.146 These
insertions preemptively remind us, before we reach Hesiod’s narration of
the battle between Zeus and Typhon near the end of the Theogony (820–
880), that humans will be the ones to finish the fight against monsters that
Zeus began. The fact that Typhon is named by Hesiod as the father of three
of the monsters (Orthrus, Cerberus, and the Hydra) underscores the point:
Zeus may have imprisoned a progenitor of monsters, but it was left to
Zeus’s mortal descendants to complete the larger task. In Homer, too,
monsters are mentioned almost always in connection with human heroes
who defeat them: we are told that Bellerophon defeats the Chimaera, that
Heracles overcomes Cerberus, that Meleager and his companions kill the
Calydonian Boar and that Odysseus overcomes several monsters, or at least
(in the case of Scylla) minimizes their impact. In contrast, we hear only
once—and even then briefly, within a metaphor—about Zeus’s defeat of
Typhon.147
Even the story of Typhon itself was adapted to bring mortals to the fore
as the Archaic period rolled on. Most later authors who refer to this story
draw on Hesiod’s version, stating that Zeus hurled Typhon into Tartarus (or
some other underground realm), where he lay forevermore, but the Homeric
Hymn to Apollo, which probably dates to the early sixth century,148 gives it a
full treatment with a different spin. To begin with, the Hymn says that
Typhon was not born from Gaea, a powerful goddess who had existed from
the very beginning of time (thus disagreeing with Hesiod), but rather from
Hera, a younger, Olympian goddess who was Zeus’s wife and his equal,
generationally speaking.149 This change already moves Typhon out of the
most cosmogonically active stage of the universe’s development and into a
period when the world and the gods’ roles within it had already been
settled. Hera’s motivation for bearing Typhon can, in fact, be understood as
a bid not to claim the cosmos but rather to reclaim a right within it that she

and other females had previously held exclusively. For, feeling insulted by
Zeus’s parthenogenic production of Athena and ashamed of her own
botched attempt at single parenthood (which produced the disabled god
Hephaestus), Hera beat the earth with her hands and begged a group of
older gods, many of whom had been displaced by the younger generation
(Gaea, Ouranos, the imprisoned Titans and Tartarus), to send her a son who
would be stronger than Zeus. Hera duly became pregnant with Typhon
without any help from Zeus, but after her first glimpse of an infant who
“looked like neither gods nor mortals,” she handed him over to Python to be
reared.150
In spite of having told us that Hera prayed for a son stronger than Zeus,
the Hymn makes no subsequent mention of any fight between Zeus and
Typhon.151 Instead, it tells us that it was mortals for whom “dreadful and
problematic Typhon” was an “affliction” (pēma). Indeed, it tells us this
twice—once at the beginning of Typhon’s story and once at the end.152
These comments strengthen the analogy between Typhon and Python, the
latter of whom the poet similarly describes, three times, as a monster who
was an affliction (pēma) or evil (kakon) for mortals and their herds.153 By
extension, the comments also strengthen the analogy between the two
monsters’ conquerors: Zeus and Apollo. But in which direction were those
analogies supposed to work? Certainly, by drawing a parallel between
Typhon and Python, the author of the Hymn can be taken to imply that
Apollo’s monster killing was one of the final acts of tidying up the cosmos,
but at the same time, by making Typhon a problem for humans (like
Python), the poet further shifts Typhon out of an early stage of cosmic
history and into a later stage when not only had the universe and the family
of gods settled into their final forms (indeed, at least two of the younger
gods, Hephaestus and Athena, were born before Typhon was created,
according to the Hymn, which contradicts Hesiod’s timeline) but the human
race had also appeared on the horizon. Neither Typhon nor Python is
presented as a cosmogonic threat, in other words, but both, instead, are
emphatically monsters who fall into line with those I mentioned earlier,
who were killed by human heroes in order to protect human interests—the
Nemean Lion, whom Hesiod called a “bane (pēma) to humans”; the Sphinx,
whom Hesiod called the “deadly destruction of the Cadmeians” and whom
Aeschylus described as eating humans raw; the Chimaera, whom Homer

called an “evil to many people”; Scylla, whom Homer called a “bane
(pēma) for mortals”; the Erymanthian Boar, who harmed the people of
Psophis; and so on. The tendency to explicitly connect monsters to the harm
they do to humans only grows stronger as time goes on, in fact—in
Euripides’s Heracles, even the Cerynitian Deer is described as a marauder
of country folk.154
Remarkably, not only does the poet of the Homeric Hymn recraft Typhon
into a monster who threatens humans and their flocks (and probably either
creates or recrafts Python into such a monster, too, although we have no
earlier version of this story with which to compare the Hymn), but he also
recrafts Apollo into someone who is similar to a human hero. Our first
indication of this comes immediately before the poet begins to tell us about
Apollo’s search for a place to build his Oracle. Briefly, the poet considers
treating another topic, instead: one of the many courtships of Apollo. But
which one? he asks himself. In listing his top choices, the poet makes one
thing curiously clear: in all of the cases, it was with a mortal man that
Apollo competed for a woman’s love.155
Having aired the theme of Apollo’s love life before us for eight lines, the
poet rejects it and decides instead to describe the foundation of the Delphic
Oracle. He begins with a lengthy list of the places that Apollo considered as
locations—Iolchus, Euboea famed for shipping, the Lelantine Plain, grassy
Teumessus, and so on, each of which is invoked with just one or two words.
But then he pauses for four entire lines on the future site of the city of
Thebes, which at the time of Apollo’s quest, the poet says, was “cloaked in
vegetation, for no mortal yet dwelt in holy Thebes, and there were not yet
any paths or roads crossing the wheat-bearing Theban plain, but it was
occupied by wild growth.”156 This is an interesting place for the poet to
linger. Thebes was reputed to be one of the oldest cities in Greece, and the
myth of its foundation was well known by the time that our Hymn was
composed. Cadmus, wandering through Greece in search of his kidnapped
sister, was told (by the Delphic Oracle, no less) to give up on finding her
and instead to follow a cow until it lay down and then to establish a city on
that spot. Having followed the cow, and having built the city of Thebes,
Cadmus asked some friends to fetch water with which he could perform a
sacrifice. When those friends approached a nearby spring, they were killed

by a gigantic snake that guarded it. Cadmus killed the snake, made his
sacrifice and became Thebes’s first king.157
The Delphic foundation myth that we get in the Hymn to Apollo mirrors
the Theban one; in each case, a giant snake, lurking at a nearby spring, must
be killed before a foundational act can be completed by an individual who
has journeyed from afar in order to perform it. If I am correct that the
Hymn’s extended meditation on the site where Thebes will soon be
established would have evoked that city’s well-known foundation myth,
then the poet of the Hymn is prompting us to have Cadmus in our minds
when listening to the tale of Apollo’s monster killing that he is soon to
narrate. Again, Apollo’s cohort—the individuals to whom we are invited to
compare him—is human.158
Once at the site of Delphi, Apollo lays the foundations of his temple with
his own hands—unusually amongst the gods, he is a do-it-yourselfer. Upon
Apollo’s foundations, the famous (human) heroes Agamedes and
Trophonius lay a stone floor, and then “tribes of humans” erect the walls of
the temple, setting the blocks in place.159 This is a collective effort, in other
words, with the god working, if not necessarily alongside humans, then at
least on a common project. This story marks a contrast with what scholars
think was an equally early, or perhaps even earlier, story, according to
which the first temple in which the Oracle was housed had been built out of
laurel branches by Apollo alone, the second by birds and bees out of wax
and feathers, the third by the gods Athena and Hephaestus out of bronze
and the fourth, finally, by Agamedes, Trophonius, and other humans. In our
Hymn, the three “magical” or “divine” temples are omitted altogether, and
the god collaborates with humans on the first and only structure that houses
his Oracle. It also marks a contrast with what was the canonical myth of
Apollo’s arrival at Delphi (a myth embraced by the personnel of the Oracle
itself), according to which he took over its functions, peaceably or not, from
earlier, female owners.160 In short, the poet of the Hymn made a very
conscious decision to present Apollo as he did.
Finally, Apollo turns his mind to staffing his new Oracle. Spotting a ship
full of Cretan sailors out at sea, he pulls it into Delphi’s harbor. Personally
leading the sailors up the hill to the Oracle, he shows them around,
describing all its features. Subsequently, he explains to them that they will
be able to feed themselves and their families—thus satisfying a very mortal

need—from the sacrifices of animals that worshippers will bring. Here, as
in his eagerness to help build that Oracle, Apollo is a god who is notably
willing to rub elbows with humans—not in disguise, as did the grief-
stricken Demeter in Eleusis, nor in disguise, under duress and with shame,
as did Aphrodite on a Trojan hillside in her Homeric Hymn, but of his own
free choice and more or less in his own form.
The Homeric Hymn to Apollo turns Typhon, a great cosmogonic monster,
into a threat to humans rather than gods and shows us one of the younger
gods, Apollo, filling the role that a human hero would fill when he defeats
Python, another monster that is markedly presented as a threat to humans.
Apollo himself is made out to be a very human god, similar to a hero such
as Cadmus. In three early Archaic compositions, then (the Hymn to Apollo,
Hesiod’s brief but pointed interjections about monster-killing heroes in the
Theogony and Hesiod’s mention of Heracles’s participation in the gods’
battle against the gigantes), we encounter subtle but clear transitions away
from the pattern of divine, cosmogonic monster killing that we find in Near
Eastern myths and in Hesiod’s own story of Zeus killing Typhon, towards
monster killing that brings human heroes, and human concerns, to the
forefront. The introduction of heroes into what had previously been
exclusively divine territory suggests that the Greek hero as we later know
him was still a relatively new phenomenon at the time that those
compositions were created and that storytellers had a free hand in how they
developed him.
Monsters and Heroes

Hesiod’s scattered catalogue of heroic monster killing is just the tip of what
is already a large iceberg by the end of the Archaic period, at latest. To give
only a partial list, Heracles killed the Nemean Lion, the Lernaean Hydra,
the Stymphalian Birds, Geryon, Geryon’s dog Orthrus, and a Trojan sea
monster, and he overcame (without killing them) the Erythmanthian Boar,
the Cerynitian Deer, the Cretan Bull, the Mares of Diomedes, and Cerberus.
Theseus conquered the Crommyonian Sow, the Bull of Marathon (who was
formerly known as the Cretan Bull), and the Minotaur. Bellerophon killed
the Chimaera. Perseus decapitated Medusa and killed a sea monster off the
coast of Joppa. Oedipus drove the Sphinx to her death. Amphitryon killed
the Teumessian Vixen—or some say that Oedipus killed her. Cadmus and

Jason killed dragons. Odysseus outsmarted a Cyclops, evaded the Sirens,
and at least stood his ground against Scylla.
These were exciting stories, even if some authors chose to narrate them
rather tersely. But who, exactly, were these monsters and how did their
identities add to the excitement? As I also discussed in Chapter 4, Greek
monsters were well woven into the network of relationships not only
because they were the victims of heroes (and occasionally of gods) but also
because many monsters were the children of gods. In Chapter 6, I discussed
how vividly, and sometimes even tenderly, the births and childhoods of
these monsters could be narrated: dreadful though they might seem to the
mortals whose lives and herds they threatened, many of these monsters
were the objects of a god’s affection, or at least the product of a god’s loins.
This lends important nuances to what became the Greek pattern of heroes
killing monsters within situations that were set well after the universe has
reached its finished form. If Hesiod’s discussion of the race of Phorcys and
Ceto proleptically implied that the heroes were destined to clear up the bits
of monster killing that Zeus had left unfinished after his victory over
Typhon, then the heroes could be understood as nearly the equals of the
gods in this one particular talent, at least. But we could also understand
Hesiod’s treatment—and particularly his choice of narrating the monsters’
deaths at the hands of the heroes immediately after his description of the
monsters’ ancestries and births—to imply that, by killing the monsters, the
heroes intruded upon the divine realm. However dim and elemental Phorcys
and Ceto may seem in contrast to the anthropomorphic Olympians, they are
gods, and their children are gods just as fully as are, for instance, the river
children of the equally dim and elemental Tethys and Oceanus, whom
Hesiod lists immediately after he finishes listing the descendants of Phorcys
and Ceto. Some of Hesiod’s monsters, moreover, are said to be lovely in
ways that evoke the loveliness of anthropomorphic goddesses or human
women. The Graeae are described as “having beautiful cheeks”; one of
them, Pemphredo, is also called “fair-robed” and her sister Enyo has “robes
dyed with saffron.” Ceto, who is the progenitor of all these monsters, is
described as having beautiful cheeks, as well, and even Echidna, who eats
raw flesh in a cave beneath the earth, is a “quick-eyed nymph, with
beautiful cheeks” (or at least her upper half is—her bottom half is a terrible
snake).161 Some of these monsters “mingle in love” with one another just as
other gods do. Indeed, one of them mingles in love with a god who has

nothing monstrous about him at all: Medusa is bedded by Poseidon in a soft
meadow amongst spring flowers. There is no indication here that Medusa is
anything but a snaky-haired Gorgon when Poseidon makes love to her;
Hesiod’s insertion of the scene in the middle of his discussion of the
Gorgons and his general discussion of monsters underscores this.162
We could choose to explain away these things as confused “holdovers”
from a time when there were alternative traditions about such creatures
(was Medusa always ugly?)163 or as attempts to raise a frisson (let’s imagine
what it would be like to have sex with a snaky-haired woman!) or simply as
a formulaically poetic way of denoting sexual intercourse and formulaically
poetic ways of describing important females, for example, but whatever the
origins of these traditions, what the audiences heard when the Theogony
was performed was that the monsters whom the heroes killed were divine
and not altogether dissimilar from other divinities.
Greek myths nuance heroic battles with monsters in another way, too.
Most heroes receive divine help in conquering monsters. Athena is their
most constant companion, followed by Hermes. Already in the Odyssey,
Heracles reports that Athena and Hermes helped him fetch Cerberus from
Hades, as shown in Figure 7.3, for example. Athena is often Heracles’s
helper in other adventures, as well, and Helios lends Heracles the cup in
which he crosses the ocean each day in order that Heracles might get to
where he needs to be (see several of the metopes shown in Figure 7.4 and
the scene in Figure 4.1).164 In the Hesiodic Shield, Perseus is described as
wearing the helmet of invisibility that Hades had lent him to combat
Medusa. Perseus received two other important divine gifts that would help
him in his quest, as well—Hermes’s winged sandals and a special bag in
which to hold Medusa’s decapitated head—and Athena and Hermes are
frequently recorded as helping Perseus in other ways.165 (Figure 7.1 shows
Perseus wearing the sandals while Hermes stands nearby, notably
barefooted, and Athena carefully holding Medusa’s head out of Perseus’s
direct line of sight.) Bellerophon received a bridle with which he could
capture Pegasus, as well as advice about how to do it, from Athena—
although in a fragment of Hesiod’s Catalogue, it seems to have been
Poseidon, Pegasus’s father, who gave the bridle to Bellerophon, who was
also his son (accentuating the odd situation of one brother riding another
into battle against the Chimaera).166 Theseus was helped by his father’s

wife, Amphitrite, when Minos sent him to the bottom of the sea to retrieve a
ring, according to Bacchylides’s seventeenth ode. A crown that was given
to him by either Amphitrite or Ariadne, the latter of whom perhaps had
received it from Dionysus, lit Theseus’s way through the Labyrinth. We see
Theseus’s underwater visit to Amphitrite in Figure 7.8, with Athena
standing by (probably, she is there because Theseus is the ancestral hero of
Athens, her special city). Sometimes Athena also shows up next to Theseus
in artistic representations of his victory over the Minotaur and other
monsters—or over humans who behave monstrously, as on the vase shown
in Figure 7.5.167 Already in the Odyssey, Jason is described as the special
favorite of Hera,168 and he is often portrayed as being helped by Aphrodite
and Athena. On the column krater shown in Figure 7.9, Athena expresses
delight, or perhaps even surprise, as Jason takes the Fleece away from the
serpent who guards it.169 We get an alternative story on an Athenian kylix
from about 480 BCE, attributed to Douris, which shows Athena standing by
while the dragon guarding the Fleece apparently disgorges Jason after
swallowing him.170 Are we to understand Athena as having caused the
dragon to do this, and thus as rescuing a hero who had been swallowed
alive?

7.8 Theseus visiting his stepmother, Amphitrite, under the sea; Athena mediates the
encounter. A small triton supports Theseus while dolphins swim around him. Interior of
an Attic red-figure cup, signed by Euphronios (the potter) and attributed to Onesimos
(the painter). Dated to between 500 and 490 BCE (detail). Now in the Musée de Louvre
inv. num. G104. Photo credit: Stéphane Maréchalle / Musée de Louvre / © RMN–Grand
Palais / Art Resource NY.

7.9 Jason snatches the Golden Fleece away from the dragon that guards it, while
Athena and an unidentified man look on. The prow of the Argo, with its speaking
figurehead, is to the far right. Attic red-figure column krater by the Orchard Painter,
dating to between 470 and 460 BCE (detail). Now in the Metropolitan Museum of Art,
New York. Purchased through the Harris Brisbane Dick Fund, inv. num. 1944.11.7.
Photo credit: © The Metropolitan Museum of Art / Art Resource, NY.
Even this partial list of the divine help given to Greek heroes who set out
to conquer monsters establishes it as a standard theme. Such alliances
confirm that the Greek hero is special, of course—that he is a sort of junior
partner of the gods. It is the hero, however, who does most of the grunt
work: the physical fighting against the monster when necessary, the
slogging across barren countrysides to find the monster in the first place,
the plowing, the digging of ditches and other mundane labors171 that enable
him to complete the task. Typically, what the god contributes to these
partnerships is a new technology or strategic insight, while staying away
from anything that raises a sweat—the castanets or rattles that Athena gives
to Heracles to ward off the Stymphalian Birds, the winged sandals given to
Perseus and the magic bridle that Bellerophon receives from either Athena
or Poseidon, for example.172 The basic identity of the hero as human, then,
is emphasized even as he is being set apart from other humans; he is

elevated by the fact that he receives divine help to complete the task, but he
does the heavy lifting. An amusing variation that underscores this point is
the portrayal of Heracles holding up the sky on one of the metopes from the
early fifth-century Temple of Zeus at Olympia: Heracles is bent over with
strain, while in back of him, out of his sight, his sister Athena is casually
carrying the bulk of the weight on one upraised hand (see metope 10 in
Figure 7.4). We could choose to understand these incidents of divine help as
merely metaphorical—just as some scholars choose to understand the
Homeric formulation of a god breathing strength or courage into a warrior
as metaphorical. But even if poetic statements of the gods’ help were meant
metaphorically (which I doubt), the expression of this idea through the
vivid presence of gods, especially in artistic representations, insistently
emphasizes this element in the hero’s story.
We once again see the Greek hero, then, being pulled in different
directions. Monster killing, by the very nature of its victims, required
heroes to intrude upon the gods’ world, but by the very nature of what was
required to kill a monster and to accomplish some of their other labors, the
heroes were once more confirmed as human beings who depended upon the
gods. The fact that the heroes killed monsters on behalf of other mortals,
whose lives and livelihoods the monsters threatened, emphasizes the human
orientation of the heroes’ deeds, too. Monsters and monster killing, then,
mark a point of uneasy balance between heroes and gods, a juncture
between human and divine through which heroes must negotiate a passage.
As we meet them in Greek myths, moreover, monsters are representatives
not of a general chaos that had once, long ago, been inflicted on a nascent
cosmos, but rather of the chaos that gods might always inflict upon mortals
and against which mortals had to be continuously on guard. In the daily
world of historical Greece, people did not expect to encounter new Hydras
or Chimaeras or Gorgons, but they did fear the ever-present demonic forces
that might bring illness, sterility, famine, and other evils. Many of these
forces were personified into characters: Ephialtes (The Guy Who Leaps on
You), who brought nightmares, for instance, and Mormo (The Frightening
Woman), a bogey who attacked children. A few of them even became
woven into the network of gods and heroes. Lamia, for example, whose
name means something such as “The Gobbler,” was both the generic term
for a demon who killed infants and pregnant women and the name of a
woman whom, according to myths, Zeus had once seduced and Hera had

subsequently persecuted, driving Lamia into a maddened state in which she
killed her children.173 According to Stesichorus, Lamia’s father was
Poseidon and Scylla was her daughter.174 In other words, Lamia became
well entrenched within the mythic web of relationships. All of these
demonic forces, whether or not they received formal narrative development,
were combatted not only by gods to whom one might pray and pay cult, as
in other ancient Mediterranean countries (thus, for instance, to combat
Lamia and other demons who thwarted successful motherhood, one might
pray to Artemis or Hera), but also, and most pronouncedly by Heracles, the
greatest heroic conqueror of monsters in myths and the greatest protection
from evil in daily life. His common cult title, “Alexikakos” (Averter of
Evils), attests to this; the declaration that “Heracles Alexikakos dwells
here!” was engraved above many a house door throughout Greek-speaking
parts of the Mediterranean. Amulets against illness that survive from later
antiquity frequently feature Heracles’s image. He is invoked for help in a
fragment of the Getty Hexameters that I discussed in Chapter 3 in a context
that equates his victory over the Hydra with his ability to ward off all of the
ills that Hexameters promise to avert.175
The End of Monsters—and of Heroes

It seems that by the time that the Trojan War took place, there were no more
monsters left to kill, other than those that Odysseus claimed to have
encountered while traveling through the outer limits of the world. The other
monsters mentioned in the Iliad and the Odyssey had been dispatched by
earlier generations of heroes by the time that the action of those poems took
place: the Calydonian Boar had been conquered by Meleager and his
friends, who predated the heroes of the Iliad by a generation,176 and the
Chimaera had been killed by Bellerophon, who predated them by two
generations.177 When the shade of Heracles met Odysseus in the
Underworld, his labors were long past; his quest to bring Cerberus to the
upper world was just a memory.178 By the time of the war, those who would
be called heroes had nothing left to kill except one another, and the war
provided excellent opportunities for them to display their skills in doing so.
Which brings me to the final topic in this chapter. According to a story
that we can put together with bits of information from the scholia to the
Iliad and Proclus’s summary of the lost epic Cypria, there came a time

when Gaea felt burdened by the flourishing human population. She
demanded a solution, and Zeus and Themis decided to goad humans (or,
more specifically, the heroes) into killing each other off. The result was the
Theban War. Yet the resulting drop in population proved to be insufficient,
and so Gaea complained again. Zeus was on the verge of using thunderbolts
and floods to thin out the ranks of humanity, but the god Momus (Blame)
suggested that he try another war, which, Momus explained, could easily be
instigated if Zeus did two things: sire a beautiful daughter upon Leda and
give the goddess Thetis to a mortal in marriage.179 Thus began the Trojan
War, which, according to a fragment of Hesiod’s Catalogue, enabled Zeus
not only to decrease the population of humans in general but more
specifically to destroy the race of hemitheoi—the “half gods” or heroes.180
Walter Burkert may be right that the kernel of this story was borrowed
from a line in the Erra and Ishum, a Mesopotamian tale that dates back at
least to the eighth century BCE. In it, Erra, the god of war, is urged by his
seven weapons to begin a new conflict. Erra subsequently claims that his
reason for instigating the war was that humans no longer respected him, but
the weapons themselves provide two other reasons: first, that they
themselves had become rusty with disuse, and second, that humans had
become too noisy for the Anunnaki (a group of gods associated with the
earth) to endure—a reason that approximates Gaea’s complaint.181
Mesopotamian influence of some kind does seem to lurk behind the
Greek idea of gods making a decision to destroy a significant portion of
humanity, for we have not only the weapons’ speech from the Erra and
Ishum but also several versions of a Mesopotamian story whereby the older
gods complain that there are too many humans and plot to wipe them out—
first by famine; then by drought and famine; then by plague; then by
plague, drought, and famine all at the same time; and then finally by a
flood. Each time, however, the god Ea (Enki) saves humanity by warning a
human confederate (variously called Atrahasis or Utnapishtim or Ziusudra)
and providing him with a plan to survive (much as Noah receives such a
plan from Yahweh in the book of Genesis). Therefore, the human race is
able to regenerate itself each time that the gods try to destroy it.182 Scholars
presume that these Mesopotamian stories were known by Greek poets and
that they helped to inspire the Greek version of the Flood story. Thus, even
if Burkert is right that the Erra and Ishum specifically spurred the Greek

poets to craft a story whereby the human race was diminished by war, the
poets are likely to have done so while having other choices in front of them
—other ways by which gods could thin the ranks of humanity and bring the
age of heroes to an end. We should look more closely, therefore, at the
differences between the tale of Erra and the tale of Troy and what they may
tell us about the Greek hero.
The Erra and Ishum focuses closely on Erra’s decision to go to war, the
attempts of his fellow god Ishum to moderate Erra’s ferocity, and the
capitulation to Erra’s decision by Marduk, the king of the gods, who must
ceremoniously leave his throne before the chaos of war can be unleashed.
Of the war itself and its effect on humans, we hear almost nothing in this
poem. In contrast, the Iliad emphasizes the individual talents and quirks of
the warriors, their personal involvements with one another, their moments
of glory on the battlefield, their losses, and their homecomings, successful
or not. The same can be said of many of the other epics narrating the Trojan
War, which we now possess only in fragments. Many of these works, as I
commented earlier, gave the poets opportunities to focus on the human
condition more generally—the inevitability of death, the personal choices
that could ruin or save one’s companions, the physical and emotional strains
of war and its effects on the families of those who fought. The first lines of
the Iliad and the Odyssey epitomize this:
Sing, Goddess, Achilles’ rage,

Black and murderous, that cost the Greeks
Incalculable pain, pitched countless souls
Of heroes into Hades’ dark,
And left their bodies to rot as feasts
For dogs and birds, as Zeus’ will was done.
Begin with the clash between Agamemnon—
The Greek warlord—and godlike Achilles.
Speak, Memory, of the cunning hero,

The wanderer, blown off course time and again
After he plundered Troy’s sacred heights.
Speak of all the cities he saw, the minds he grasped.
The suffering deep in his heart at sea
As he struggled to survive and bring his men home.

But he could not save them, hard as he tried—
The fools—destroyed by their own recklessness
When they ate the oxen of Hyperion the Sun,
And that god snuffed out their day of return.183
In each case, there are gods at work—what happened in the Iliad unfolds
according to the will of Zeus, and in the Odyssey, Hyperion, and also
Poseidon, wreak a terrible vengeance upon humans who offend them.
Nonetheless, it is Achilles (and to a lesser degree, Agamemnon) who is
responsible for the pains and deaths of the Greeks, it is human bodies that
rot and are consumed by scavengers, and it is Odysseus who, in a very
human way that no god ever shared, has to learn to understand how the
minds of other people work and use that knowledge to save himself.
In choosing to end the age of heroes with a war, then, the poets who were
responsible for the evolution of this story once again underscored the
centrality of the hero—the remarkable human—within the mythic network
that they were creating. There are plenty of gods on the Trojan stage as
well, but their actions are significant primarily apropos the humans. As
deviously enchanting as Hera’s deception of Zeus may be in Iliad 14—and
as much as it “shows divinity in a naturalistic, cosmic setting which is not
otherwise a feature of Homeric anthropomorphism,” as Burkert suggests184
—the scene makes sense within the poem as a whole only insofar as Zeus’s
postcoital slumber provides an opportunity for Ajax to knock Hector to the
ground with a huge rock, which enables the Greeks to trounce the Trojans.
Hera’s machinations, divinely charming though they may be, underscore the
fascination that a human conflict could exert over the gods. However much
those gods pursued their selfish aims in Greek myths, the myths, which
were narrated by humans for human audiences, returned time after time to
the ramifications that those divine pursuits had on the humans of earlier
generations—the heroes.

Epilogue
GREEK MYTHS—precisely as they were created by the ancient poetic

narratives that we still admire today for their beauty and force—were
central elements in the formation of the Greeks’ belief in their gods and
heroes. Understanding this should alter not only the way in which we study
the myths themselves but also the way in which we reconstruct and
understand Greek religion more generally. The study of rituals (that is, what
one “does” in religion, as opposed to what one “thinks” or “feels”) should
no longer dominate the field as it did in the twentieth century; we must
begin to give appropriate weight to the important work that the myths
themselves did and strive to understand it better.
To do so, we need to approach these myths not only with the tools of
philology, literary criticism, and cultural studies that we have always used,
but also with the tools that narratologists, folklorists, sociologists,
psychologists, and cognitive scientists have developed to understand how
modern narratives—as found in novels, films, television shows, and other
cultural forms—affect audience members’ cognition, emotions, ethics, and
beliefs. Recently, the ability to tell stories has itself been identified as one of
humanity’s most important evolutionary advantages: stories enable us to
imagine situations that we have not experienced and to play through the
ways in which we might react if confronted by those situations ourselves
one day. As such, stories are a particularly important means of creating and
sustaining religious beliefs—beliefs in entities that we cannot access
through ordinary means. The Greek gods and heroes are excellent figures
against which to test this idea because we possess a rich trove of narratives
—of myths, that is—about them. Most of these myths are about what
happened when gods and mortals interacted. As such, they served not only
to expand and intensify ideas of who the gods and heroes were but also to
limn the possible effects that the divine world could have upon the mortal
world. Myths did this more memorably, and therefore more effectively, than
any set of doctrines or dogmas ever could.

In borrowing tools from scholars of modern narratives to better understand
how Greek myths did these things, I selected those that engage with aspects
of human cognition and emotion that are unlikely to vary from time to time
or place to place. Thus, for example, the tendency to continue pondering an
intriguing experience after it is over would seem to be intrinsic to humans
(and, indeed, perhaps also to other higher primates). Therefore, I assumed
that what narratologists had concluded about the effects of modern episodic
narration—that it causes an audience to think about the story’s characters in
between episodes and to bond with them more deeply than they would if the
story had been narrated in one fell swoop—would also be true for episodic
narration in antiquity.
I also chose to use, with one exception, methodologies that focus
primarily on those cultural forms that Dorothy Noyes, in her model as I
presented it in Chapter 1, includes in her first category: forms that are
deliberately sought out by, and retain the focused attention of, their
audiences. Noyes characterizes these forms as having maximal internal
coherence, maximal salience, and maximal indexical connections to their
audience, and as being produced with the assumption that they will be
evaluated for their potency of effect and skillful execution. Given that the
Greek myths that come down to us are the highly polished products of
professional poets, which were intended to be performed by experienced
choruses or actors, they match these criteria closely. Therefore I assumed,
as I looked for methodologies to borrow, that by and large, what makes
first-category forms successful in our own time and place would have
worked in Greece, as well, mutatis mutandis. My one exception was the use
I made of what folklorists have discovered about the techniques that people
use when telling memorates, a narrative form that I suspect belongs in
Noyes’s fourth category (that is, unsought experiences that nonetheless
demand focused attention). As I briefly showed, however, successful
authors of modern literary tales about the supernatural (which belong in
Noyes’s first category) often choose to adopt the narrative techniques used
in memorates. Therefore, it did not seem to me to be a stretch to look for
those techniques within ancient literary tales about the gods and heroes, as
well.
Other than the two guidelines I have just sketched, I set no limits for
myself as I borrowed my methodological tools. For the past ten years or so I
have read widely in fields that seemed likely to produce results, picked out

approaches that seemed useful and then tested them against ancient
materials. I clove to no particular school of theory, but was sometimes
surprised and delighted to discover that different scholars whose work I
liked—scholars who were, sometimes, from very different fields—had a
tendency to know and admire one another. This gave me confidence that the
cumulative approach I was slowly building was not just the product of my
own perceptions.
As I step back now from my work, I know that I leave some issues open.
One is the question of to what degree it makes a difference for a narrator to
be aware of the cognitive and emotional effects that his or her narrative will
have. Or in other words: do the techniques that I have discussed in this book
work better when there is a driver at the wheel?
I suspect that the answer to this depends on the technique. Joshua
Landy’s formative fictions, which I discussed in Chapter 3, would work
even if an individual narrator weren’t conscious that he or she was wielding
this tool (long-standing modes of narration that had been noticed to have
desirable effects could be adopted by new narrators who knew that they
were successful but did not know, or care, why). Yet, as in Landy’s case of
Jean Eugène Robert-Houdin, for instance, a narrator who aims for a specific
effect and knows that certain techniques will achieve it is likelier to have a
higher degree of success. Tanya Luhrmann’s Vineyard Christians are
another excellent example of this. That said, however, narrators do not
always purposefully aim for all the effects they end up achieving. William
Peter Blatty did not set out to make more Americans believe in demons
when he wrote The Exorcist, yet the book and the movie that followed it
accomplished just that, at least temporarily.
Greek narrators of the archaic or classical period did not set out to induce
belief. Their job was to tell engaging stories, which often carried
institutional or personal agendas, as well. Sometimes, the agenda was to
attract people to a particular cult of a particular god or hero—but the poets
were not evangelizing the very existence of those gods and heroes. If they
consciously wielded the tools that I have described as being conducive to
belief, then, they did so primarily with other goals in mind.
Other phenomena that I discussed, such as parasocial relationships
(Chapter 3) and plurimediality and accretive characters (Chapter 5), are by
their very nature impossible to wield deliberately; they develop under
circumstances that cannot be controlled by a single individual or group of

individuals. In still other cases, such as that of constructing a story world
(Chapter 4), I have argued that there is a difference between modern
narrators and ancient narrators. Modern narrators who set out to create a
story world typically do so from scratch, and therefore must make
deliberate, calculated choices; ancient narrators, in contrast, reacted to and
helped to develop a story world that was already well established by the
time it began to leave a record for us to consider. In either case, certain
effects of a well-wrought story world will work (for instance, crossovers
and hyperseriality), but the perceived connection between that story world
and the real, “historical” world will be different.
The other major issue that I leave open is that of specific application.
Each time that I introduced a technique or phenomenon, I gave enough
examples of how it worked in ancient narratives, I hope, to persuade my
readers that I was on to something, but I have by not exhausted the possible
texts against which my proposals can be tested and weighed. I look forward
to seeing other scholars try their hands.
In Chapters 6 and 7, I explored two subjects that I suggest are
distinctively characteristic of Greek myths, metamorphosis and heroes, and
used them as opportunities to extend some of the methodological arguments
I made earlier in the book. In Chapter 6, I suggested that the primary
purpose of narrating stories about marvelous creatures and events was to
sustain belief in the gods and heroes. I also noted that ancient Greek
narratives of the marvelous, like those of some other cultures, including our
own, provide an opportunity to articulate the negative side of divinity,
which is avoided in more formal liturgical contexts such as prayers. In
Chapter 7, I returned to the topic of episodic narration that I introduced in
Chapter 3 and proposed that myths about heroes partook of two different
kinds of episodic narration, the series and the serial. This had the effect of
sharpening the inherent contradiction of the Greek hero. Like a god, the
hero could be suspended in time and his adventures could be multiplied
infinitely, unfolding like the separate episodes of a series. But like other
mortals, the hero lived a life that was known to have a beginning and an
inevitable end, as in a serial. This hero was peculiarly Greek in this and
other ways that I discussed that set his myths (and, correspondingly, his
cult) apart from almost anything we find in the Near Eastern cultures whose
myths otherwise influenced those of Greece so significantly.

Whatever else I have accomplished in this book, I hope that I have
stimulated further conversations about how religious beliefs not only in
ancient Greece but also in other cultures are created and sustained by the
consumption of engaging narratives. Until now, the two groups of people
who have the best potential to shed light on this topic—the social scientists
and psychologists who have studied the construction of belief, on the one
hand, and the humanists who have studied the particular ways in which
narratives affect audiences, on the other hand—have interacted with one
another very little. There has been almost no interaction, moreover, between
that second group and the scholars who study religions. Here I have tried to
integrate ideas from each of these fields. Whether or not my readers agree
with my particular conclusions, I hope that I have motivated them to try to
do the same.

Notes
1 THE STORY OF MYTH

1. Burkert 1979: 23.
2. Csapo 2005: 9. Csapo draws on Bascom 1965.
3. Lincoln 1999: xii.
4. Mills 1990: 28–29.
5. I take some of my observations about Sleeping Beauty from Tatar 2002: 95–104.
6. The rape interpretation that I offer here draws on comments made by Maleficent star
Angelina Jolie, as quoted in Holmes 2014. The quotation is taken from Shapiro 2014 (and
was widely reposted on the web).
7. It is only in Walt Disney’s 1959 animated movie version (Sleeping Beauty) and the 2014
reworking of it (Maleficent) that all of the characters are given personal names. Some earlier
versions bestowed a personal name on the main character—e.g., Giambattista Basile’s Sole,
lune e Talia (1934), in which she is named Talia.
8. Od. 12.69–72.
9. I borrow the term from social and cultural anthropologists, who use it as a concise way
of referring to a variety of nonhuman entities whose presence is not always detectable.
10. As stated, for instance, in a conversation between George Lucas and Bill Moyers
(Moyers & Company 1999).
11. Od. 4.214; Od. 19.502.
12. Pl. R. 377a3–d6.
13. For more on this large topic, see Morgan 2000; Buxton 1999b; P. Murray 1999; Rowe
1999.
14. Most famously, probably, Pi. O. 1.28–29.
15. Further, see Fowler 2016; Calame 2015: 23–62; Csapo 2005: 10–30; Lincoln 1999: 51–
75; Most 1999; Graf 1993: 9–34.
16. Claude Calame has several times fought the battle against looking for an absolute
description of myth: most recently, Calame 2015: 17–19 and 403–414, but also, for example,
the “Avant-Propos II” of the second edition of his Mythe et histoire dans l’Antiquité grecque
(Calame 2011b). See also the review of recent books on Greek myths by Philippe Matthey
(2016), which summarizes the state of the question.
17. See Graf and Johnston 2013: 66–93; Johnston 1999: 161–199.
18. I have chosen to limit myself to these periods because the ways in which myths were
narrated changed significantly after approximately 300 BCE. Most importantly for my
purposes, there was far less creative reshaping of myths by authors, as N. J. Lowe has
observed (2000: 97–99).
19. Burkert 1979: 23–24.
20. Hansen 2017 takes as its quest the side-by-side assembling of a great many stories from
a wide variety of these genres, which are divided up and discussed according to their topics
(for example, ‘ “Kings and Princesses”). Precisely as such, the book demonstrates my point
nicely. See also Buxton 1994, esp. 40–44.

21. E.g., Pl. R. 2.381d1–e6 and Lg. 10.887c5–e1.
22. Od. 23.301–341. Odysseus, not surprisingly, seems to have done the lion’s share of the
talking, although Atwood 2005 begins to make up for this.
23. I say “almost” because recent observations suggest that some of the higher primates
can engage in sustained periods of what we would call make-believe, carrying around a log as
if it were a baby, for example (Jolly 1999: 290–292).
24. In my formulation here, I rely particularly on Harari 2015: 31–37; Sjöblom 2011; and
Boyd 2009, esp. 1–16 and 191–196.
25. Cf. O’Flaherty 1988: 27–28.
26. D.S. 5.4.7.
27. Noyes 2016: 127–178.
28. Noyes 2016: 141–148 for the basic description of the categories; for the quotation, 141.
29. I borrow the image from the literary critic Alessandro Portelli (1997).
30. I am thinking here of the fact that King Demetrius of Macedon had a mistress named
Lamia—a name she shared with a monster who was reputed to steal and eat infants and to
seduce men, and sometimes to eat them, too (on the monster, see Johnston 1999: 173–179).
The relationship was parodied by Alciphron in his “Letter from Lamia to Demetrius” (4.16)
and mentioned by several other ancient authors, e.g., Plut. Dem. 27; Ath. 101e, 128b, 253a
and (most interestingly) 577d–f.
31. Noyes 2016: 144.
32. The only extant example is A. Pers., but we know that Phrynichus wrote a Sack of
Miletus (notably, it met with public disapproval). Phrynichus also wrote a Phoenician Women,
set at the court of Xerxes. There was probably a tragedy treating the near death of Croesus on
the pyre, and there are fragments of a few more historical plays, as well. See Edith Hall
(1996: 7–10), who rightly points out that although the Greeks distinguished between what we
would call a “mythic” age and a “historical age,” they felt that certain events (such as the
battle of Salamis) were of a highly significant nature that raised their importance near to those
of the mythic age. Events connected with the Persian Wars, she notes, were particularly likely
to be included in that category.
33. Od. 19.570–575; A. Ag. 611–612.
34. Pi. O. 1.23–89 and P. 9. Cf. Medea’s appearance as a surprisingly Pythia-like figure in
P. 4, and see O’Higgins 1997.
35. Bacch. 17. There is debate about how much of this story was new at the time that
Bacchylides narrated it. We have no earlier textual evidence for Theseus jumping into the sea.
A vase painting showing Theseus amongst undersea divinities (the Onesimos cup—see
Figure 7.8 and further at Gantz 1993: 263–264) dates to between 490 and 500 BCE, which
has encouraged some scholars (e.g., Maehler 1997: 170) to date the ode to the early 490s and
make the entire leap into the sea Bacchylides’s invention, on the assumption that artists are
likelier to illustrate an existing story than to create a new one. Other scholars argue, on the
basis of possible political allusions in the poem, that Bacchylides’s ode dates to the 470s (e.g.,
Fearn 2007: 242–244), which would make Onesimos’s painting, and probably also some
now-lost textual narration of the story, earlier than Bacchylides’s treatment. Even when
accepting the later date, however, most scholars agree that the detail of Theseus retrieving
Minos’s ring was Bacchylides’s own invention. See Pavlou 2012 for a recent discussion, with
summary and references to previous discussions.

36. Cf. N. J. Lowe (2000: x–xi), who notes that the “Classical” plot (i.e., that endorsed by
Aristotle, as found in epic and tragedy) became the “classical plot” (i.e., the one still used
frequently in storytelling today, even if, at times, through media that Aristotle never dreamt
of).
37. Hufford 1982. Or as the classicist and folklorist Andrew Lang put it nearly a century
earlier than Hufford, modern people have been habituated to hastily offer “a very mixed
theory in which rats, indigestion, dreams, and of late, hypnotism, are mingled much at
random,” in order to dismiss those who claim they have experienced something outside of the
ordinary (Lang 1894; the quotation is taken from chapter 6, “Cock Lane and Common-
Sense”).
38. Calame 2015: 493–503.
39. Schaeffer 1999: 327–335. See also an interview with Schaeffer about his book by
Alexandre Prstojevic (2017) in Vox-Poetica.
40. “Hogwarts Extreme puts you in the world of Harry Potter. From the Common Room to
Diagon Alley: interact and role-play. Win the Quidditch World Cup, engage in the wizard
economy, take classes, and more! Your Hogwarts letter has arrived. Enroll now. The Sorting
Hat awaits” (HEXRPG, LLC 2017).
41. Let us not forget, either, that even historians used such techniques: Thucydides (1.22)
tells us that telling a true story sometimes requires the narrator to provide details and dialogue
that would otherwise be missing, if the story is to be convincing.
42. Landy 2012.
43. Luhrmann 2012.
44. E.g., Giles 2010.
45. E.g., Saler 2012.
46. Tolkien 1947.
47. Bettini 2013.
48. Scheid and Svenbro 1996.
2 RITUAL’S HANDMAID
1. E.g., Harrison (1924) 1963: ix, as cited by Csapo 2005: 157. See also Harrison (1912)
1927, e.g., 16, 327.
2. He was also a minister of the Free Church of Scotland, which had once defrocked him
on charges of heresy: Beidelman and Segal 2005.
3. Later published as Lectures on the Religion of the Semites (1889).
4. W. Smith 1889: 18, as quoted in Segal 1998: 2.
5. First published in two volumes in 1890, then in three in 1900, and finally in twelve in
1911–1915.
6. Frazer 1921: appendix 13 (quotation from 404); cf. Csapo 2005: 57–67. Interestingly, in
spite of Frazer’s declared intention to simply provide parallels to demonstrate the diffusion of
this myth, he seems to have taken great relish in narrating each of the stories in vivid detail.
Something deep in his soul seems to have cried out for a well-told tale.
7. See R. Parker 2007 for an excellent analysis of Murray’s work on this topic and its
entanglements with that of Harrison. I take the quotations from Parker (95), who cites Murray
1925: 72–78.
8. On the glamour of Harrison, see Beard 2002, esp. 37–84.

9. This was first used as a term, as far as I can tell, at Harrison (1912) 1927: 225.
10. Harrison (1912) 1927: 1–74; cf. Csapo 1995: 146–149.
11. Frazer 1901 (non vidi). Frazer is cited by Harrison (1912) 1927: 18.
12. Harrison (1912) 1927: 18.
13. E.g., Furley and Bremer 2001: 1.65–75; Perlman 1995.
14. Murray (1912) 1927.
15. Harrison and Verrall 1890: iii; cf. page xxxiii.
16. Harrison (1912) 1927: 16, 330; cf. Harrison 1921: 27.
17. Harrison (1912) 1927: 16 (shifting, manifold); Harrison and Verrall 1890: xxxiii
(unsatisfactory, absurd).
18. On Nietzsche and Harrison, see further Csapo 1995: 145–146; Peacock 1991: 170–171.
A crucial passage is Harrison (1912) 1927: viii.
19. The phrase is taken from the title of the fifth chapter of Harrison (1903) 1922: 163.
20. This is the general course taken in Harrison (1903) 1922 and to a great extent also in
Harrison (1912) 1927.
21. Harrison (1925) 1965: 344.
22. Harrison moved in the same social circle as the Stokers, sharing the stage with Mrs.
Stoker one evening for an amateur production of The Tale of Troy (Beard 2002: 39–40).
23. Ackerman 1991: 11–12, 15; and, on Murray, Lowe 2007.
24. Kripal 2010: 36–91; Ackerman 1991: 12.
25. Magliocco 2004: 40–43; Hutton 1999: 122–127.
26. Beard 2002: 135–136.
27. Harrison (1924) 1963: xviii.
28. Harrison (1924) 1963: 144. Or yet again: “[The Greeks] could not tolerate the Gorgon
form of the Earth-Mother. It was the mission of the Greek artist and the Greek poet to cleanse
religion from fear” (71).
29. The heyday of the approach lay between 1890, when the first edition of The Golden
Bough appeared, and 1912, when Harrison’s Themis appeared. In 1921, Harrison published
her short (forty-one pages) Epilegomena to the Study of Greek Religion, in which she touched
on the issue of myth and ritual again in passing, but she focused there primarily on applying
the comparative approach and the Durkheimian concept of collective group emotion to
interpreting ancient Greek religion. In 1913, Greek Divination: A Study of Its Methods and
Principles was published by W. R. Halliday—formally a student of Gilbert Murray and L. R.
Farnell at Oxford but closely mentored by Harrison as well, as he makes clear in the preface
to this book. Halliday embraced ideas that were dear to the Cambridge Ritualists’ hearts, most
notably that there was an inner core of primitivity in Greek religion, that Greek religion could
therefore be elucidated through cross-cultural comparisons to contemporary primitive
religions and that the figure of the sacred king was central to Greek religious thought and
social practice (more at Johnston 2008: 18–19). One of the few acknowledgments of the
book’s appearance was an anonymous review in a 1913 issue of the Journal of Hellenic
Studies, which praises Halliday for using “comparative spectacles” and knowledge of
“primitive cultures” to understand the Greeks and for looking for the “pre-Olympian element”
in Greek religion. In other words, the reviewer valued Halliday’s work precisely because it
had taken up the Ritualists’ banner. Another figure who was to loom very large in the study of
Greek religion, the Swede Martin P. Nilsson, wrote on the historical origins of Greek myths
early in his career (in the course of which he rejected the comparative approach as outdated:

Nilsson 1932: 10–11) and in the maturity of his career, placed ritual so firmly at the center of
ancient Greek religion, and myths so firmly at its margins, as to perform what now seems like
an impossible feat: he managed to write his two-volume, 1,573-page Geschichte der
griechischen Religion almost without mentioning myths at all (Nilsson [1950] 1961 and
[1941] 1967).
30. On Hooke, see also Harrelson 2005: 6380–6381; Segal 1998a: 5–7; Segal 1998b: 83–
84 (editor’s introduction to the reprinted introduction of Hooke’s book). Myth and Ritual was
the first of what became a trilogy of edited volumes on the topic: see also Hooke 1958, 1935.
31. Hooke 1933b. Hooke eschewed, however, any specific connection with Frazer and his
ideas, particularly the idea that a myth might be invented independently by different cultures:
Segal 1998a: 6–7.
32. Hyman 1955; further in Segal 1998b: 231 (the editor’s introduction to Hyman’s
reprinted article); Weston (1920) 1993; Murray 1914. Fascinating information about the
development of Hyman’s adherence to the ritualist approach can be found throughout
Franklin 2016—a biography of Hyman’s wife, the novelist Shirley Jackson (who was herself
deeply interested in the ritualist approach); see especially 338–341 but also 179–180 and
292–296.
33. Kluckhohn 1942: 68.
34. Leach 1954: 13. Further on anthropologists’ acceptance of the approach at Versnel
1993: 37–41.
35. Fontenrose 1959: 3.
37. Hyman 1960: 127.
41. Kirk 1974: 68.
42. Kirk 1974: 229–230.
43. Kirk 1974: 231.
44. Kirk 1974: 234.
45. Kirk 1974: 241.
46. Kirk 1974: 252, 246–247.
47. Kirk 1974: 235.
48. Better than Pindar had, for example, when narrating the story of Neoptolemus in
Delphi at N. 7 and Pa. 6.
49. Particularly important in this regard is Burkert (1966) 2001.
50. See, e.g., Burkert 1979, especially 26–29 and 56–58.
51. On Burkert, see also Csapo 2005: 161–180; Versnel 1993: 51–60.
52. I think here particularly of Burkert 1983: 1–82 (first published in German in 1972); and
Burkert 1979: 35–58.
53. Gernet 1981; Brelich 1969; Jeanmarie 1939; Lorenz 1963.
54. Graf 2003.
55. Burkert 1983: 130–134, 168–179; Burkert 1979: 123–142.
56. Burkert (1966) 2001: 50 (emphasis added).
57. E.g., Graf 2000; Leitao 1995; Bremmer 1978.
58. Propp (1928) 1958. Burkert also draws on Dundes 1964.

59. Burkert 1979: 6–18.
60. Burkert 1979: 23 (emphasis added).
61. Heracles, for example, may have begun as one instance of a broader type that scholars
call the “master of animals”—this may be the “original” reason that he fights with and defeats
the Nemean Lion and numerous other beasts—but when he was subsequently drafted into
being the ancestor of the Dorian kings, his victory over a lion (which was considered the most
valiant and noble of animals), and his habit of wearing that lion’s skin gained a new
resonance (Burkert 1979: 78–98).
62. Versnel 1993: 15–89.
63. Versnel 1993: 15–88; the quotation is from 87.
64. See particularly Lévi-Strauss 1958: 233–236; cf. Csapo 2005: 219–237.
65. Vernant 1974: 177–194 (= Vernant 1980: 55–77).
66. Detienne 1979: 7.
67. Detienne 1979: 5–6. In setting this Straussian course, he was also reacting to something
that Geoffrey Kirk had said a few years previously. Kirk had claimed that Greek myths as we
knew them were defective, because they represented reworkings—sometimes many times
over—of whatever the “original” myth had been. In other words, Kirk, like many earlier
scholars, ardently desired to find the “real” myth buried beneath narrative elaboration, even if
he was less sanguine than they had been about succeeding.
68. A colloquium organized by Richard Buxton at the University of Bristol in 1996 marked
a culmination; many of the scholars who participated argued that, for the particular author,
genre, or period they had chosen to treat, there simply was no easy way of dividing Greek
“mythic” thinking from Greek “rationality” and therefore no way of defining what a Greek
myth was (published as Buxton 1998a). See also Calame 1995; Graf 1995.
69. E.g., Johnston 2002; Graf 1997; Dowden 1989. See also Bremmer 2005.
70. Calame 2003: 27.
71. See also Calame 1991.
72. See particularly Calame 2003: 29: “Only by abstraction, by bracketing of the ritual
situations in which they are represented, by exclusion of the poetic forms that are the medium
of their communication, is one able to constitute a myth of Oedipus or a legend of the
Atreidae.” See also Calame 2009, the translation of a French book published in 2000. The
French version has recently been revised and expanded as Calame 2015. Where possible, I
refer in this book to the 2009 English translation, referring to the 2015 French revision only
when it presents new material.
73. Johnston 2002.
74. Lloyd-Jones 1985.
75. Bacch. 16.
76. Pi. Pa. 8, Aristonous’s paean to Apollo (Käppel 1992: no. 42), Athenaeus’s paean to
Apollo (Käppel 1992: 387–391), and HHAp. 514–419. Generally on paeans, see Furley and
Bremer 2001: 1.76–102; Rutherford 2001.
77. Pi. Pa. 12.
78. As in HHHerm. and Alcaeus’s Hymn to Hermes. The Homeric Hymn to Demeter is a
notable exception to this rule, providing as it apparently does the aitia for several rituals
performed during the Mysteries and / or the Thesmophoria.
79. Nilsson (1941) 1967: 548.
80. Furley and Bremer 2001: 1.99–102.

81. Rutherford 2001: 312–315.
82. Bacch. 5; Pi. O. 2 and O. 3; HHAp. 207–215.
83. Looms: E. Ion 190–200. Symposia: this is the setting of Call. Aet. frs. 43 and 178 (and
see Harder’s commentary ad loc.).
84. Isoc. Paneg. 4.27–31.
85. There is debate as to whether we can trust Euripides to be transmitting correct
aitiologies or even correct information about the rituals. See L. Parker’s comments on the
aitia in the Iphigenia in Tauris (Parker 2016: 346–347), and also Seaford 2009 and Scullion
1999. My opinion is that the audiences of Euripides’s tragedies would have been used to
hearing more than one aition for a given ritual and would not have been shocked by
Euripides’s presenting a different one from those they already knew. See also Christine
Hamilton (2017), who fruitfully adopts a similar approach. In any case, for my immediate
purposes, the answers to these questions do not matter much; the point is that aitia were
transmitted by Euripides and other tragedians.
86. Hellanicus: FGrH 4F*38; Androtion: FGrH 324F16.
87. Call. Aet. fr. 43b–c Harder.
3 NARRATING MYTHS
1. O’Flaherty 1988: 26 (this scholar now uses the name “Doniger”). On “mythological
zombies,” see 26 and more generally chaps. 2 and 3.
2. The neologism historiola was coined by scholars of folklore, from whom it was adopted
by scholars of ancient magic: Maas 1942: 37, 37n22; Heim 1892–1893: 495 with notes.
3. Further on mythic names, see Johnston 2013 and Chapter 5 of the present book.
4. Taken from P. British Museum 10059 [37]; Borghouts 1978: 24–25, no. 34, slightly
modified.
5. I adopt the term “paradigm” from Frankfurter 2001; cf. Frankfurter 2009.
6. Pócs 2009: 29, from a Romanian example that is still in use today.
7. A large number of European historiolae from a range of periods can be found in the
essays contained in Roper 2009 and 2004a.
8. For example, the spell including the historiola involving Isis and Nephthys, cited in note
4, further specifies that the milk must be that of a woman who has borne a male child (as Isis
bore Horus) and that the milk must be mixed with resin from an acacia, dough made of
barley, and other materials, over which the historiola itself has been recited. Once completed,
the mixture must be applied with a leaf of the ricinus plant. Another historiola from
Borghouts’s collection, to be used against headache, specifies that it must be recited “over the
buds of a Unique Bush. To be twisted leftwise, to be soaked in mucus, and the bud of the snb-
plants laced to it. To be fitted with 7 knots and to be applied to a man’s throat” (Borghouts
1978: 30–31, no. 43). A third example, to protect a house against snakes, other reptiles, and
spirits of the dead, specifies that the words must be spoken over garlic that has been ground
and pulverized with beer; the house must be sprinkled with this mixture in the night, before
daybreak (Borghouts 1978: 82–83, no. 121). PGM VI.1–47, a divinatory spell that uses
laurel, tells about how Apollo first discovered the divinatory properties of the plant; it also
specifies the date, moon phase, and time of day when the spell is to be spoken. Further on
felicity conditions, see Roper 2004b: 2.

9. Excerpted from PGM IV.94–153, a love spell in a fourth-century CE Greco-Egyptian
magical papyrus.
10. Excerpted from Davies 2004: 98, from a manuscript belonging to a French bonesetter,
dating to about 1900.
11. I find it impossible to come up with a really satisfying alternative to the word “realm”;
most of the words we might be tempted to substitute, such as “sphere,” evoke place at least as
strongly as does “realm.” “Mythic mode,” perhaps?
12. There are a few exceptions, although most of these cannot be called completely
“Greek” and are substantially later in date than the period on which I am focusing in this
book. In addition to the Getty Hexameters that I will discuss in this section, they include the
Philinna Papyrus (P. Berol. 7504 plus P. Amherst II col. ii (A), which dates to the first century
BCE and narrates a historiola of Egyptian origin in the Greek language (Ritner 1998; cf.
Faraone 2001, 1995); and a few historiolae found in the late antique Greek magical papyri,
which were found in Egypt. These owe a great deal to Egyptian models and are almost
always populated by Egyptian gods—except for PGM VI.1–47, a divinatory spell that tells of
how Apollo first tasted the divinatory laurel.
13. The tablet from Himera that shares three lines with col. II Side A (Bernabé 2013: 82, F)
is dated to the early fifth century, and two other tablets from Selinus itself (Bernabé 2013: 82,
G and H) that share phrases with the Getty text are dated to the fifth century as well. Janko
2013 posits that the Getty tablet copies an earlier archetype and suggests that at least some of
the material may be as early as the late sixth century.
14. A fuller, although differently oriented, version of my arguments here can be found in
Johnston 2013. See also Caliva 2016; Waller 2015.
15. Translation from Faraone and Obbink 2013: 12, slightly altered.
16. Johnston 2013: 133–138.
17. Some examples are Gold Tablets 3, 5, 26a and 26b G&J; Ov. Met. 1.111; Verg. G.
1.132; Lucian True History 2.3, 2.13, 2.26; Hippocr. De morbis mulierum 1.120 and
Dioscorides De Materia Medica 5.99. See further at Johnston 2013: 139–144; cf. Graf 1980.
18. Lakoff and Johnson 1980; the example “Love is a journey” is first introduced on 44.
19. Richard Janko (2013: 34) suggests that it means a tomb or chthonic shrine (i.e., a place
like those where defixiones were deposited). It may also be poetic language for some kind of
carved vessel or box into which the inscription should be placed.
20. Nor would it have worked well, in this case, for a festival, ritual or sacred place to have
more than one aition, as with, for instance, the origin of the Delphic Oracle (Johnston 2008:
38–60). The myths of the Proetides, which were associated with the festival of the Agriania
or Agrionia and with offerings of hair at the tomb of the maiden Iphinoë, offer another
example. According to some of them the girls offended Hera and according to others they
offended Dionysus; in some stories they were cured by Melampus, and in others they were
cured by waters sacred to Artemis. Even if Ken Dowden (1989: 71–96) is correct to stress
that an “initiatory” pattern underlies all of these myths, the fact remains that there was
apparently no problem, in the opinion of the Greeks, with changing the story considerably.
See also Fowler 2013: 169–178; Johnston 1999: 66–70; Gantz 1993: 311; Seaford 1988.
21. Calame 2009: 98–99.
22. Calame 2011b; Calame 2009: esp. 53–93, 98–99, 116–118; Calame 2003: 29–34, 86.
Specifically on melic poetry: Calame 2009: 98–99.

23. Johnston 1995; Braswell 1988: 359–361. It’s not clear whether Damophilus’s plea for
return was granted; if it was, it may have been granted before Pindar’s performance, which
then would have been an elaborately staged opportunity for Arcesilas to demonstrate his
generosity—a different purpose but pragmatic nonetheless.
24. Il. 24.599–627.
25. It is possible to imagine that, if we are correct in assuming that the story of Demeter
and Persephone was narrated or performed at the Eleusinian mysteries, and if the episode in
which Demeter drinks a kykeon or in which Iambe makes her laugh were narrated / performed
at the same moment as the initiates experienced their own version of such an event, the myth
may have been understood to (further) empower the ritual action. But this depends on a great
deal of speculation, and in any case, the other instances in which we can imagine something
like this happening are very few in number and are all (so far as I can see) associated with
mystery cults, which may have developed the use of myths in new directions for their own
purposes.
26. Calame 2011a; Calame 2011b; Calame 2009: 63.
27. Bacch. 17. It’s worth noting that there are also instances in which we witness the
failure of a narrative that was intended to effect change. In Iliad 9.529–655, Phoenix tells a
myth about Meleager in order to persuade Achilles to return to battle, but Achilles continues
to sulk in his tent.
28. Antiphanes fr. 189. I thank Heinz-Günther Nesselrath for pointing me towards this
passage.
29. On the Erechtheus, see Calame 2011a. On Pythian 4, see Calame 2003: 35–66; Felson
1999; cf. Felson 2004b.
30. Krummen 1990: 33–95.
31. On deixis in general, see Arethusa 37, no. 2 (2004), a special issue edited by Nancy
Felson, and particularly for the purposes of deixis that I emphasize them here, the articles by
Athanassaki, Calame, Felson, and Martin.
32. Athanassaki 2004: 330.
33. Calame 2011a.
34. Calame 2003: 2, 68–74; Greimas 1983.
35. Pi. I. 4.12 (Pillars of Heracles in the west), 52 (Antaeus’s home in the west) and cf.
zophos in line 18, a word meaning “dark” but cognate with zephyros, “west.”
36. Calame 2003: 43–66; cf. Athanassaki 2004; Felson 2004b, 1999; Martin 2004.
37. Pi. N. 5.46; and cf. Pi. P. 9.91.
38. Calame (2009: 116–117 and cf. 98) does suggest that, in the case of epinicians, the
pragmatic force of mythic narratives was so powerful as to make them into performative
utterances: that is, as the myth within an epinician was narrated, the audience began to
perceive the victor differently, which simultaneously changed his status within the
community at that very moment. This stretches the concept of “performative utterance”
further than I am comfortable with, however.
39. Landy 2012. Of course, we might consider a work of fiction to be both “formative” and
“informative” (pace Landy, who thinks that fiction can never be truly informative in the sense
of changing minds in more than the simplest ways).
40. Brakke 1999.
41. Cf. the example offered by William Doty (2000: 51–52) of the use made of proverbs by
the Lele, a Bantu people of East Zaire: adult males who have learned thousands of traditional

proverbs during their initiation period as youths view the entire universe through the
proverbs’ cumulative lens.
42. Landy 2012: 10.
43. Barkun 2013: 29–33; Kripal 2011: 2; Partridge 2004: chap. 6, esp. 119, 124–138. Or as
Graham Harvey (2000) says, “Fantasy does not necessarily misdirect people away from
consciousness raising, it need not be an opiate, but can be the much-needed catalyst for
change.” Cf. also Clark 2003.
44. Cf., too, the remarks of E. M. Forster: “People in a novel can be understood completely
by the reader, if the novelist wishes, their inner as well as their outer life can be exposed. And
this is why they seem more definite than characters in history, or even our friends” (Forster
[1927] 1985: 57, as cited by Eder, Jannidis, and Schneider 2010: 53). Cf. Hoorn and Konijn
2003: 253–255.
45. As Wendy Doniger [O’Flaherty] (1988: 27–28) notes.
46. Cuneo 2001: 27–28, 127. Cf. Christopher Partridge (2004: 119–141), who
demonstrates that popular culture (films, television shows, novels) is a “key sacralizing
factor” that plays a highly influential role in shaping spiritual and religious beliefs,
“contributing to a sense of what the real world is all about” (119).
47. Cf. Arist. Poetics 9 (1451a37–1451a38) and 24–25 (1460a12–1461b25); Phelan 2017:
32–36.
48. Luhrmann 2012.
49. Luhrmann 2012: 189–226; the quotation is from 221. For the study, see 202–215.
50. Luhrmann 2012: 48, 74–86.
51. Biblical metaphors of God as a lion include Hosea 5.14 and Rev. 5.5. On Evangelical
fiction reading, see Luhrmann 2013c; and Luhrmann 2012: 83, 73–74, 129–131; cf.
Luhrmann 2013a, 2013b.
52. Luhrmann 2012: 39, 52, 94.
53. MRI studies have shown that personal prayers to God (as opposed to formalized, rote-
memorized prayers such as the Lord’s Prayer) activate the same regions of the brain as does
social interaction. In other words, cognitively, God is experienced as a social relationship.
Luhrmann 2012: xvi and additional information in her note 10; see also Schjoedt et al.
(2009), who show that whereas informal, personal prayer to God stimulates the social areas of
the brain, formal prayers and expressing wishes that Santa will bring you something, do not.
54. For a general introduction to PSI and PSR, see Giles 2010, 2002. The term was
introduced by Horton and Wohl 1956. On the differences between PSI and PSR, see Schmid
and Klimmt 2011; Giles 2002. What researchers call PSI might sail less formally under a
variety of other names. In a piece in the New York Times Magazine (November 22, 2013),
JoAnn Klimkiewicz describes how she and her sister, fresh from reading Rhoda Byrne’s self-
help manual The Secret, “secreted” the supermodel Christy Turlington right into their lives.
55. On PSI with fictional characters, see, for example (in addition to Schmid and Klimmt
2011; Giles 2010, 2002), Mittell 2015: 124–149; Eder, Jannidis, and Schneider 2010; Reicher
2010; Schmid and Klimmt 2010; Hoorn and Konijn 2003.
56. The anecdotes about the reception of Little Nell’s death are mentioned by many
treatments of Dickens and of seriality; see, for example, Gardner 2012: 56–57; Walsh 2007:
148–169. More generally on audience reaction to serials, see Mittell 2015: especially 118–
163; Gardner 2012: 29–60; Nussbaum 2012. The anecdote about Daniel O’Connell is taken
from an article published in the Irish Times on January 7, 2012

(https://www.irishtimes.com/opinion/an-irishman-s-diary-1.950636). And then there is
Dickens camp, held for a week each summer at the University of California at Santa Cruz,
where adults dress up as their favorite characters: Lepore 2011. Similarly, when Sidney
Smith, author of a long-running daily comic strip called “The Gumps,” killed off a beloved
character named Mary Gold in 1929, the collective grief amongst readers brought an
avalanche of letters to newspapers that carried the strip, demanding that Mary somehow be
revived. Like Dickens, Smith refused to relent (Gardner 2012: 53–55).
57. Letters to Holmes: Rule 1989. When the BBC’s interpretation of Holmes’s death was
aired as “The Reichenbach Fall” on January 15, 2012, it met with a similar (even if much
more self-consciously constructed) response and generated an enormous number of memes
(of which “I Believe in Sherlock,” perhaps orchestrated by the BBC itself, was the most
prominent), discussions on social networks and articles in traditional print media. Not
surprisingly, given that any twenty-first-century audience could feel secure that Holmes
would eventually return, many of these focused more on the question of how Holmes’s
apparent death would be explained away than on lamenting the death itself.
58. Gunn 2013; cf. Arngrim 2010: 130–134 (a memoir by the actress who played Nellie
Oleson on Little House on the Prairie, which includes descriptions of viewers accosting her
for what “she” did to Laura Ingalls). Further on PSI with unpleasant characters, see Mittell
2015: 142–163, with 364n2; Hoorn and Konijn 2003.
59. Giles 2010: 454.
60. As I was writing the first version of this chapter, I was struck by a review, posted on
December 16, 2013, of the third-season finale of Homeland, in which the television critic
Willa Paskin says, “I am more surprised than anyone about what happened while watching
the season finale of Homeland, ‘The Star.’ I cried. Three times” (Paskin 2013).
61. Luhrmann (2012) and Schjoedt et al. (2009) certainly treat the topic but without ever
mentioning the terms PSI and PSR. Similarly, John Caughey (1984) discusses social
relationships in the realms of what he calls fantasy and hallucinations or delusions but never
mentions PSI and PSR.
62. Verity Platt (2010) has convincingly made the argument that we must similarly erase
the distinction between viewing portraits of the gods and heroes as art and as sacred objects
during antiquity.
63. The iamata and the confession stelai provide the best-known examples, but we might
also think of various other reports of encounters with gods, such as that by the nympholeptos
Archedamus (IG I3 976–980) or the story of Phidippides’s encounter with Pan, which was
said to have prompted the Athenians to establish new cult to the god (Hdt. 6.105–106; I return
to the incident later in this chapter).
64. Of course, very few people actively, consciously, and consistently hold any religious
belief, all of the time. Nor do the majority of people who follow politics, who campaign to
protect the environment or who engage in any number of other things to which they are
sincerely committed think about them actively, consciously, and consistently, all the time. As
Paul Veyne has noted, the mind changes its programs of truth as it changes its interests;
whether you “believe” in something can only be measured by whether you are committed to
what you are doing or saying at the moment that you are doing or saying it. (Veyne 1983: 86–
87, drawing on the work of Paul Pruyser). Cf. Partridge 2004: 125, discussing TV viewers:
“Some people will receive what is communicated uncritically, others will reject the message
and yet others will enter into a relationship of negotiation. Some viewers of a news report, for

example, or even a television series such as The X-Files, may uncritically accept everything
communicated. Others will accept parts of what is presented as an accurate representation of
reality and reject other elements as distortion or simply fiction. Yet others will reject
everything presented. Such decoding, of course, is influenced to a large extent by prior
commitments, and plausibility structures.”
65. On the seriality of The Old Curiosity Shop and Middlemarch and their similarities to
serial television—and on the allures and frustrations of serial storytelling more generally—
see O’Sullivan 2013.
66. Mittell 2015: 124–149; Schmid and Klimmt 2011: 254; Giles 2002; Murray Smith
1995. Cf. O’Sullivan 2013.
67. Cf. O’Flaherty 1988: 49.
68. On episodic serials, see, e.g., O’Sullivan 2013; Mittell 2010: 228–242.
69. In Euripides’s version of Antigone’s story, Haemon and Antigone marry (hypothesis to
Sophocles’s Antigone and the scholia to line 1350). Ion of Chios (fr. 1 Page) had Antigone
and Ismene burned to death inside of Hera’s temple by their nephew Laodamas. Cf. Johnston
2006; Gantz 1993: 519–522. Euripides’s Electra presented a heroine who had been married
off to a peasant; Sophocles’s lost Aletes had her married off to Pylades.
70. O’Sullivan 2013. I will discuss series and episodic serials in more detail in Chapter 7.
71. Jason Mittell (2015: 165–166) discusses the experiential differences between what he
calls “ideal” and “competent” viewers.
72. Hitchcock, as quoted in Chatman 1978: 60. I take the citation from Mittell 2015: 177.
Cf. also Bordwell 1985: 57–61.
73. Bacch. 17; Pi. P. 4.
74. And even epics are themselves only a smaller section of a larger story—a couple of
weeks out of the Trojan War, for example. Had we Homer’s sources, we would probably
discover that he was playing the same games as Pindar and the tragedians.
75. Hes. Th. 1011–1018.
76. Phelan 2017: x, 5.
77. Hänninen 2017; Bennett 1999.
78. I should also note that “ghost story” is a term into which both literary authors and
scholars collapse a wider range of stories about the supernatural—not just those explicitly
involving ghosts.
79. Wooffitt 1992: 114–152.
80. Bennett uses capital letters to indicate that her informants raised the volume of their
voices.
81. Bennett 1999: 61–62.
82. Hänninen 2017: 131.
83. Hänninen (2017: 128) adds, “For readers or listeners, it is more plausible to hear or
read extraordinary, uncanny and marvelous events taking place in a distant tale world than in
everyday reality. When supernatural experiences take place in everyday reality, it may leave
the readers or listeners perplexed. When that happens, narrators need to express that they
realize the distinction between the ordinary realm of everyday life and the supernatural realm
and locate themselves in the ordinary realm of rationality by the end of the story at latest.”
84. The quotation is from James’s preface to his collection More Ghost Stories of an
Antiquary, first published in 1911. He made similar remarks in his introduction to Collins
1924. These and similar short essays by James on the writing of ghost stories are gathered by

Darryl Jones at the end of his edition of James’s work (James 2011). The present quotation is
from page 406 of that edition. Some of Joseph Sheridan Le Fanu’s work, which preceded
James by several decades, began to introduce elements that fit the X/Y model (e.g., “Green
Tea,” 1872). At about the same time as James, E. F. Benson, Algernon Blackwood, and
Arthur Machen often used the X/Y model in their ghost stories, too, as did Sir Arthur Conan
Doyle in his supernatural fiction (which has recently been collected in Doyle 2017). Later
works that notably use the X/Y format include F. Scott Fitzgerald’s “A Short Trip Home”
(1927), Daphne Du Maurier’s “The Blue Lenses” (1959), Peter Straub’s Ghost Story (1979),
many of Stephen King’s novels, and Joyce Carol Oates’s The Accursed, for instance. Authors
such as James were reacting to the Gothic horror story in its purest form, which tended to be,
if not supernatural all the way through, certainly not a portrait of everyday life—e.g., Horace
Walpole’s The Castle of Otranto: A Gothic Story (1764) or Matthew Lewis’s The Monk
(1796).
85. Bennett 1999: 16.
86. Bennett 1999: 115.
87. M. R. James, “Canon Alberic’s Scrapbook,” first published in Ghost Stories of an
Antiquary (1904) = James 2011: 3–13; the quotation is from page 13.
88. I use the word “diegetic” here not as Plato does in R. 3 (392d3) to mean simply a
narrative, but rather as scholars of narrative do, to refer to the world within which a story, or
series of related stories, unfolds. “Non-diegetic” comprises external elements that might be
used to enhance the story, such as the background music in a film.
89. Cf. Hänninen (2017: 128–129), who develops the work of Cohn 1978. Cohn
envisioned a sliding scale that ran from “consonance” to “dissonance” in the relationship
between what the Narrating I and the Experiencing I reported. Hänninen notes that when
people report their supernatural experiences, those whose two I’s are consonant typically are
still processing their experience; they are unsure what happened, what it meant and whether it
is worth telling. Conversely, those whose reports are dissonant are more confident in the
accuracy of what they are telling.
90. I paraphrase a statement from Bennett 1999: 38. By “traditions of disbelief,” she refers
to the post-Enlightenment assumption, widely held in the Western world (and increasingly
outside of the Western world), that religious or supernatural experiences can be explained
away with reference to something such as (as Andrew Lang long ago put it) “rats, indigestion,
dreams and of late, hypnotism” (Lang 1894: 173, quoted in Bennett 1999: 32). “At the last
ditch,” Bennett continues, “rationalists fall back on the argument that, even if none of their
arguments will fit the case now, given time and the advance of scientific knowledge, a
‘rational’ cause will eventually be found.” On “traditions of disbelief,” see also the seminal
piece by David Hufford (1982).
91. Bennett 1999: 3 (emphasis added); cf. Hänninen 2017: 127–128.
92. I take the term from William Hansen (2017: 24), who cites Labov and Waletsky 1967:
see esp. 32, 34.
93. Hdt. 6.105–106.
94. Healed people would dedicate ex-voto inscriptions that narrated their stories in a
formalized first- or third-person voice; some of these were collected by the priests and
inscribed on the stelai, which were erected in the sanctuary during the second half of the
fourth century BCE. Pausanius mentioned that there were six such large stelai when he
visited in the second century CE; we have found two of these and fragments of two more

(Paus. 2.27.3). For the Greek texts and a translation, see Edelstein and Edelstein 1945:
testimony 423; and cf. discussion in LiDonnici 1995. I cite individual iamata from the stelai
by the numbers that the Edelsteins use.
95. There is also a long healing inscription from Epidaurus, dated to about 160 CE, in
which the patient, a man from Mylasa in Asia Minor, tells us that Asclepius called him to
Epidaurus (presumably in a dream): “During the sea voyage, when in Aigina, he told me not
2
to be so angry” (IG IV 1, 126 = Edelstein and Edelstein 1945: testimony 425, the Edelsteins’
translation). Here we may be getting a less-mediated, personal voice, i.e., something closer to
a memorate.
96. Iamata nos. 27 (abdominal abscess), 30 (arrow in lung), and 14 (kidney stone).
97. The word I am translating as “it seemed” is ἐδόκει. The phrase is found in testimonies
nos. 2, 3, 4, 6, 7, 8, 9, 17, 18, 21, 23, 24, 27, 28, 29, 30, 31, 32, 34, 35, 37, 38, 39, 40, 41, 42.
The phrase clearly became formulaic but nonetheless would have functioned to remind the
audience of the distance between narrator and the events being narrated.
98. Hes. Th. 22–34.
99. Th. 26–28, trans. Most, slightly modified.
100. See West 1966: 158–164; and, more recently, with more nuancing of the Hesiodic
experience against Near Eastern comparanda, López-Ruiz 2010: 48–83.
101. West 1966: 160–161.
102. López-Ruiz 2010: 54.
103. Joshua Katz and Katharina Volk (2000; followed by Carolina López-Ruiz 2010: 77–
78) have persuasively argued that in “mere bellies” we are meant to hear an allusion to
engastrimythoi, “belly-talkers,” who claimed to have spirits in their stomachs and who
therefore could offer a cheap, easy-to-access local version of the prestigious form of prophecy
offered by the Delphic Pythia. I think this is correct, but it does not contradict (indeed, it
enhances) the contrast that Hesiod’s Muses are drawing between shepherds and themselves.
104. The expression has Near Eastern precedents, as López-Ruiz (2010: 56–73), amongst
others, has argued, which helps us understand how Hesiod is using it to subtly underscore the
divine source of his information, but nonetheless, as an interjection, the expression plays the
dialectical role that I describe here.
105. Hesiod’s quotation of the Muses’ claim that they can lie persuasively when they want
to but also tell the truth is interesting in this respect, as well, since it introduces ambiguity on
yet another level: even if we decide to believe that Hesiod really did meet the Muses and
received their gift, we can’t be sure that the gift itself is not riddled with misleading
statements. This means that, at least initially, attentive listeners will weigh every word that
Hesiod sings, hoping to catch the Muses out at their game. Notably, the seventy lines or so
that immediately follow Hesiod’s remark about the oak and rock are a treatise on the nature of
the Muses themselves—their parentage, their birth, their names, their central role in the
creation not only of poets but also of kings, or in other words, topics on which we would least
doubt the Muses’ honesty. Will attentive listeners have let down their guard by the time we
get to the creation of the cosmos itself?
106. The Hymn to Hermes also includes encounters between a mortal and the gods: an old
man, who sees Hermes stealing Apollo’s cattle, is told by Hermes not to admit that he knows
anything but later reports what he has seen to Apollo (lines 87–93 and 187–211). In contrast
to humans in the other episodes that I am discussing here, the old man expresses no alarm or
even surprise during either encounter and suffers no ills. This fits the ludic nature of this

Hymn, which humorously showcases the young god’s craftiness. The typical awe and terror
associated with divine encounters in the other Hymns would strike a wrong note in this
atmosphere. Notably, other myths about this old man give him a name—Battus (“The
Babbler”)—and end with him being turned to stone by Hermes: Ant. Lib. 23, citing a number
of earlier treatments, including Hesiod’s Catalogue fr. 256 MW.
107. HHDem. 98–295.
108. HHAp. 388–447.
109. HHAph. 75–184.
110. HH 7.1–53.
111. HHDem. 235
112. Buxton 2009: 29–37, his translation.
113. Od. 3.371–373, trans. Buxton 2009: 29.
114. Buxton 2009: 43.
115. This terseness was first noted by Allen 1908. The Chimaera: Il. 6.179–183 and
16.326–329; Cerberus, Il. 8.368, cf. Od. 11.623, where Cerberus is called simply a dog;
Typhon: Il. 2.782–783.
116. I am also attracted by the suggestion of Marianne Govers Hopman (2012: 24–25) that
Odysseus emphasizes the monstrousness of Scylla because it is an episode in which he feels
that he failed to win the day.
117. Scylla: Od. 12.85–99 (fifteen lines); and Typhon: Th. 823–835 (twelve lines).
118. I think particularly of the passages that immediately precede the episodes with
Polyphemus and Circe (Od. 9.62–233 and 10.133–188), each of which shows us the
workaday world of Odysseus and his men.
119. Od. 12.235–245.
120. Od. 12.245–250.
121. Od. 12.251–260.
122. Od. 13.1–6.
123. Although in I. 4.49–51, to flatter Melissus, a victor who was small in stature, Pindar
does claim that Heracles was small in stature as well.
124. Bacch. 17.117–178, 3.54–59. I am counting Croesus as a hero here, although he was a
historical figure, because Bacchylides’s treatment of his story, and particularly the
interventions of Zeus and Apollo, align him with the heroes treated elsewhere by
Bacchylides, such as Theseus and Heracles.
125. Pi. O. 13.83, P. 2.49–51, P. 10.48–50, P. 3.11–12.
126. For example, Pi. O. 7.47, I. 4.52, I. 8.59.
127. Pi. O. 1.28–36.
128. “It is proper for a man to speak well of the gods, for less is the blame. Son of
Tantalus, of you shall I say, contrary to my predecessors, that when your father invited the
gods to his most orderly feast and to his friendly Sipylos, giving them a banquet in return for
theirs, then it was that the Lord of the Splendid Trident seized you, his mind overcome with
desire, and with golden steeds conveyed you to the highest home of widely honored Zeus”
(Pi. O. 1.35–42, translation William Race).
129. Pi. O. 1.24–25: “[The victor’s] glory shines in the settlement of fine men founded by
Lydian Pelops, with whom the mighty holder of the earth Poseidon fell in love.”
130. E. Hercl. 844–866; E. Alc. 1578–1612; S. OC 1586–1666.
131. E. Hipp. 1173–1254.

132. E. Hipp. 1173, 887–898.
133. E. Hipp. 1185–1198.
134. E. Hipp. 1201–1243.
135. Hippolytus is named already in the (sixth-century?) epic Naupactica as someone
whom Asclepius brought back to life, which typically is said to have happened after his
chariot wreck (fr. 10, West). Plutarch, in his Theseus (28.2) says, “Theseus did, indeed, marry
Phaedra, but this was after the death of Antiope, and he had a son by Antiope, Hippolytus, or,
as Pindar says, Demophon. As for the calamities which befell Phaedra and the son of Theseus
by Antiope, since there is no conflict here between historians and tragic poets, we must
suppose that they happened as represented by the poets uniformly” (trans. Bernadotte Perrin).
Cf. Gantz 1993: 285–288.
136. Bacch. 16.
137. Outside of Athens, tragedies could be (re)performed at the festivals of other gods—
but the subject matter had just as little to do with those gods, most of the time, as it did with
Dionysus: Scullion 2002.
4 THE GREEK MYTHIC STORY WORLD

1. Lewis 1947; cf. remarks made in the essays collected in Lewis 1966.
2. Coleridge 1817: chap. 14.
3. Tolkien 1947: 60.
4. Saler 2012: 28.
5. M. Wolf 2012: 25.
6. Kalidah: a vicious beast that has the body of a bear, a head like a tiger and claws sharp
enough to tear a lion in two, first featured in The Wonderful Wizard of Oz (1900) and
appearing in other Oz books thereafter. Mangaboos: an underground race of people made of
vegetable material, introduced in Dorothy and the Wizard in Oz (1908).
7. M. Wolf 2012: 33.
8. Fourteen books about Oz were written by L. Frank Baum between 1900 and 1920; after
his death, his publishing house, Reilley and Lee, commissioned twenty-six more. In addition,
Baum published eleven books that many people consider as canonically belonging to the “Oz
World,” and others published an additional six books and two short stories. A complete and
accurate list, so far as I can establish, is to be found at
http://oz.wikia.com/wiki/List_of_Oz_books. Potentially more complete, but also more
confusing, is the information at http://en.wikipedia.org/wiki/List_of_Oz_books. Vidal 1977a
and 1977b are helpful and insightful.
9. First printed on the end papers of Faulkner 1946.
10. Mittell 2015: 261–291; M. Wolf 2012: 153–197. Other examples are The Languages of
Tolkien’s Middle-Earth: A Complete Guide to All Fourteen of the Languages That Tolkien
Invented (Noel 1980); Stephen King’s “The Dark Tower”: The Complete Concordance,
Revised and Updated (Furth 2012); and (just published as I write this) Flora of Middle-Earth:
Plants of J. R. R. Tolkien’s Legendarium (Judd and Judd 2017).
11. And then there are Herodotus’s gigantic, gold-digging Indian ants—which he assumes
are real (Hdt. 3.102–105). Herodotus also mentions that some people believe that each year, a
single man turns into a wolf for a few days and then turns back again. But these shape-
shifting people are the Neuri—living far away from Greece, beyond even the distant

Scythians, and he reports that it’s only the Scythians and Greeks who have lived amongst the
Scythians who believe that the Neuri actually can do such a thing (Herodotus certainly
doesn’t believe it: Hdt. 4.105). Should we call these tales myths and add the ants and the
werewolves to our bestiary? By a criterion that I used for my heuristic definition of myth in
Chapter 1 and that I will explore more later in this chapter, we should not: they have no place
in the complex web of relationships amongst heroes and gods.
12. Epimenid. frr. 27–29, Bernabé; cf. Plut. Dinner of the Seven Wise Men 157d.
13. Il. 6.171–183, trans. Lombardo, slightly modified; cf. Il. 16.328–329, where the
Chimaera is described simply as “raging” and “an evil for many people.”
14. Ο. 13.90. HHAp. 368 calls her merely “ill-named.” Hes. fr. 43a MW 84 calls her “fire-
breathing,” although it’s possible that another adjective was squeezed into the six syllables
that are missing from the line. Hesiod takes fuller advantage of the possibilities at Th. 319–
322: “she breathes forth invincible fire, is terrible, swift and mighty with three heads: a
fierce-eyed lion’s, a she-goat’s, and a snake’s—a mighty dragon’s—breathing forth the
terrible strength of burning fire.” On textual problems in the lines that immediately follow
these (323–324), continuing the description in a contradictory way, see West 1966: ad loc.
Most editors regard these two lines as interpolation from Il. 6.181–182.
15. S. Tr. 573–574; A. Ag. 1234; E. Med. 1359.
16. See my comments in Chapter 3 on Odysseus’s narration of the episode with Scylla and
Charybdis.
17. Hes. Th. 820–868.
18. It is possible that one reason that ancient authors could get away with restrained
descriptions of the monstrous is that art portrayed many of these creatures fairly frequently
and with gusto. We know this not only from the representations that remain to us but also
from ancient reactions to those representations that are embedded in literature: e.g., E. Ion
184–218 and Paus. 2.27.2, which tells us that the throne on which sat the large statue of
Asclepius in Epidaurus was decorated with scenes of Bellerophon fighting the Chimaera and
Perseus beheading Medusa.
19. Cf. my discussion of memorates in the final section of Chapter 3, where I make the
same point but with reference to how myths work to establish credibility more generally.
20. Gaiman (1996) 2001: 214.
21. Heracles in the Hesperides: attested as early as Pherecyd. FGrH 3 F 16 and 17. See
further Fowler 2013: 291–299 and Gantz 1993: 410–413. The visit, in all its narrative
variations, is well attested in art from the early sixth century on. The vase shown in Figure 4.1
is especially interesting because the painter has managed to combine so many elements from
what are different parts of the story of Heracles’s labors as we best know it: Atlas holding up
the sky, one of the Hesperides (so labeled), the tree with apples guarded by the snake and
Heracles in what appears to be the cup he borrowed from Helios (which is more typically
associated with the Labor to obtain the cattle of Geryon, although Pherecyd. FGrH 3 F 18 has
him use it for both Labors: Fowler 2013: 294–299). Hermes and Athena have been added on
either side, ready to aid Heracles. Perseus in the Hesperides: Hesiod places the Gorgons in the
same place as the Hesperides, “beyond glorious Ocean, towards the edge of Night” (Th. 270–
271) and Perseus’s visit to their garden is attested by a red-figure hydria from Paestum now in
Lisbon, dated to 340–330 BCE, showing the hero with three nymphs, a dragon and an apple
tree (RVP 258, 1022 = LIMC Hesperides 62). See further Gantz 1993: 305–306.

22. Pi. P. 10.29–30—a poem that gives us our earliest attestation of Perseus in Hyperborea.
Heracles: Pi. O. 3.16–33 and perhaps also in Pherecydes, although in association with his
quest for the apples of the Hesperides: Fowler 2013: 295–297. Further on Hyperborea in the
early Greek tradition, see Fowler 2013: 132–133, 606–607.
23. Hawes 2017a appeared too late for me to make full use of it, but I would single out as
particularly relevant to this discussion Hawes’s own contribution to the volume and that of
Clarke.
24. In using the word “network,” I don’t mean formally to invoke “network theory,”
although some of the implications of the subdivision known as “social network theory”
certainly might be applied to my material.
25. HHAp. 353–354, 368; Il. 6.179–182; Hes. Th. 304–325.
26. Gantz 1993: xxxix; Hes. Th. 280–283; Il. 6.179–182.
27. Hellanic. FGrH 4F51; Musae. fr. 100 Bernabé; Pherecyd. FGrH 3F88 and Gantz 1993:
467–468; Il. 14.321–322, Il. 12.292 with scholia = Hes. Cat fr. 140; Bacch. fr. 10 SM;
Pherecyd. FGrH 3F89; Hes. Cat. frs. 141 and 145; Bacch. 26; Sapph. 206 LP and Gantz
1993: 260–268. See also Gantz 1993: 811.
28. Od. 11.630–631; Paus. 9.31.5 and 10.28.3 (citing early works now lost); Hes. fr. 280
and Gantz 1993: 291–295; Hellanic. FGrH 4F168a; Alcm. 21 PMG; Stesich. fr. 191 PMG;
Bacch. 5; Hes. Cat. fr. 25 MW; Od. 11.601–627. Note that there is also a tradition of Theseus
meeting Meleager’s ghost in the Underworld, according to Hes. fr. 280 MW.
29. Pherecyd. FGrH 3 F 88 and 22; Pherecyd. FGrH 3 F 112 and 31; Pi. P. 4.220–242;
Call. fr. 233 Pf. But the story is much earlier—see Gantz 1993: 255–266; Ibyc. 291 PMG and
Lyc. 174 and 798.
30. First told fully at Apollod. 3.14.4, but Reed 1996 traces the story of Adonis’s death to
the late fifth / early fourth century author Antimachus of Colophon and artistic
representations suggest this part of the story is earlier, too: Gantz 1993: 102–103 and 729–
731 and cf. Reed 1995. See also E. Hipp. 1420–1422, Theocr. 3.46–48, Sapph. 140 LP,
Apollod. 3.14.4 and Hyg. astr. 2.7.3.
31. Il. 14.153–360; HHHerm.
32. Pi. O. 13.63–92; Hes. Cat. fr. 124 and 126; Gantz 1993: 199 and 811.
33. Ov. Met. 6.148–149.
34. Lyc. 143, 174 and 798.
35. Hes. Cat. frs. 190 and 191; Bacch. 5; E. Hipp. 1416–1422.
36. Graf and Johnston 2013: 66–93.
37. Gantz 1993: 248–245; Kearns 1989: 117–124; but see also Fowler 2013: 470–472 for a
different view.
38. Medea: our first mention of her involvement in Theseus’s story is Call. Hec. frr. 232
and 233 Pf. Vases from the mid-fifth century seem to show her attempting to poison him
(details at Gantz 1993: 255–256, and see also Sourvinou-Inwood 1979). Heracles: earliest
evidence is a shield band relief from Olympia dated to about 560 BCE, showing figures
labeled “Theseus” and “Pirithous” sitting on a chair, while a third figure approaches with a
drawn sword; this figure’s name has been worn away (inv. B 2198). See discussion at Fowler
2013: 487–488 and Gantz 1993: 292. Theseus’s attempt to kidnap Helen seems to be alluded
to at Il. 3.143–144, where Theseus’s mother is Helen’s servant and a scholium A to Il. 3.242
says the story was told in the epic cycle.
39. Graf and Johnston 2013: 167; Graf 1987.

40. Apollod. 1.9.16; Apollod. 1.82–83, probably derived from Euripides’s lost Meleager.
41. Cf. Eco 1990: 198.
42. Od. 19.394–398 says that Hermes taught Autolycus thievery and the art of making
slippery oaths in return for generous sacrifices. Pherecydes (FGrH 3 F 120) says that Hermes
was Autolycus’s father.
43. M. Wolf 2012: 205–12.
44. On crossovers, see, for example, Mittell 2015: 267–270; Mittell 2010: 28, 46.
45. Amongst the earliest of our own era were the Houseboat books by John Kendrick
Bangs (1895 and 1897), which posited gatherings of the ghosts of famous men and women,
real and fictional, in a houseboat sailing down the Styx (Sherlock Holmes was a central figure
in the second volume, The Pursuit of the Houseboat).
46. Paus. 1.28.7.
47. Or perhaps by Euripides in his Phoenissae, depending on how you judge some
controversial lines; on both matters, Gantz 1993: 295–297; Kearns 1989: 208–209.
48. Johnston 1999: 219–220 with notes.
49. Notably including one in which a girl passes through the wardrobe in a spare room of
her aunt’s house and enters another world: “The Aunt and Amabel” (included in Nesbit 1912;
original pagination is not available to me).
50. Th. 1.3.
51. Johnston 1999: 279–287; Lardinois 1992.
52. Calame 2003: 68.
53. Pi. P. 9.26–65.
54. Pherecyd. FGrH 3 F 22; A. Eu. 727–728; E. Ion 985–1010.
55. In a letter to Robert E. Howard, dated August 14, 1930, as cited at
http://www.hplovecraft.com/creation/necron/letters.aspx.
56. Further on the topic that follows, see the excellent treatments at Fowler 2016, 2011.
57. Cf. Fowler 2016: 201–203.
58. Th. 1.8.4, 1.9.4. Thucydides subsequently stresses that a good historian should not
accept every detail that comes down through tradition even in connection with relatively
recent events such as the Athenian tyrannicide, and all the more so when it comes to the sorts
of earlier events that are favored by poets and some prose writers. Such narrators, he charges,
are more interested in catching the public’s attention than in discovering the truth, and their
subject matter, in any case, has been “lost in the unreliable streams of mythology.” Yet he
puts what he himself has said about Troy and Minos into a different category (1.21).
59. This principle permeates Eliade’s work: e.g., Eliade 1963, 1959, 1954. See also Rennie
1996: 77–225.
60. Green 1997: 30; see also 14–15.
61. Parian Chronicle: FGrH 239. See Oxford fr. entries 1, 3, 10–12, 17, 40, and Paros fr.
entries 8 and 10; Lindian Chronicle = IG XII, 1 Lindos II.2 = SEG 39.727. Generally on the
Lindian Chronicle, see Higbie 2003.
62. D. 60.6–8, 60.28 (the Funeral Oration). Cf. Th. 2.29.3.
63. Pl. Phdr. 229b4–c3.
64. Paus. 1.39.1, 2.35.11, 8.2.1–4; generally on the question of which myths Pausanias held
to be true, see Pirenne-Delforge 2008. Some of the monuments that our ancient sources
describe provide tantalizing glimpses of myths that we no longer have in any full narrative

form: Philochorus, most enticingly, tells us of a tomb in the Delphic precinct on which was
inscribed “Dionysus, son of Semele” (FGrH 328 7).
65. Further on Asclepius’s cures, see the section “Ancient Narrations of Remarkable
Incidents” in Chapter 3. For an overview of Amphiaraus, see Johnston 2008: 90–95. Pan:
Hdt. 6.105–106. Apollo: Cic. Div. 1.37, D.S. 22.9, SIG 398, Pomp. Trog. 24.8, Paus. 10.23–
24, 1.4, and cf. Call. H. 4.171–184.
66. The description of Athena bringing rain is at IG XII, 1 Lindos II.2. D 1–59. On
manifestations of the gods, see Henrichs 2010; Graf 2004; Bravo 2003; Versnel 1987.
5 CHARACTERS
1. In my summary here, I am most indebted to Maria Reicher (2010), who provides a good
bibliography of earlier treatments, and to a lesser extent to Eco 2009. In contrast, see, for
example, Richardson 2010.
2. Reicher’s (2010) maximal character is in some ways similar to Eco’s “fluctuating
character,” which had been introduced the previous year in an Estonian journal (Eco 2009). I
find Reicher’s articulation of the relationship between maximal and submaximal characters
more compelling than that between Eco’s fluctuating and what, I presume, he would call non-
fluctuating characters, because of Reicher’s emphasis on the functions served by submaximal
characters.
3. Reicher (2010: 129–130) suggests that there are also cases in which a character has
become so diversely represented as to produce two maximal characters, between whom there
are few overlaps and each of whom has its own submaximal characters. (For an example, she
offers Marlowe’s Faust as a maximal character and Goethe’s Faust as a separate maximal
character, each of which has spawned submaximal Fausts.)
4. See Rule 1989.
5. This is similar to what the social psychologist Robert Cialdini (2016) calls “pre-
suasion,” although he is interested in how individuals’ actions and choices (particularly
purchasing choices) may be affected by front-loading particular opinions.
6. Th. 1.22, trans. Radice.
7. I am reminded of E. M. Forster’s remarks about people in novels, as I cited them in
Chapter 3, note 44. I am also reminded of an anecdote told by Alexandre Dumas and repeated
by Umberto Eco (2009: 83). When, in 1860, Dumas joined a group of tourists visiting the
Chateau d’If, where his hero Edmond Dantès, the Count of Monte Cristo, had languished, he
was surprised to hear the guides speaking of Dantès and other characters in the novel as if
they had been real, and completely ignoring a real, historical figure who actually had been
imprisoned there. Dumas comments in his Memoirs, “It is the privilege of novelists to create
characters who kill those of the historians, the reason is that historians only evoke mere
ghosts, while novelists create persons in flesh and bones.”
8. Il. 24.334–335.
9. Il. 5.370; Hes. Th. 188–192.
10. An actor portraying Elizabeth II parachuted out of a helicopter at the opening
ceremonies of the 2008 London Olympics—alongside an actor portraying James Bond—with
the enthusiastic support of the real Elizabeth, who played herself in scenes leading up to the
point when “she” climbed into the helicopter. This stunt presented an Elizabeth who was far

outside of her usual portrayals but it was greeted enthusiastically—perhaps because her
accompaniment by James Bond clearly marked it as fictionalizing.
11. As is the case with mythic characters, most of these treatments are careful to plant their
roots within some trait or episode that is already part of the canon—that is, a trait or episode
that has already been mentioned by Doyle himself: Holmes’s intention to retire into
beekeeping, his admiration for Irene Adler or his obsession with Moriarty, for example.
12. The neologism “plurimediality” and its cognates are used in different ways by different
scholars, as are the terms “transmedial” and “intermedial,” which sometimes seem to be
regarded as synonyms for “plurimedial.” I adopt “plurimedial” (and my initial example here)
from Denson 2011. Generally, see also Mittell 2015: 292–318; Ryan 2012; Eder, Jannidis,
and Schneider 2010: 17–20, 28–30; Richardson 2010; W. Wolf 2007, 2005; and Jenkins 2006:
95–134. Mittell (2015: 310–318) notes that transmediality most often refers to a phenomenon
that has been orchestrated by the creator of the “mother-ship” representation—that is, the
relationship between different representations is purposefully built to be one between what is
perceived as a “main narrative” and “spin-offs” (e.g., Star Wars movies on the one hand and
Star Wars action figures on the other hand). In cases of plurimediality, in contrast, as I will
stress, there are no orchestrators other than the individual consumers’ minds, and it is usually
harder to pinpoint anything that is clearly a “mother-ship” narrative. Henry Jenkins talks
about “balanced transmedia” as a form of transmediality that is similar to plurimediality but
argues that it is seldom realized.
13. Cf. Denson 2011.
14. Denson 2011.
15. As Denson (2011) has made especially clear, “original” portrayals have no particular
authority as a character becomes plurimedial. To return to Holmes as an example, his
calabash pipe was never mentioned by Doyle, who described him using only a clay pipe, a
briarwood pipe and a cherrywood pipe; the calabash was introduced by the actor William
Gillette, who portrayed Holmes on the stage in the early 1900s.
16. Plu. Thes. 36.4; cf. Ar. fr. 594 Edmonds and Pherecr. fr. 49.
17. Cf. Plu. Thes. 36.4, and see Kearns 1989: 120–124, 168–169. It is likely that Theseus
had formal cult sites in at least four other Attic locations, as well (Philoc. FGrH 328 F 18).
18. Pi. O. 3; Bacch. 17. On whether Bacchylides was the first to tell this story, see note 35
in Chapter 1.
19. Pi. O. 1.24–89.
20. Hes. frs. 37.10–15, 131a and b, 132–133 MW; Acus. FGrH 2F28; Pherecyd. FGrH
3F114, Bacch. 11.53–58, 92–95. See also Fowler 2013: 169–178; Gantz 1993: 311; Dowden
1989: 71–96; Seaford 1988.
21. Pi. I. 4.35–38.
22. Hdt. 2.44.
23. Plu. Thes. 1, Kilvert’s translation.
24. Stesich. frs. 192–193. Cf. Pi. O. 1.34–35.
25. On “diegetic,” see Chapter 3, note 88.
26. Lycaon: already glimpsed in Hes. frs. 161–168 and Pherecyd. FGrH 3F156; fuller story
at Eratosth. Cat. 8. See also Fowler 2013: 104–109, and Gantz 1993: 728. Croesus: Hdt.
1.46–51.
27. Poseidon in Ethiopia: Od. 1.25–30. Polyphemus calling to his father: Od. 9.526–535.

28. Zeus needs Prometheus’s advice: A. Pr. 167–195, 907–917, 944–953, and cf. Pi. I.
8.26a–45a, which puts Themis, traditionally an oracular goddess, in Prometheus’s role.
29. Persephone lured by flower: HHDem. 15–18. Hecate hears her: HHDem. 22–25, 54–
58. Helios tells who kidnapped Persephone: HHDem. 74–81. None of the gods heard her cry:
HHDem. 22–23. Zeus spares himself the sound: HHDem. 27–29. Persephone eats something:
HHDem. 371–374, 405–413.
30. Eileithyia in a cloud: HHAp. 97–101. Telphousa tricks Apollo: 254–274, 375–381.
31. HHHerm. 190–200.
32. Demeter distracted by grief: implied by the story that Pindar rejects at O. 1.27, fully
told first by Lyc. 152–155, and cf. Ov. Met. 6.403–411. Children of Helios forget fire: Pi. O.
7.39–48. Eos forgets to ask for youth: HHAphr. 218–238.
33. Il. 16.431–461.
34. Il. 5.330–352, 392–404, 439–442, 846–863.
35. Hermes uses special sandals: e.g., Il. 24.339–345, which is repeated at Od. 5.43–49
(and note that Athena has similar sandals at Od. 1.96–98). Hermes’s sandals are more
specifically described as winged in Hes. Shield 220. Hermes is represented as wearing
winged sandals already on vases dated between 550 and 520 BCE (for example, Musée du
Louvre, F 19 and F 30). See also Figures 7.2 and 7.3, in the latter of which Perseus, rather
than Hermes, wears the winged sandals because Hermes has lent them to him. Note, however,
that in Figure 6.3, Hermes’s feet are bare, and in Figure. 4.1, Hermes wears ordinary sandals.
Iris uses special sandals, too, on occasion, although she is usually described and represented
as being winged herself: discussion at Gantz 1993: 17–18.
36. Hades owns a cap of invisibility and Athena borrows it: Il. 5.844–845.
37. Athena shakes her aegis: Il. 2.450–452; and see Gantz 1993: 84–85. Hermes uses
special shoes: HHHerm. 75–86. Hermes throws them into the Alpheus: HHHerm. 138–140.
Hermes slips through the latch-hole: HHHerm. 145–147.
38. Boyer 2001: 87–90, drawing on Barrett 1996.
39. “Officially” includes what canonical texts say, but it also includes ideas that become
canonical in other ways. The Bible never describes Satan as having horns on his head, yet
artists’ renditions show him that way in stained-glass windows and other church art from the
medieval period onward; by now, therefore, the horns approach canonicity.
40. Boyer 2001: 87–88, drawing on Barrett 1996.
41. Boyer 2001: 59.
42. E. Ba. 1388–1392.
43. Pi. P. 3.27–29. Another example comes from one of the myths that Plato invents. When
Poseidon creates a precinct for his lover Clito, we are told that “he adorned the island at the
center with his own hand—in the way that gods easily do—causing two fountains to flow …”
(Pl. Cri. 113e2–6). This is typical of Plato’s myths, in which the gods are omnipotent as well
as morally unimpeachable.
44. Reicher 2010: 130. Reicher uses “vague” here in the sense that analytical philosophers
do, to mean something whose limits cannot be reliably defined, the classic example being a
cloud made up of water molecules. She is responding to scholars who have argued that each
instantiation of a character is an ontologically separate entity: thus, the Julius Caesar we see
in Shakespeare’s Julius Caesar is a separate character from the Caesar of Shakespeare’s
Antony and Cleopatra, for example.
45. As is well explored in Pirenne-Delforge and Pironti 2016.

46. See further Henrichs 1994, 1991.
47. Pi. P. 4; cf. O’Higgins 1997.
48. Barthes 1972: 110.
49. Notably, when Disney takes on a fairy tale whose main characters are nameless and
turns it into a ninety-minute feature film, it bestows personal names upon those characters, as
one step of giving them enough depth to carry the lengthier action. The Little Mermaid
becomes Ariel, Beauty (of Beauty and the Beast) becomes Belle (which, for the average
American, has no association with the word “beauty” and is just another moniker), the Snow
Queen becomes Elsa, the Prince in Sleeping Beauty becomes Prince Philip, Cinderella’s evil
stepmother becomes Lady Tremaine and the wicked fairy of Sleeping Beauty becomes
Maleficent—who becomes so well developed as a character, in fact, as to have merited a live-
action, megabudget backstory (Maleficent) fifty-five years later (Holmes 2014).
50. Jenkins 1992: 99.
51. See Leitao 1995.
52. Cf. Philippe Borgeaud’s study of the names of early mythic inhabitants of Athens and
Crete (2004: 46–64) and Marianne Govers Hopman’s different approach to mythic names
belonging to more than one individual (2012).
53. A. Ag. 160–176.
6 METAMORPHOSES
1. Gen. 19:26 and Ev. Luc. 17:32; Aitareya Brahmana 3.33–34, as taken from O’Flaherty
1988: 86–87; ponies: e.g., Boas 1917: 53 and Dorsey 1904: 294–295, 295n. Dorsey collected
a number of other Pawnee stories about people turning into animals, sometimes after having
sex with an animal.
2. Piraro 2012.
3. Ovide Moralisé 2.1365–1819, 10.3478–3795; see also Dimmick 2002: 264–287.
4. See Hopman 2012.
5. In addition to snaky-haired Medusa and bedogged Scylla, as examples of
transformational hybridity I can offer only Salmacis, who starts out as a woman but ends up
as a hermaphrodite (Ov. Met. 4.285–388).
6. Hes. Th. 270–336.
7. First mentioned, apparently, in the fragments of Bacch. 26; see Gantz 1993: 260–261.
8. Pi. P. 2.42–43; see also Gantz 1993: 718–721.
9. Hes. Th. 278–279.
10. Hydra, Cerberus and Geryon: Hes. Th. 288–289, 306–318. Callirhoe as ocean nymph:
Th. 351, 979. Cyclops: Od. 1.71–72.
11. Polyphemus: Od. 9.528–535.
12. Stesich. Ger. frr. 10–13. Another case is Scylla; at Od. 12.124–126, Circe advises
Odysseus that when he sails past Scylla, he should call out to Crataeis, the mother of Scylla,
asking her to hold back her daughter.
13. POxy 2461 = fr. 81 Aus.
14. HHAp. 351; Hes. Th. 824–828.
15. Hes. Th. 313–315, 328–329.
16. Hes. Th. 270–336; cf. Gantz 1993: 19–25.
17. Cf. West 1966: ad loc.

18. We get the whole story from Eratosthenes (Catastr. 8), who cites Hesiod as a source (=
fr. 163 MW).
19. One reason that the story was told was that it could be used to explain a cult established
near Mount Lykaos in Arcadia (see Pl. R. 565d; Paus. 8.38.7; Pliny HN 8.81 and discussion in
Buxton 1987 and Burkert 1983: 83–10). From an early time, however (apparently already
Hes. frr. 161–168; see Fowler 2013: 104–109 and Gantz 1993: 728), the story was well
known outside of Arcadia, suggesting that its appeal was not due only to the fact that it
explained the existence of a local cult.
20. Iynx: Call. fr. 685 (= schol. Theocr. 2.17); cf. schol. Pi. N. 4.56a and Forbes-Irving
1990: 243–244. Anthos: Ant. Lib. 7, citing Boeos.
21. Io turned into cow by Hera: A. Supp. 299. Io as priestess of Hera: Hes. fr. 124 MW and
later sources including A. Suppl. 291–293. Io equated with Isis: first with certainty Call. Epig.
57 but perhaps also Hdt. 1.1. Cf. also Fowler 2013: 235–248; Gantz 1993: 198–203; Forbes-
Irving 1990: 212–215; Seaford 1980.
22. Artemis transforming Callisto and Callisto as mother of Arcas: Eratosth. Cat. 1, citing
Hes.; Amphis fr. 47. See also Fowler 2013: 107–108.
23. Lycian herdsmen muddying water: Ant. Lib. 35, crediting Nicander and Menecrates the
Xanthian (FrGH 769F2); Ov. Met. 6.339–381.
24. Midrash Aggadah [Buber] Gen. 19:26.
25. Zeus turning Io into a cow: Apollod. 2.1.3–4, probably drawing on Hesiod; see Gantz
1993: 199–201; Fowler 2013: 235–248.
26. Zeus turning Callisto into bear: Apollod. 3.8.2. Hera turning Callisto into a bear: Call.
fr. 632 Pf. and later sources, including Ov. Met. 2. 476–488 and Fast. 2.177–178. Cf. Fowler
2013: 107–108.
27. On playing the bear, with full ancient sources and citations to earlier scholarship, see
Faraone 2003.
28. Our sources for Mintha are late but numerous enough that it is hard to imagine that the
story was not popular from an early time: Oppian H. 3.486; schol. Nic. Alex. 375; Photius s.v.
Mintha; Strabo 8.3.14. Ov. Met. 10.728 hints at it. See also Detienne 1977: 72–98.
29. Ov. Met. 6.1–145.
30. Discussions of the stories to be analyzed in this section can be found at Fowler 2013:
365–366; Levaniouk 2008; Johnston 1994; Gantz 1993: 239–241; Forbes-Irving 1990: 248–
249; and Burkert 1983: 201–207.
31. Od. 19.518.
32. See, e.g., Parth. 11, quoting Apollonius Rhodius (nightingales mourn Adonis).
33. Hes. Op. 568; Sapph. 135 LP, cf. 136 LP.
34. Swallow’s song as barbarian: LSJ s.v. chelidōn.
35. Hes. fr. 312 MW = Ael. VH 12.20.
36. Od. 19.518–523; Pherecr. fr. 124 = schol. Od. 19.518; Paus. 9.5.9.
37. Helladius ap. Phot. Bibl. 531; Ant. Lib. 11.
38. A. Suppl. 58–65 and Ag. 1144–1145; S. fr. 585; Th. 2.29; Ov. Met. 6.424–676;
Apollod. 3.14.8; Paus. 1.5.4, 1.41.8.
39. Ant. Lib. 11.
40. Helladius ap. Phot. Bibl. 531.
41. E.g., Od. 19.518–523.
42. S. fr. 585; Th. 2.29; Ov. Met. 6.424–676; Apollod. 3.14.8; Paus. 1.5.4, 1.41.8.

43. This meaning cannot be correct (“Philomela” would more accurately mean “lover of
sheep”) but it would have been heard nonetheless.
44. The cutting of the tongue appears first in Sophocles but is developed most elaborately,
and most famously, in Ovid.
45. Hyg. Fab. 45; Serv. Ec. 6.78 (both of whom omit the tongue cutting); Myth. Vat. I 4
and II 261 (who retain it).
46. Eust. Od. 1875.15–27 and Od. 19.518.
47. Doniger 1996: 119.
48. D. 60.30.
49. The meaning of Hecabe’s transformation has been treated by Franco 2014: 109–116;
Buxton 2009: 58; Burnett 1994; Forbes-Irving 1990: 207–210.
50. E. Hec. 1259–1273.
51. Nic. fr. 62 = schol. E. Hec. 3.
52. Lyc. 333.
53. Lyc. 1174–1188.
54. Ov. Met. 13.565–575.
55. Q. S. 14.347.
56. Apollod. Ep. 5.23.
57. Some but not all of them manage to connect Hecabe’s transformation, or at least her
burial, with the Cynossema on the Chersonese—but that may well have been Euripides’s
innovation. (Notably, moreover, there were plenty of other Cynossemata in antiquity that
didn’t manage to attract such a myth.)
58. PMG 965.
59. Il. 24.212–213.
60. See also Franco 2014 for the full range of associations.
61. E.g., Burnett 1994.
62. E. fr. 968 Nauck.
63. Ar. fr. 567 Edmonds: “I will become a nasty-tempered bitch, the agalma of light-
bearing Hecate.”
64. Lyc. 1174–1188.
65. Johnston 1999: 203–215.
66. E.g., Od. 20.17–18; Semon. fr. 7 West lines 32–36. I thank Tom Hawkins for pointing
this out. See also Franco (2014: 108), who understands the trope as a more strongly negative
one than I do, and Burnett (1994), who argues that the canine transformation is Euripides’s
own development on Hecabe’s previously existing reputation as a mother.
67. Buxton 1987.
68. Pi. O. 1.61–64.
69. Nic. Ther. 12a.
70. Theophilus can be dated to the Hellenistic period—probably the late fourth and early
third centuries. For more on this, see Johnston 2009.
71. As we learn both from Nicander and from other ancient writers, the term phalangion
could designate any member of a whole family of venomous spiders. In some discussions,
ancient authorities distinguish between phalangia and non-venomous spiders, to which they
apply the term arachnai, but other authorities understand both phalangia and non-venomous
spiders to be subgroups of a more inclusive family, to all of which the term arachnai can be
applied. Further on methods of categorization, see Bevis 1988: 34–35; Scarborough 1979.

72. Gibson 1979. The example of the stick is mine.
73. Bettini 2013: 123–230.
74. Of course, the original meaning of symbolon relies on this restricted one-to-one
relationship; see Struck 2004: 78–84. The essentialism of symbols was developed by
Neoplatonists; subsequently, during the Middle Ages and the Renaissance, a greater range of
potential meaning might be admitted to a given symbol, but in any particular instance, its
meaning was set by the artist or author. Works such as Vincenzo Cartari’s Le imagini degli dei
antichi (ca. 1400) were handbooks intended to aid in the assignment of proper meaning (cf.
Graf 2008: 153–157).
75. Except where noted, for the rest of this discussion, I subsume both arachnai and
phalangia under the word “spider.”
76. E.g., Hes. Op. 777; Arist. HA 622b23; Ael. NA 1.21; further at Bevis 1988: 39.
77. Although this could have a negative valence (Odysseus’s marital bed is said to be
covered with webs after his twenty years of absence), it could also be positive (Bacchylides
describes peace as a time when shields are covered with webs). Bacch. fr. 4.69–70; Od. 16.35.
See also Hes. Op. 475; E. fr. 369 Nauck; Pherecr. fr. 142; S. fr. 264; Theoc. 16.96; Philostr.
Imag. 2.28.2; further at Bevis 1988: 40.
78. Od. 8.280; Arist. HA 623a8; AP 9.372; Plut. Mor. De soll. an. 966e–f; Philostr. Im.
2.28.1; Plin. HN 11.79–82; cf. Paus. 6.26.7. “A spider-like thread” (arachnaios mitos)
proverbially meant a “very fine thread,” e.g., AP 6.39.3.
79. Arist. Ph. 199a20–22; Ael. VH 1.2 and NA 1.21 (but cf. NA 6.57, which is more
nuanced); Pliny HN 11.80.2; Sen. Epist. 121.23. Cf. Feeney 1991: 193–194. Arist. HA
622b28–623b1 distinguishes amongst different types of spiders, some of which spin webs
that are sloppy and crude and others of which spin webs that are clever and polished. Plut. De
soll. an. 966e–f lauds the fineness of the thread and regularity of the weaving but notes that
there is no warp—i.e., he confirms its simplicity even as he admires it. Pliny, on the other
hand, mentions both warp and a woof (tela and subtemina) at HN 11.80. Further at Bevis
1988: 39.
80. E.g., Pl. Com. fr. 22 line 2 Kock; many other citations, mostly from later antiquity, are
offered at Bevis 1988: 39–40. Cf. Plut. De Is. et Os. 358f, where the spider’s web is compared
to hasty, poorly developed thoughts.
81. A. Ag. 1492 and 1516, Suppl. 887; Xen. Mem. 3.11.6; AP 9.372; Philostr. Im. 2.28.3–4.
82. Arist. HA 555b10–15, 555a23–26 (where he uses arachnai, not phalangia); cf. Antig.
Mir. 87 and schol. Nic. Ther. 715a; Plin. HN 11.85. In the fuller version of this discussion at
Johnston 2009, I include a discussion of spider cannibalism based on modern arachnological
data. (I thank the arachnologist Daniel Gloor for his help with information concerning real, as
opposed to mythic, spiders.)
83. E.g., Pliny HN 24.61–63; Ael. NA 17.11; schol. Nic. Ther. 721–724; Paul. Aeg.
Epitomae medicae libri septem 5.6t1; [Ps.] Dioscorides De iis 4; Philum. Ven. 15.6; Aët.
13.20; Eutecnius Soph. Paraph. Nic. Ther. 59.
84. E.g., Scarborough 1979: 7–8. Species are found throughout the world; the American
variety is the black widow (Latrodectus mactans [Fabricius, 1775]).
85. Scarborough 1979: 8n70; websites for physicians mention it, e.g.,
http://precordialthump.medbrains.net/2008/12/06/problems-in-toxicology-003/and
http://medbrains.net/tag/toxicology/.

86. Further at Johnston 2009; Von Staden 1993: 36–37; Scarborough 1979: 3–6; Gow and
Scholfield 1953: 18.
87. The basic meaning of phalang- is “beam” or “plank” (IE bhelg-); from this, it comes to
refer to a number of things that are long and relatively slender. Most notably, the word
phalanx can also mean a finger or toe bone. The application of the words phalanx and
phalangion to spiders is established by the Classical period (e.g., Ar. Ra. 1314, Pl. Euthd.
290a and Xen. Mem. 3.11.6), but it is unclear why; perhaps it is because their legs, which
have two joints, look like fingers (this is especially so in the Latrodectus genus, in which the
legs tend to be longer and more slender than in other genuses). Ovid appreciates this
similarity: Met. 6.143 (cf. Ar. Ra. 1314; Ov. Am. 1.14.7; AP 9.372).
88. The significance of weaving within Athenian society has been well studied by several
scholars, most importantly by Scheid and Svenbro 1996. I will not repeat the details of their
analyses here but will instead summarize those of their conclusions that will be of the greatest
significance for the present discussion.
89. The fact that words for the warp have masculine connotations and words for the woof
have feminine connotations adds strength to the metaphor: Scheid and Svenbro 1996: 13.
90. Further on all of this at Johnston 2009 and Scheid and Svenbro 1996.
91. Further examples at Johnston 2009.
92. Reviews of the issues and evidence in Barber 1992 and Burkert (1966) 2001.
93. Our main source is Paus. 1.27.3; further at Goff 2004: 198–205; Kearns 1989: 21–27;
and Burkert (1966) 2001.
94. Barbara Goff (2004: 198–205) offers an attractive interpretation that downplays the
secrecy of the myth and ritual; her differing conclusions do not affect my argument here.
95. Phot., Hsch. and Sud. s.v. protonion; Phot. s.v. Plynteria; and Hsch. s.v. Kallynteria
and Plynteria.
96. Pl. Lg. 833d–e; cf. La. 181e–182a, Grg. 456d. See also Kyle 1992: 87–88; Wheeler
1982.
97. Kyle 1992: 88–89.
98. Further on the complementarity of the roles at Johnston 2009.
99. According to I. Agnarsson, director of the Zoology Museum at the University of Puerto
Rico, only twenty to twenty-five species of spiders (out of about 39,000) are “quasi-social”;
most of these are tropical. See Agnarsson Lab 2017.
100. Arist. HA 542a12–17. If Plutarch’s opinion that a spider’s webs were all woof and no
warp was shared, then this, too, would have signaled failure on the spider’s part to weave
opposites into a single whole (see note 79 above).
101. Purg. 12.43–45.
102. Buxton 2009.
103. A. Supp. 565–570, Pr. 687–692. Cf. also, for example, Il. 2.308–320. Buxton (2009:
29) comments on passages in which characters express their astonishment at a god’s
transformation, e.g., Od. 3.371–373.
104. E.g., Corinna sang of the daughters of Minyas (fr. 665) and Bacchylides of Io and
Niobe (19 and fr. 20D).
105. E.g., Pherecydes told of Aedon (FrGH 3F124) and the Hyades (fr. 90); Acusilaus told
of Io and Actaeon (FGrH 2F26, 27, 33); Menecrates told of Leto turning peasants into frogs
(FGrH 769F2); several authors told of Niobe (see Fowler 2013: 366); and Deilochus told of
how Cleite’s tears turned into a spring named after her (FGrH 471F6).

106. The vase showing pirates turning into dolphins recently was returned to Italy by the
Toledo Museum of Art (Toledo 1982: 134); its current whereabouts have not been disclosed.
It is Etruscan and has been dated to 510–500 BCE. It is attributed to the Painter of Vatican
238. The vase showing a horned Actaeon in the Underworld is Toledo 1994: 19, RVAp. Supp.
2, p. 508 18 / 41a1, attributed to the Darius painter and dated to 340–330 BCE.
107. For a study of the two Scyllas that aims at showing that they developed from a single
collocation of ideas, see Hopman 2012.
108. Verg. G. 4.246–247; Bacon, The Wisedome of the Ancients (1619) (STC 1130), X.
109. Updike 1963: chap. 1.
110. The earliest source for Callisto as the mother of Arcas is Charon of Lampsacus FGrH
262F12, but the story probably goes back to Hesiod; a fuller story is at Paus. 8.3.6–7. See also
Fowler 2013: 105–108; Gantz 1993: 725–729.
111. D. 60.30.
112. Ant. Lib. 17.
113. Od. 6.229–231.
114. The story seems to be as old as Hesiod and Pherecydes. See Hes. fr. 68 MW;
Pherecyd. FGrH 3F98, 99 and cf. discussions at Fowler 2013: 195–201 and Gantz 1993:
176–180. A complete version of the story appears at Apollod. 1.80–83.
115. For an overview of Asclepius at Epidaurus and similar healing shrines, see Johnston
2008: 90–95; for specific instances, see the section of Chapter 3 entitled “Ancient Narrations
of Remarkable Incidents.”
116. Apollo: Cic. Div. 1.37; D.S. 22.9; SIG 398; Pomp. Trog. 24.8; Paus. 1.4.4 and
10.23.4–6; and cf. Call. H. 4.171–184; Demeter and Persephone: Plut. Tim. 8.
2
117. St. Elmo’s fire: HH 27; Alc. fr. 34; Isyllus’s Paean to Apollo and Asclepius = IG IV
128; and see Furley and Bremer 2001: 1.227–239, 2.180–192. Demeter: Rehm and Harder
1958: no. 496. See also Henrichs 2010; Bravo 2003; Graf 2003; Versnel 1987.
118. Nelson 2001. The specific reference is to page viii, but the topic is one of the foci of
the book as a whole.
119. D.L. 59.
120. HHDem. 243–249, 305–313.
121. IG XI.4 1299; see also Engelmann 1975.
122. Ov. Met. 2.484–488, 3.200–201; Miller’s translations, slightly altered.
123. Il. 24.617; A. Pr. 673–674; E. fr. 930 Nauck.
124. A. Pr. 673–674.
125. E. Hipp. 732.
126. E. IT 1089–1092.
127. E.g., Callisto at Ov. Met. 2.485; Actaeon at Ov. Met. 3.203.
128. Ov. Met. 15.158. At 167, he calls it the spiritus instead; at 172 and 175, he again uses
anima.
129. Bremmer 1983: 24–53.
130. Johnston 1999.
131. Io and Odysseus’s men are rare exceptions, as is, in later times, the hero of Lucian’s
Ass and Apuleius’s Golden Ass. Tiresias has a similar experience, given that he is turned from
a man into a woman and then back again in some versions of his story: Hes. fr. 275 MW and
see Gantz 1993: 529–530.

132. Mestra’s full story is at Ov. Met. 8.738–878. Mestra as lover of Poseidon: Hes. fr. 43a
MW; as shape-shifter: Palaeph. 23, Lyc. 1391–1396; see also Gantz 1993: 68–69.
Periclymenus: Hes. frs. 33a and 35 MW; see also Gantz 1993: 184–185.
133. Hdt. 4.105; Pl. R. 565d; Paus. 8.38.7; Plin. NH 8.81. Discussion at Buxton 1987 and
Burkert 1983: 83–10.
134. HHAp. 400–450.
135. Kemmerer 2011: 43, citing Erdoes and Ortiz 1984: 392.
136. Kemmerer 2011: 43, citing Brown 1979: 22.
137. Valladolis and Apfell-Marglin 2001: 639–670; cf. Kemmerer 2011: 43–44.
138. I thank my former student Karen Ravelli for discussing this idea with me.
139. Ov. Met. 2.485–494.
140. Atrahasis Tablet III; Epic of Gilgamesh Tablet XI; Gen. 6–9. The Hindu myth of the
Flood, the hero of which is Manu, is from the Matsya Purana 1.11–34, 2.1–19 (well narrated
in translation by Doniger 2004: 181–185).
141. For a complete account, we must go to Apollod. 1.7.2, but it is clear that the basic
story was around already in the Archaic period. We get fragments of it from Hellanicus
(FGrH 4F117a) and Pindar (O. 9.44–53). Andron fr. 8 and Epicharmus fr. 113 each explicitly
mention the larnax. Later, Ov. Met. 1.319 calls the craft a parva ratis, “tiny boat.” See also
Mark Smith 2014; Fowler 2013: 113–121; and Bremmer 2008.
142. Ov. Met. 1.416–421.
143. Hes. Th. 535–564.
144. For complete citations and discussion of the ancient sources for the Buphonia and its
myth, see Burkert 1983: 136–143 (although his tendency to export conclusions about the
Buphonia to the rest of Greek sacrifice are overdone). The myth seems to be at least as old as
the late fifth century.
145. Clem. Al. Protr. 2.17; schol. Lucian 275–276 (Rabe).
146. See Pirenne-Delforge 1994: 388–393.
147. From the Shatapatha Brahmana, as cited in O’Flaherty 1988: 84.
148. Telephus: Apollod. 2.7.4; S. fr. 89; Ael. VH 12.42; Hyg. Fab. 252. Peleus and Neleus:
schol. Il. 10.334; Eust. ad Od. 11.253, p. 1681; Ael. VH 12.42. Atalanta: Apollod. 3.9.2 and
Ael. VH 13.1. Paris: Ael. VH 12.42.
149. Cf., too, the story of the Iranian hero Zal as told in the Shahnameh. Zal is abandoned
at birth and raised to adulthood by a bird. When he is a young man, they part, but the bird
later saves the life of Zal’s wife, Rudaba, when she is giving birth to the hero Rustam.
150. E.g., Pi. O. 6.44–47; Anticlides FGrH 140F17, POxy 56.3830.
151. Apollod. 3.3.1–2.
152. In Vergil’s fourth Eclogue, we come close—but the paradise is a paradise by virtue of
the lions staying away from the flocks and the serpents dying.
7 HEROES
1. Ovid, in his Metamorphoses—and in many cases, probably, Ovid’s Hellenistic sources
—showed a greater interest in ordinary people than did poets who narrated myths during the
Archaic and Classical periods. For example, Baucis and Philemon, who are a poor, elderly
Phrygian couple, become the stars of one of Ovid’s myths (Met. 8.620–724) and his Arachne
is set apart from other women only by her extraordinary skill at weaving (Met. 6.5–145).

Ascabulus, who is turned into a gecko by Demeter, begins as an ordinary boy, remarkable
only for his bad manners (Met. 5.446, cf. Ant. Lib. 24, which follows Nic. Ther. 484).
2. Ganymede: the earliest story makes it a whirlwind that carries the boy away (HHAphr.
202–217); the eagle first appears in the fourth-century-BCE poet Leochares, as cited by Pliny
HN 34.79, after which it becomes a common element in the story: Gantz 1993: 559–560.
Prometheus: Hes. Th. 523–525. Argus: Od. 17.290–327. Arion: Hdt. 1.23–24.
3. At least parts of the story are found already in Hes. Th. 280–281 and fr. 135 MW of his
Catalogue. Full discussions at Fowler 2013: 248–259 and Gantz 1993: 399–311.
4. Thymoetes: Hellanic. FGrH 4F*38; Diod. Ath. FGrH 372F28; Paus. 2.18.9; Demon
FGrH 327F1; Nic. Dam. FGrH 90F48. See also Kearns 1989: 170.
5. Paus. 3.12.5.
6. Thoricus: Hsch. s.v.; Lupu 2005: 115–149 (number 1 = SEG xxxiii 147); Kearns 1989:
169.
7. Sphinx: first attested in Corinna fr. 672 and A. Th. 773 but alluded to, perhaps, at Hes.
Th. 326–327. Further at Gantz 1993: 490–506.
8. Hes. fr. 10a.20–24 MW; E. Ion; Hdt. 8.44. Further at Gantz 1993: 244–245 and Kearns
1989: 108–110, 174–175.
9. Menelaus’s afterlife: Od. 4.561–569. Generally on Menelaus: Gantz 1993: 564–547,
572–664.
10. I give a list of these oracles at the end of Johnston 2005.
11. During the Hellenistic period, however, the designation and status of “hero” began to
be applied to some amongst the recently dead—particularly rulers and other prominent
people. See most recently Jones 2010.
12. Farnell 1921: 23–30.
13. D. 60.31; Apollod. 1.8.3; Paus. 1.5.2, 2.4.3; D.S. 4.37; cf. Kearns 1989: 80–91, 149.
14. Hdt. 6.52, 7.204, 8.131.
15. On this, see Lyons 1997; Larson 1995; Dowden 1989.
16. I do not include the Homeric Hymns in my category of epics. Although they share
meter and diction with epic poems, their function as hymns was clear already in antiquity.
Discussion of the problems—but ultimate necessity—of distinguishing hymns from epics in
Clay 1989: 3–16.
17. Geryoneis = frs. S7–87 as in David Campbell’s Loeb edition, which uses the order also
found in the S.L.G. Oresteia = frs. 210–219 of the Loeb. Stesichorus apparently had some
interest in the early days of the gods as well; he described Typhon as the son of Hera, who
bore him to spite Zeus (fr. 239). On his invention of the hymn: Clem. Al. Strom. 1.16.78.5 (ii
Stählin).
18. Perseus: fr. 543 PMG; Jason: frs. 544–548 PMG; Theseus: fr. 550 PMG.
19. Helen and Menelaus: Bacch. 15, Heracles: 16, Theseus: 17 and 18, Io: 19, Idas: 20,
Cassandra: 23, Meleager: 25, Pasiphae: 26, Achilles: 27, Orpheus: 28.
20. Pi. dithyrambs 1 and 4 (Perseus), 2 (Heracles) = frs. 70a–d.
21. Myrtis: fr. 716 PMG; Corinna: fr. 671 PMG.
22. There were some other comedies that starred Dionysus, the god of theater, as well:
Dover 1997: 39.
23. The titles of lost tragedies suggest a few more that may have focused on gods: from
Aeschylus, the Prometheus Unbound, the Nurses of Dionysus, the Daughters of Phorcys, and
the Daughters of Helios; and from Sophocles, the Inachus and the Muses.

24. Heracles and the ghost of Meleager: 5, the daughters of Proetus: 11, the Trojan War:
13, Croesus’s story: 3, the Telchines and their daughters: 1.
25. Castor and Polydeuces: frs. 509 and 510; Heracles: 509.
26. The others are Pi. P. 2 (Ixion’s pursuit of Hera and Zeus’s punishment of him), P. 3 (the
conception and birth of Asclepius, in which Apollo plays a large role), P. 5 (Apollo as
founder of Cyrene, as healer and as prophet), P. 7 (Apollo’s foundation of Delphi), P. 12
(Athena’s invention of the aulos) and I. 8 (Zeus’s seduction of Aegina and Zeus and
Poseidon’s quarrel over Thetis).
27. P. O. 7.35–38 and fr. 34 SM; earlier vases seem to refer to Hephaestus’s role as well:
Gantz 1993: 51–52, 78. In a few cases, these stories come down to us only through those
glancing treatments: without Pindar, we would know nothing of how and why Helios
received the island of Rhodes as his special realm (P. O. 7.54–76), for example.
28. In these Homeric Hymns, indeed, the gods themselves are often portrayed much as
heroes would be: we hear of them battling a giant serpent, stealing cattle and, in the case of
female divinities, disguising themselves to serve as the nursemaid or lover of heroic humans.
It is possible that, in the Homeric Hymns’ role as prooimia for the performance of episodes
taken from epics, they not only borrowed meter and diction from epics but also leaned
towards selecting stories of the gods that were typologically similar to those found in epics.
29. Alc. fr. 307. See also Men. Rh. (iii.340), who mentions “genealogical” hymns by
Alcaeus and other poets, describing the birth of various gods.
30. E.g., Pl. R. 2.381d1–e6, Lg. 10.887c5–e1.
31. For instance, Timotheus (an Eleusinian exegete) narrated a cult myth for the mysteries
of the Mother: Paus. 7.17.10–12 and Arn. 5.5–7 with comments at Burkert 1987: 73; the
Parian Chronicle (= FGrH 239) included a history of divine and human deeds stretching from
the early days of the gods down to 299 BCE, including, for instance, entry no. 3, which notes
the dispute between Poseidon and Ares that led to the foundation of the court at the
Areopagus, and entry no. 12, Demeter’s invention of seed corn. See also Jacoby’s
introductory comments to the fragments of the local historian Philochorus of Athens (FGrH
328) = IIIB vol. 1. 226; Jacoby argues that Philochorus treated the myths of the gods
systematically.
32. This is implied by a number of ancient sources that claim that the women explain their
actions at the Thesmophoria with reference to the story (e.g., D.S. 5.4.7 and Apollod. I.5.1)
and by Call. Hymn 6.1–6 (his hymn to Demeter), where women on their way to the
Thesmophoria begin to tell the story of Demeter and Persephone and then switch to the story
of Demeter and Erysichthon, lest the first story sadden Demeter.
33. Il. 6.130–141, 24.602–609; A. Ag. 160–176; E. Hel. 1301–1366 and Ion 985–988.
34. E.g., Darshan 2013a, which includes a sketch of recent controversy over the topic at
517–518, and cf. Darshan 2013b; Schüle 2009. Bruce Louden (2011: 10–15) describes the
relationship between Greek myths and the Old Testament as “dialogic,” emphasizing that
influence went both ways. Brian Doak (2012) argues that a group of figures in the Old
Testament who are associated with giants share features both with the giants of Greek myth
and with Greek heroes, and that both the Israelite and Greek traditions draw on a “pan-
Mediterranean style of religious thought regarding heroic warriors and their fate and meaning
after death,” which he suggests was anchored in Canaan (esp. 195–199).
35. Enuma Elish = Dalley 1989: 228–277; Anzu = Dalley 1989: 203–227.

36. Lugul-e and the Angim: ETCSL at http://etcsl.orinst.ox.ac.uk/cgi-bin/etcsl.cgi?
text=t.1.6.2# and http://etcsl.orinst.ox.ac.uk/cgi-bin/etcsl.cgi?text=t.1.6.1#.
37. Theogony of Dunnu = Dalley 1989: 278–281.
38. Descent of Ishtar (Inanna) = Dalley 1989: 154–162 and
http://etcsl.orinst.ox.ac.uk/section1/tr141.htm.
39. Inanna and Enki = Farber 1997.
40. Enki and Ninhursag = ETCSL: http://etcsl.orinst.ox.ac.uk/cgi-bin/etcsl.cgi?
text=t.1.1.1#.
41. Enki and Ninmah = Klein 1997.
42. Nergal and Ereshkigal = Dalley 1989: 163–181.
43. Erra and Ishum = Dalley 1989: 282–316.
44. Kumarbi Cycle = CTH 344; The Disappearance of Telipinu = CTH 324; Story of
Illuyanka = CTH 321; Song of Hedammu = CTH 428; Telipinu and the Daughter of the Sea
God = CTH 322; Inara = CTH 366; Kamrusepa = CTH 457.1; The Disappearance of
Hannahanna = CTH 334.
45. Epic of Gilgamesh Tablet I, i and ii.
46. Lugalbanda in the Wilderness and The Return of Lugalbanda = Vanstiphout 2003: 97–
166. The two other poems, Enmerkar and Ensuhgirana and Enmerkar and the Lord of Aratta,
are at Vanstiphout 2003: 23–96. See also the relevant ETCSL translations:
http://etcsl.orinst.ox.ac.uk/cgi-bin/etcsl.cgi?text=t.1.8.2.1# (and t.1.8.2.2#, t.1.8.2.3#,
t.1.8.2.4#).
47. Dalley 1989: 1–38.
48. Dalley 1989: 189–202.
49. Dalley 1989: 154–162; and ETCSL: http://etcsl.orinst.ox.ac.uk/section1/tr141.htm.
50. “The Birth Legend of Sargon”: Westenholz 1997: 39–49, esp. 41 and cf. 35.
51. “The Sargon Legend,” segment B: ETCSL:
http://etcsl.orinst.ox.ac.uk/section2/tr214.htm.
52. The Res Gestae Sargonis: Westenholz 1997: 57–139; with The Chronicle of Early
Kings, Tablet A, lines 20–23, as at Glassner and Foster 2004: 270–271.
53. “Erra and Naram-Sin”: Westenholz 1997: 189–201; “The Curse of Agade”: ETCSL:
http://etcsl.orinst.ox.ac.uk/cgi-bin/etcsl.cgi?text=t.2.1.5#.
54. “The Kirta Epic” and “The Aqhatu Legend”: Pardee 1997. Mark S. Smith (2014: 154)
has suggested that it was in the interest of the Ugaritic monarchy to recall important kings and
warriors of the past, to whom they paid cult under the collective name rephaim (more on this
shortly), but in terms of full-blown storytelling, we have only the tales of the warriors Danel
and Aquat and of King Keret.
55. CTH 7.1, 360, 363, 361. All four stories are found in Hoffner 1990: 62–68.
56. CTH 789; López-Ruiz 2018: sec. 4.2.
57. CTH 321; Hoffner 1990: 11–12.
58. Brian Doak (2012: 187) has conjectured that there was once a larger corpus of Ugaritic
heroic epic, which does not survive, describing the exploits of the rp’um (= rephaim, warrior
ancestors from the distant past).
59. “The Tale of Sinuhe” and “The Tale of the Shipwrecked Sailor” can be found at
Parkinson 1997: 21–53 and 89–101. See also Pinch 2002: 86.
60. Birth of triplets: “The Tale of King Cheops’ Court”: Parkinson 1997: 102–127. “The
Tale of the Herdsman”: Parkinson 1997: 287–288. See also Pinch 2002: 16–17.

61. “The Tale of Two Brothers” can be found in Maspero and El-Shamy 2004: 1–16.
“Inpu” is a variant of “Anubis” and Bata was the city god of Saka; he was sometimes
understood as a manifestation of Seth.
62. Pinch 2002: 43–44.
63. I am not counting the Egyptian “Tale of Sinuhe,” “Tale of the Shipwrecked Sailor,” or
“Tale of the Two Brothers,” for the reasons given earlier, or the tale of Setna because of its
late date and likely indebtedness to Greek myths. I have otherwise erred on the side of
generosity in what I am counting as a hero myth from the Near East. For example, I count all
four stories about the conflicts between the kings of Uruk and Arata, although only the two
about Lugalbanda include significant aspects of the marvelous. To reflect the popularity of
stories about Gilgamesh, and the varying ways in which they circulated, I have counted the
Epic of Gilgamesh as one story and two Sumerian stories that independently circulated about
Gilgamesh (The Envoys of Akka and The Great Bull Is Lying Down) as two more (thus, I
count “Gilgamesh” as three). Similarly, I count each story about Sargon and Naram-Sin
separately.
64. My arguments here about the relative abundance of hero myths in Greece and the
relative paucity of them in cultures of the ancient Near East is not meant to refute the
convincing arguments that many scholars have made for the influence of Near Eastern
narratives upon the Homeric epics or other early Greek myths. See, for example (to take only
three notable recent works), Louden 2011, 2006; López-Ruiz 2010. Louden’s emphasis on the
dialogic nature of interaction between Greek myths and Near Eastern myths is especially
important to remember; influence traveled both ways (Louden 2011: 10–15). The decades-
long dedication to this topic of Walter Burkert and M. L. West was extraordinarily important
and a list of their relevant publications would be too long to include here; I will mention only
West 1999 and Burkert 1992.
65. Gunnel Ekroth (2015) provides an excellent overview and up-to-date bibliography for
this large topic. Ekroth (2002) and many of her other articles are also foundational resources.
66. Doak (2012: 164) notes a recently discovered eighth-century stele from Sam’al (part of
the Hittite empire) with an Aramaic inscription that implies the notions that “the dead live on
in specific locales (i.e., at the monument), periodic feasting at the tomb and the designation of
monuments only for certain, important individuals.”
67. Mark Smith (2014) discusses this group particularly in his fifth and twelfth chapters;
see 145 for a summary description of them. The word rephaim enters the Hebrew Bible as a
term for “giant,” i.e., threatening outsider, as discussed in M. Smith 2014: 314–232 and Doak
2012.
68. Cf. Antonnacio 1994: 409–410.
69. See, for example, Assmann 2009.
70. Cf. Doak 2012: 198.
71. I thank Irad Malkin for this suggestion.
72. Dalley 1989: 51 = tablet I, ii. I am not counting the three Egyptian kings who were
sons of Ra in the story mentioned earlier because the formal attribution of a king’s paternity
to Ra was a standard practice in Egypt.
73. Volsungsaga, chap. 1. Other sources, such as Snorri Sturluson’s Edda, made Odin the
father of descendants who became the first kings of Scandinavian countries. Some of these
figures were originally thought of as gods themselves, but Snorri, good Christian that he was,
euhemerized them: for example, Yngvi (= Frey), from whom the kings of Sweden are

descended; Veggdegg, who ruled over East Saxony; Beldegg (= Balder), who ruled over
Westphalia; Skjöldr, from whom the kings of Denmark descend. Other figures, such as Frodi,
a descendant of Skjöldr, are more clearly mortal kings who in many cases have at least a foot
in historical reality. Some of their stories include extraordinary elements—for instance, the
fifth book of the Gesta Danorum tells us that a wicked woman turned herself into a sea cow
and gored King Frodi to death. The Danes concealed his death by embalming his corpse and
carrying it around in a cart for three years.
74. Mahabharata I (Adi Parva), part 6 (Adivansavatarana Parva). But Hinduism primarily
works to create an ontological category that falls between gods and humans in a different
way, positing that each of the gods has numerous avatars, or incarnations, some of which are
human. Thus Rama, a prince who is the protagonist of the Ramayana, is an avatar of Vishnu,
for instance, and his friend Hanuman is an avatar of Shiva. Even if we decide to count such
human avatars amongst the heroes, there are more narratives about the Hindu gods and their
affairs than about the heroes. The two closest things that India has to heroic epic—the
Mahabharata and the Ramayana—weigh in at a combined 140,000 verses or so, but the
eighteen major Puranas and eighteen minor Puranas, which take myths about the creation of
the cosmos and the affairs of the gods as their main foci, come in at about 400,000. And, as in
Greek narratives, tales about the gods are also embedded in stories that are mainly about
heroes, most notably the Mahabharata.
75. For a more complete list, see later in this chapter.
76. Epic of Gilgamesh, Tablets III–VI.
77. Calvert Watkins (1995: 297–518) gives a detailed analysis of Indo-European poetic
formulae for slaying dragons, with many examples of dragon slayers along the way. Almost
all of them who are not Greek are divine rather than human. On Thraetoana, see 313–320.
78. Shahnama 1.5.1.
79. Ramayana, Book 6 (the Yuddha Kanda), chap. 108.
80. Gregory 1902: 9–10, 69–70, 75–81.
81. Beowulf, lines 766–823, 1492–1650, 2712–3182.
82. E.g., Volsungsaga, chap. 18.
83. Cat: This story is told in the document now called the Prose Merlin; an easily
accessible text is at the University of Rochester’s database The Camelot Project:
http://d.lib.rochester.edu/teams/text/conlee-prose-merlin-defeat-of-lucius-and-arthur-and-the-
devil-cat. Boar: for an overview of the several boars against whom Arthur and his knights
fight, with links to texts, consult the Rochester database at
http://d.lib.rochester.edu/camelot/theme/boar. Further Arthurian foes, with links to texts, can
be found in this database as well.
84. Epic of Gilgamesh, Tablets X–XII.
85. Christ’s harrowing of Hell is first attested in the Gospel of Nicodemus, from the fourth
century CE. Other late antique journeys to Hell include the Apocalypse of Peter (perhaps
second century CE) and Apocalypse of Paul (perhaps third century CE). See Frankfurter
2000; Himmelfarb 1983.
86. Kalevala, Song 15.
87. Leucothea’s rescue of Odysseus at Od. 5.333–375 is a rare exception. Although one
might emphasize the fact that at the time that she rescues him, she is no longer human but
rather a god (-thea), Homer stresses the fact that she used to be Ino, the daughter of Cadmus,
which would have inevitably evoked the tragic tale of her life and its violent end. Historians

and cult documents such as the Asclepian iamata sometimes tersely describe a hero’s
intervention in human life after he has risen to superhuman status, but full-blown narratives
that were developed with the intention of entertaining an audience focus closely on the hero’s
mortal life—and on his very mortal death.
88. In using “series” and “serial,” I am adopting the system used by Mittell (2015) and
most other scholars who work on contemporary media.
89. Pi. P. 10.31–48.
90. E.g., the ghost of Polydorus at the beginning of E. Hec., Athena at the end of E. IT and
Heracles at the end of S. Ph.
91. Il. 6.130–141, 145–211; Il. 9.527–605.
92. Ancient narrators occasionally chose to weave stories that were complete in and of
themselves into a longer, known timeline. For example, when Heracles enters the action of
Euripides’s Alcestis, the chorus asks him what he’s doing in the city of Pherae. He explains
that he is just stopping by on his way to Thrace, where he must capture the mares of
Diomedes (thus completing Labor eight in his series of twelve). If, as some scholars suspect,
it was Euripides who invented the idea of Heracles wrestling with Death to save Alcestis,
then Heracles’s reference to the mares anchors this new episode within his established story.
93. E.g., in Homeric Hymns 3 (Apollo), 4 (Hermes), 19 (Pan), 28 (Athena).
94. E. Hipp. 10–59, 1416–1422. It is not clear whether Aphrodite’s involvement in the
death of Hippolytus was an innovation of Sophocles’s or not; Artemis’s involvement certainly
seems to be his invention: Gantz 1993: 287–288. See also Reed 1996, 1995.
95. E.g., Il. 5.330–352, 392–404, 439–442, 846–863.
96. Gantz 1993: charts 12, 17.
97. HHAp. 186–206.
98. Nothing underscores the nature of Holmes as a character in a series better than the fact
that Doyle was compelled by fans to resurrect his hero.
99. As told, e.g., at S. Tr. 555–581, 1141–1172.
100. As told, e.g., at S. OT 717–833, 998–1085, 1110–1176.
101. Paus. 8.34.
102. Eco 1985, 1972. Eco himself did not admire the use of iteration very much; he
thought it was a cheaply easy way of hooking audience members.
103. Il. 14.153–377.
104. HHHerm. 190–210.
105. Pi. P. 9.30–50.
106. Od. 9.318–330, 5.234–262, 23.187–204.
107. Cleaning manure from Augeas’s stables implied at Pi. O. 10.28 (where the word for
Heracles’s labor is latrion, which denotes particularly menial work) and is illustrated on one
of the metopes from the fifth-century Temple of Zeus at Olympia: see metope 12 in Figure
7.4. Scholiasts on the Iliad tell us that the tale of Heracles cleaning the stables was more fully
narrated by Callimachus in his Aitia (schol. Il. A 2.629, 11.700); the use of the diverted
rivers, for which Heracles had to dig ditches, is fully described at D.S. 4.13.3 and Apollod.
2.5.5.
108. Digging ditch to launch the Argo: A.R. 1.367–393; carrying the Argo across the
Libyan desert: first at Pi. P. 4.25–27.
109. Cadmus plowing a field and sowing dragon’s teeth: as early as Pherecyd. FGrH 3F22.
The same fragment tells us that half of these dragon’s teeth went to Aeetes in Colchis, where

Jason later plowed the ground with fire-breathing bulls (Pherecyd. FGrH 3F30, 31, 112; Pi. P.
4.233–238) and planted the teeth (Eumelus fr. 21 West). Discussion at Fowler 2013: 225–226,
360.
110. Odysseus plowing the seashore: Hyg. Fab. 95 and Apollod. Epit. 3.7.
111. See, for example, Figure 7.5, from an Athenian cup dating to about 480 BCE, which
shows a tortoise that ate human flesh as being no larger than a normal tortoise; at the right
edge of the image, we see the hindquarters of the Crommyonian Sow, which similarly look to
be normal in size. Another portion of this vase shows the Marathonian Bull, which is of
normal size as well.
112. Fowler 2013: 250–251.
113. Il. 19.418–424; Od. 5.204–224, 11.488–503.
114. E. Alc. 77–212.
115. E. Med. 1–88.
116. A. Ag. 1–20.
117. E. Ion 102–184.
118. S. Phil. 254–316.
119. E. HF 1–86.
120. S. Tr. 28–35.
121. S. Aj. 492–525.
122. Bacch. 5, 13, 17, 18.
123. Pi. P. 4.232–238, 250; Eumelus fr. 21 West.
124. Pi. P. 4.235–236.
125. The dream: Pi. O. 13.63–82; the deeds: 97–100.
126. Pi. P. 10.46–48.
127. Perseus: Pi. P. 10.49–50. Bellerophon: Pi. O. 13.83–84. The two translations are by
William H. Race. Emphasis added.
128. Bacch. 17.117–118 (trans. David A. Campbell).
129. Dalley 1989: 237–255, Tablets I–IV.
130. CTH 321 = Hoffner 1990: 10–13.
131. Song of Hedammu = CTH 428; Hoffner 1990: 48–51.
132. Song of Ullikummi = CTH 345; Hoffner 1990: 52–57.
133. See Assmann 1995: 49–57.
134. Ps. 74; Is. 27.1.
135. Ritner 1997: 32; Enuma Elish, Tablets IV and V; Dalley 1989: 254–257.
136. Apophis, Illuyanka, Hedammu, Leviathan.
137. Watkins (1995) has demonstrated the similar popularity of the god-fighting-dragon
topos in Indo-European poetry.
138. Ophion: Pherecydes of Syros FGrH, 7B4; cf. A.R. 1.503–506. Discussion at Gantz
1993: 739–741.
139. Argus: Hes. fr. 126 MW.
140. Od. 10.118.
141. Further discussion of the appearances of the gigantes at Gantz 1993: 445–454; the
mention that the neoteroi (poets of the fourth century and later) showed them with snaky legs
comes from schol. Od. 7.59.
142. E. Ion 989–996.

143. Hes. frs. 43a65 MW and 195.28–29 MW and cf. Th. 954; E. HF 177–180; discussion
at Gantz 1993: 445–454.
144. Cf. Burkert (1979: 8–9), who suggests that Hupasiya’s story seems to have started out
as a kind of novella of its own, which eventually became loosely attached to the story of
Illuyanka.
145. Hes. Th. 270–336.
146. Perseus: Hes. Th. 280–281; Heracles: Hes. Th. 289–294, 313–318, 332; Bellerophon:
Hes. Th. 325.
147. Bellerophon: Il. 6.183; Heracles: Il. 8.362–369, Od. 11.620–626. Zeus and Typhon: Il.
2.780.
148. For a résumé and analysis of the controversy about whether our present Hymn to
Apollo is actually two hymns joined together and on the date of the Hymn or at least the
“Pythian” portion of it, see the comments in West 2003: 9–12. West tentatively suggests that
the Hymn was performed at the first Pythian games in 586 BCE.
149. HHAp. 305–354. Stesichorus also made Hera the mother of Typhon, but we know
nothing else about his treatment: fr. 239, taken from the Et. Gen. see Typhōea.
150. Most scholars consider this excursus to be an interpolation into an earlier version of
the Hymn, but that doesn’t let us off the hook as far as understanding the role it played in the
Hymn as we now have it. Part of the reason that it exists, I suggest, is that it neatly balances
out another story about Hera that was told earlier in the Hymn, concerning the birth of Apollo.
In that story, Hera persecutes Apollo’s pregnant mother, preventing her from giving birth for a
long time (HHAp. 88–125).
151. Andrew Miller (1986: 88) suggests that this is because any description of the fight
between Zeus and Typhon would inevitably put Apollo’s defeat of Python in the shade. But
whatever the motivation, the elision of the battle has the effect of muting Typhon’s power.
152. HHAp. 306, 352. Like most recent editors of the Hymns, for line 352, I accept the
reading of all the manuscripts except M, which substitutes theoisin for brotoisin, making
Typhon an affliction to the gods instead of mortals.
153. Python as an affliction for mortals: HHAp. 302–304, 355–356, 364–366.
154. Sphinx: Hes. Th. 326, A. Th. 541, cf. 777; Chimaera: Il. 16.328; Scylla: Od. 12.125;
Boar: Hecat. FGrH 1F6; later descriptions: e.g., Apollod. 2.77–80 and Hyg. Fab. 30 (the
Hydra); D.S. 4.42.1 (the Trojan sea monster); Cerynitian Deer: E. HF 375.
155. HHAp. 207–215. Moreover, as an audience familiar with these stories would have
known, in each case, Apollo lost the woman to that mortal man.
156. HHAp. 216–228 (translation by Martin West).
157. For the myth and our textual evidence, Gantz 1993: 467–473. Gantz concludes that
the tale was known by the seventh century, if not earlier, although the full story does not
appear until Pherecydes in the fifth century (FGrH 3F22). Further discussion at Fowler 2013:
351–352.
158. Jessica Lamont has suggested to me that the allusion to the foundation of Thebes
additionally was meant as a “tacit nod” to the Ismenion in that city, a nearby oracle with
which the Delphic Oracle probably understood itself to be in competition. By describing
Thebes as not yet even founded, the Hymn thus emphatically asserts the priority of Delphi. I
find this suggestion persuasive.
159. HHAp. 294–300 and cf. 254–255, where Apollo lays the foundations of a temple near
Telphousa, which he subsequently abandons when Telphousa deceptively sends him away.

Later, Apollo returns to bury Telphousa under stones and then builds himself an altar on top:
HHAp. 382–385. He is not a god who worries about getting his hands dirty.
160. Earlier temples of Apollo: Pi. Pa. 8; and see discussion at Furley and Bremer 2001:
93–95. Of the earlier (female) owners, we first hear from A. Eum. 1–8 and Pi. fr. 55. See the
important discussion at Sourvinou-Inwood 1987. See also Graf 2008: 30; Furley and Bremer
2001: 95–97; Clay 1989: 61–62.
161. Graeae: Hes. Th. 270–273; Ceto: Hes. Th. 238; Echidna: Hes. Th. 297–298. Medusa is
called fair-cheeked at Pi. P. 12.16.
162. Chrysaor and Callirhoe: Hes. Th. 288; Echidna and Typhon: Hes. Th. 306–307; Ceto
and Phorcys: Hes. Th. 333; Medusa and Poseidon: Hes. Th. 277–278.
163. According to Ovid, Medusa was originally beautiful and was punished by Athena for
her alleged enjoyment of Poseidon’s rape; Athena turned her into the monster we now know
her as (Met. 4.790–803; cf. Apollod. 2.4.3).
164. Od. 11.625–626. The metopes on the Temple of Zeus at Olympia, built during the
second quarter of the fifth century BCE, show Athena supporting Heracles in his labors
(Figure 7.4). Athena is frequently shown in other artistic representations of Heracles’s labors,
too (e.g., Figure 7.3); these are discussed in Gantz’s treatment of each labor (1993: 381–416).
Helios: Pisander fr. 5 PEG and discussion at Fowler 2013: 294–295, 298 and Gantz 1993:
404–405.
165. Hes. Shield 216–237; Pherecyd. FGrH 3F11; Pi. P. 10.45; for artistic representations
of the gods helping Perseus and other textual references, see Gantz 1993: 304–307.
166. Pi. O. 13.65–87; Hes. fr. 43a MW.
167. Bacch. 17; and see Gantz 1993: 263–264 on artistic representations of the undersea
scene. The tradition concerning the crown is confused but may be as early as Epimenid. fr. 3;
see Gantz 1993: 264–266 for discussion and 266–268 on artistic representations of Athena,
and see also Fowler 2013: 472–473.
168. Od. 12.72.
169. Aphrodite: Pi. P. 4.216–217; Athena: our earliest attestation is Apollonius of Rhodes
(e.g., 1.526–527 etc.).
170. Vat. 16545 = Beazley Archive 205162; see also Gantz 1993: 359–360.
171. Plowing: note 109 above; ditch digging: notes 107 and 108 above.
172. Castanets: Apollod. 2.5.6. On sandals and bridle, see notes 165 and 166 above.
173. D.S. 20.41, 3–5; Duris FGrH 76F17; Phot. and Suid., s.v. “Lamia,” and schol. Ar. Pax
758; cf. Apostol. ap. Leutsch Paroim. Gr. 2.497–498 and schol. Aristid. p. 41 Dindorf. For
further discussion, see Johnston 1999: 161–183.
174. Stesich. fr. 220.
175. Getty Hexameters fr. i, Side B, lines 45 and 46: “O son of Zeus, and whoever …
Forceful … with your bow … and of the Hydra, many-.”
176. Il. 9.543–605. Meleager was the half brother of Tydeus, Diomedes’s father.
177. Il. 6.197–211. Glaucus himself describes the lineage; he is the grandson of
Bellerophon.
178. Od. 11.602–626.
179. Cypr. fr. 1 West with schol. A Il. 1.5–6; cf. Hes. Op. 156–165, which says that Zeus
decided to destroy the race of heroes (hemitheoi) with two wars—the one at Thebes and the
one at Troy. Further at Gantz 1993: 567–569.
180. Hes. fr. 204 MW lines 98–101.

181. Burkert 1992: 100–113. Erra and Ishum = Dalley 1989: 288. See also the recent
analysis of George 2013.
182. For these stories, see Dalley 1989: 1–38.
183. Il. 1.1–7 and Od. 1.1–9, both translations by Stanley Lombardo, slightly altered.
184. Burkert 1992: 92.

References
Ackerman, Robert. 1991. “The Cambridge Group: Origins and Composition.” In The Cambridge
Ritualists Reconsidered: Proceedings of the First Oldfather Conference, Held on the Campus of
the University of Illinois at Urbana-Champaign, April 27–30, 1989, edited by William M. Calder
III, 1–19. Atlanta.
Agnarsson Lab. 2017. “Social Spiders.” http://theridiidae.com/Social-Spiders.html.
Allen, T. W. 1908. “The Epic Cycle.” CQ 2: 64–74, 81–88.
Antonaccio, Carla. 1994. An Archaeology of Ancestors: Tomb Cult and Hero Cult in Ancient Greece.
Lanham, MD.
Arngrim, Alison. 2010. Confessions of a Prairie Bitch: How I Survived Nellie Oleson and Learned to
Love Being Hated. New York.
Assmann, Jan. 2009. “Der Mythos des Göttkonigs im Alten Ägypten.” In Menschen-Heros-Gott:
Weltenwürfe und Lebensmodelle im Mythos der Vormoderne, edited by Christine Schmitz and
Anja Bettenworth, 11–26. Stuttgart.
________
. 1995. Egyptian Solar Religion in the New Kingdom: Re, Amun and the Crisis of Polytheism.
London. German original published in 1983.
Athanassaki, Lucia. 2004. “Deixis, Performance and Poetics in Pindar’s First Olympian Ode.”
Arethusa 37, no. 2: 317–341.
Atwood, Margaret. 2005. The Penelopiad. Edinburgh.
Barber, Elizabeth J. Wayland. 1992. “The Peplos of Athens.” In Neils 1992a, 103–118.
Barkun, Michael. 2013 A Culture of Conspiracy: Apocalyptic Visions in Contemporary America. 2nd
ed. Berkeley, CA.
Barrett, J. L. 1996. “Anthropomorphism, Intentional Agents and Conceptualizing God.” PhD diss.,
Cornell University.
Barthes, Roland. 1972. Mythologies. London. (= Barthes, Roland. 1957. Mythologies. Paris.)
Bascom, William. 1965. “The Forms of Folklore: Prose Narratives.” Journal of American Folklore
78: 3–20.
Beard, Mary. 2002. The Invention of Jane Harrison. Cambridge, MA.
Beidelman, T. O., and Robert A. Segal. 2005. “W. Robertson Smith.” In Encyclopedia of Religion,
edited by Lindsay Jones, 8451–8453. Detroit.
Bennett, Gillian. 1999. Alas Poor Ghost: Traditions of Belief in Story and Discourse. 2nd expanded
ed. Logan, UT.
Bernabé, Alberto. 2013. “The Ephesia Grammata: Genesis of a Magical Formula.” In Faraone and
Obbink 2013, 71–96.
Bettini, Maurizio. 2013. Women and Weasels: Mythologies of Birth in Ancient Greece and Rome.
Chicago. (= Bettini, Maurizio. 1998. Nascere: Storie di donne, donnole, madri ed eroi. Torino.)
Bevis, I. C. 1988. Insects and Other Invertebrates in Classical Antiquity. Exeter, UK.
Black, Jeremy, Graham Cunningham, Jarle Ebeling, Esther Flückiger-Hawker, Eleanor Robson, Jon
Taylor, and Gábor Zólyomi, eds. 1998–2006. The Electronic Text Corpus of Sumerian Literature.
Oxford. http://etcsl.orinst.ox.ac.uk/.

Boas, Franz. 1917. Folk-Tales of Salishan and Sahaptin Tribes. Lancaster, PA.
Bordwell, David. 1985. Narration in the Fiction Film. Madison, WI.
Borgeaud, Philippe. 2004. Exercises de mythologie. Geneva.
Borghouts, J. F., trans. 1978. Ancient Egyptian Magical Texts. Leiden.
Boyd, Brian. 2009. On the Origin of Stories. Cambridge, MA.
Boyer, Pascal. 2001. Religion Explained: The Evolutionary Origins of Religious Thought. New York.
Brakke, David. 1999. “Parables and Plain Speech in the Fourth Gospel and the Apocryphon of
James.” Journal of Early Christian Studies 7: 187–218.
Braswell, Bruce. 1988. A Commentary on the Fourth Pythian Ode of Pindar. Berlin.
Bravo, Jorge. 2003. “Heroic Epiphanies: Narrative, Visual, and Cultic Contexts.” Illinois Classical
Studies 29: 63–84.
Brelich, Angelo. 1969. Paides e Parthenoi. Rome.
Bremmer, Jan. 2008. “Near Eastern and Native Traditions in Apollodorus’ Account of the Flood.” In
Greek Religion and Culture, the Bible and the Ancient Near East, edited by Jan Bremmer, 101–
116. Leiden.
________
. 2005. “Myth and Ritual in Ancient Greece: Observations on a Difficult Relationship.” In
Griechische Mythologie und frühes Christentum, edited by Raban von Haehling, 21–43.
Darmstadt.
________
. 1983. The Early Greek Concept of the Soul. Princeton, NJ.
________
. 1978. “Heroes, Rituals and the Trojan War.” Studi storico-religiosi 2: 5–38.
Brisson, Luc. 1982. Platon, les mots et les mythes. Paris.
Brown, Dee. 1979. Folktales of the Native American: Retold for Our Times. New York.
Burkert, Walter. 1992. The Orientalizing Revolution. Cambridge, MA. (= Burkert, Walter. 1984. Die
orientalisierende Epoche in der griechischen Religion und Literatur. Heidelberg.)
________
. 1987. Ancient Mystery Cults. Cambridge, MA.
________
. 1983. Homo Necans: The Anthropology of Ancient Greek Sacrificial Ritual and Myth.
Berkeley, CA. (= Burkert, Walter. 1972. Homo Necans. Berlin.)
________
. 1979. Structure and History in Greek Mythology and Ritual. Berkeley, CA.
________
. (1966) 2001. “Kekropidensage und Arrhephoria: Vom Initiationsritus zum
Panathenäenfest.” Translated by Peter Bing as “The Legend of Kekrops’ Daughters and the
Arrhephoria: From Initiation Ritual to Panathenaic Festival,” in Burkert, Savage Energies:
Lessons of Myth and Ritual in Ancient Greece, translated by Peter Bing, 37–63. Chicago.
Citations refer to the 2001 translation.
Burnett, Anne Pippin. 1994. “Hekabe the Dog.” Arethusa 27: 151–164.
Buxton, Richard G. A. 2009. Forms of Astonishment: Greek Myths of Metamorphosis. Oxford.
________
, ed. 1999a. From Myth to Reason? Studies in the Development of Greek Thought. Oxford.
________
. 1999b. Introduction to Buxton 1999a, 1–24.
________
. 1994. Imaginary Greece. Cambridge.
________
. 1987. “Wolves and Werewolves in Greek Thought.” In Interpretations of Greek Mythology,
edited by Jan Bremmer, 60–79. London.
Calame, Claude. 2015. Qu’est-ce que la mythologie grecque? Paris.

________
. 2011a. “Myth and Performance on the Athenian Stage: Praxithea, Erechtheus, Their
Daughters and the Etiology of Autochthony.” Classical Philology 106, no. 1: 1–19.
________
. 2011b. Mythe et histoire dans l’Antiquité grecque: La création symbolique d’une colonie.
2nd ed. Paris.
________
. 2009. Greek Mythology: Poetics, Pragmatics and Fiction. Cambridge. (= Calame, Claude.
2000. Poetiques des Mythes dans la Grèce Antique. Paris.)
________
. 2006. “La fabrication historiographique d’un passé héroïque et classique: Archaia et palaia
chez Hérodote.” Ktèma 31: 39–49.
________
. 2004. “Deictic Ambiguity and Auto-Referentiality: Some Examples from Greek Poetics.”
Arethusa 37, no. 2: 415–443.
________
. 2003. Myth and History in Ancient Greece: The Symbolic Creation of a Colony. Princeton,
NJ. (= Calame, Claude. 1996. Mythe et histoire dans l’Antiquité grecque: La création symbolique
d’une colonie. Lausanne.)
________
. 1995. “Mythe et mythologie: Taxonomies indigenes et categories occidentales.” In Presenze
classiche nelle letterature: Occidental il mito dall’età antica all’età moderna e contemporanea,
edited by Margherita Rossi Cittidini, 57–76. Perugia, Italy.
________
. 1991. “Mythe et rite en Grèce: Des categories indigenes?” Kernos 4: 179–204.
________
. 1990. Thésée et l’imaginaire athénien. Lausanne.
Caliva, Kathryn. 2016. “Speech Acts and Embedded Narrative Structure in the Getty Hexameters.”
Archiv für Religiongeschichte 17: 139–164.
Caughey, John L. 1984. Imaginary Social Worlds. Lincoln, NE.
Chatman, Seymour. 1978. Story and Discourse: Narrative Structure in Fiction and Film. Ithaca, NY.
Cialdini, Robert. 2016. Pre-suasion: A Revolutionary Way to Influence and Persuade. New York.
Clark, Lynn Schofield. 2003. From Angels to Aliens: Teenagers, the Media, and the Supernatural.
New York.
Clarke, Katherine. 2017. “Walking through History: Unlocking the Mythical Past.” In Hawes 2017a,
14–31.
Clay, Jenny Strauss. 1989. The Politics of Olympus. Princeton, NJ.
Cohn, Dorrit. 1978. Transparent Minds: Narrative Modes for Presenting Consciousness in Fiction.
Princeton, NJ.
Coleridge, Samuel Taylor. 1817. Biographia Literaria. London.
Collins, V. H., ed. 1924. Ghosts and Marvels. Oxford.
Csapo, Eric. 2005. Theories of Mythology. Malden, MA.
Cuneo, Michael W. 2001. American Exorcism: Expelling Demons in the Land of Plenty. New York.
Dalley, Stephanie, ed., trans., and comm. 1989. Myths from Mesopotamia: Creation, the Flood,
Gilgamesh, and Others. Oxford.
Darshan, Guy. 2013a. “The Biblical Account of the Post-Diluvian Generation (Gen. 9:20–10:32) in
the Light of Greek Genealogical Literature.” Vetus Testamentum 63: 515–535.
________
. 2013b. “The Reinterment of Saul and Jonathan’s Bones (II Sam. 21,12–14) in Light of
Ancient Greek Hero-Cult Stories.” Zeitschrift für die alttestamentliche Wissenschaft 125, no. 4:
2–6.
Davies, Owen. 2004. “French Charmers and Their Healing Charms.” In Roper 2004a, 91–112.

Denson, Shane. 2011. “Marvel Comics’ Frankenstein: A Case Study in the Media of Serial Figures.”
Amerikastudien 56, no. 4: 531–553.
Detienne, Marcel. 1986. The Creation of Mythology. Chicago. (= Detienne, Marcel. 1981.
L’invention de la mythologie. Paris.)
________
. 1979. Dionysus Slain. Baltimore. (= Detienne, Marcel. 1977. Dionysus mis à mort. Paris.)
________
. 1977. The Gardens of Adonis. Princeton, NJ. (= Detienne, Marcel. 1972. Les jardins
d’Adonis. Paris.)
Dimmick, Jeremy. 2002. “Ovid in the Middle Ages: Authority and Poetry.” In The Cambridge
Companion to Ovid, edited by Philip Hardie, 264–287. Cambridge.
Doak, Brian R. 2012. The Last of the Rephaim: Conquest and Cataclysm in the Heroic Ages of
Ancient Israel. Boston.
Doniger, Wendy, ed., trans., and comm. 2004. Hindu Myths: A Sourcebook. New York.
________
. 1996. “Minimyths and Maximyths and Political Points of View.” In Myths and Method,
edited by Laurie Patton and Wendy Doniger, 109–127. Charlottesville, VA.
Dorsey, George A. 1904. Traditions of the Skidi Pawnee. London.
Doty, William G. (1986) 2000. Mythography: The Study of Myths and Rituals. Tuscaloosa, AL.
Citations refer to the 2000 edition.
Dover, Kenneth, ed., trans., and comm. 1997. Aristophanes. Frogs. Oxford.
Dowden, Ken. 1989. Death and the Maiden: Girls’ Initiation Rites in Greek Mythology. London.
Doyle, Arthur Conan. 2017. Gothic Tales. Edited by Darryl Jones. Oxford.
Dundes, Alan. 1964. The Morphology of North American Indian Folktales. Helsinki.
Eco, Umberto. 2009. “On the Ontology of Fictional Characters: A Semiotic Approach.” Sign Systems
Studies 37, nos. 1–2: 82–97.
________
. 1990. Travels in Hyperreality. San Diego.
________
. 1985. “Innovation and Repetition: Between Modern and Post-Modern Aesthetics.”
Daedalus 114, no. 4: 161–184.
________
. 1972. “The Myth of Superman.” Diacritics 2, no. 1: 14–22.
Edelstein, Emma J., and Ludwig Edelstein. 1945. Asclepius: Collection and Interpretation of the
Testimonies. Baltimore.
Eder, Jens, Fotis Jannidis, and Ralf Schneider. 2010. “Characters in Fictional Worlds: An
Introduction.” In Revisionen 3: Characters in Fictional Worlds: Understanding Imaginary Beings
in Literature, Film and Other Media, edited by Jens Eder, Fotis Jannidis, and Ralf Schneider, 3–
64. Berlin.
Ekroth, Gunnel. 2015. “Heroes: Living or Dead.” In The Oxford Handbook of Ancient Greek
Religion, edited by Esther Eidinow and Julia Kindt, 383–396. Oxford.
________
. 2002. The Sacrificial Rituals of Greek Hero-Cults in the Archaic to the Hellenistic Periods.
Liège.
Eliade, Mircea. 1963. Myth and Reality. San Francisco.
________
. 1959. The Sacred and the Profane: The Nature of Religion. San Diego.
________
. 1954. The Myth of the Eternal Return. New York. French original published in 1949.
Engelmann, Helmut. 1975. The Delian Aretalogy of Sarapis. Leiden.
Erdoes, Richard, and Alfonso Ortiz, eds. 1984. American Indian Myths and Legends. New York.

Faraone, Christopher. 2003. “Playing the Bear and the Fawn for Artemis: Female Initiation or
Substitute Sacrifice?” In Initiation in Ancient Greek Rituals and Narratives: New Critical
Perspectives, edited by David Dodd and Christopher Faraone, 43–67. London.
________
. 2001. “Handbook or Anthology? The Collection of Greek and Egyptian Incantations in Late
Hellenistic Egypt.” Archiv für Religionsgeschichte 2: 195–214.
________
. 1995. “The Mystodokos and the Dark-Eyed Maidens: Multicultural Influences on a Late-
Hellenistic Incantation.” In Ancient Magic and Ritual Power, edited by Marvin Meyer and Paul
Mirecki, 297–333. Leiden.
Faraone, Christopher, and Dirk Obbink, eds. 2013. The Getty Hexameters: Poetry, Magic, and
Mystery in Ancient Selinous. Oxford.
Farber, Gertrude, ed., trans., and comm. 1997. “Inna and Enki.” In Hallo and Lawson 1997, 522–526.
Farnell, L. R. 1921. Greek Hero Cults and Ideas of Immortality. Oxford.
Faulkner, William. 1946. The Portable Faulkner. Edited by Malcolm Cowley. New York.
Fearn, D. 2007. Bacchylides: Politics, Performance, Poetic Tradition. Oxford.
Feeney, Denis. 1991. The Gods in Epic: Poets and Critics of the Classical Tradition. Oxford.
Felson, Nancy. 2004a. Introduction to Arethusa 37, no. 2: 253–266.
________
. 2004b. “The Poetic Effects of Deixis in Pindar’s Ninth Pythian Ode.” Arethusa 37, no. 2:
365–389.
________
. 1999. “Vicarious Transport: Fictive Deixis in Pindar’s Pythian Four.” Harvard Studies in
Classical Philology 99: 1–31.
Fontenrose, Joseph. 1966. The Ritual Theory of Myth. Berkeley, CA.
________
. 1961. “Some Observations on Hyman’s Review of Python.” Carleton Miscellany 2: 122–
125.
________
. 1959. Python: A Study of Delphic Myth and its Origins. Berkeley, CA.
Forbes-Irving, P. M. C. 1990. Metamorphosis in Greek Myths. Oxford.
Forster, E. M. (1927) 1985. Aspects of the Novel. San Diego. Citations refer to the 1985 edition.
Fowler, Robert. 2016. “History.” In The Oxford Handbook of Ancient Greek Religion, edited by
Esther Eidinow and Julia Kindt, 195–209. Oxford.
________
. 2013. Early Greek Mythography. Vol. 2, Commentary. Oxford.
________
. 2011. “Mythos and Logos.” Journal of Hellenic Studies 131: 45–66.
________
. 2000. Early Greek Mythography. Vol. 1, Text and Introduction. Oxford.
Franco, Cristiana. 2014. Shameless: The Canine and the Feminine in Ancient Greece. Berkeley, CA.
Frankfurter, David T. M. 2009. “The Laments of Horus in Coptic: Myth, Folklore, and Syncretism in
Late Antique Egypt.” In Antike Mythen: Medien, Transformationen und Konstruktionen, edited
by Ueli Dill and Christine Walde, 229–247. Berlin.
________
. 2001. “Narrating Power: The Theory and Practice of the Magical Historiola in Ritual
Spells.” In Ancient Magic and Ritual Power, edited by Marvin Meyer and Paul Mirecki, 457–476.
Leiden.
________
. 2000. “Early Christian Apocalypticism: Literature and Social World.” In The Encyclopedia
of Apocalypticism, vol. 1, The Origins of Apocalypticism in Judaism and Christianity, edited by
John J. Collins, 415–456. New York.
Franklin, Ruth. 2016. Shirley Jackson: A Rather Haunted Life. New York.

Frazer, James G., ed., trans., and comm. 1921. Apollodorus: The Library; With an English
Translation. Cambridge, MA.
________
. 1901. “On Some Ceremonies of the Central Australian Tribes.” Paper read before the
Australian Association for the Advancement of Science. Melbourne.
________
. (1890) 1911–1915. The Golden Bough: A Study in Magic and Religion. 3rd expanded ed.
London.
Furley, William, and Jan Maarten Bremer, eds., trans., and comms. 2001. Greek Hymns: Selected
Cult Songs from the Archaic to the Hellenistic Period. 2 vols. Tübingen.
Furth, Robin. 2012. Stephen King’s “The Dark Tower”: The Complete Concordance, Revised and
Updated. New York.
Gaiman, Neil. (1996) 2001. Neverwhere. New York. Citations refer to the 2001 edition.
Gantz, Timothy. 1993. Early Greek Myth: A Guide to Literary and Artistic Sources. Baltimore.
Gardner, Jared. 2012. Projections: Comics and the History of Twenty-First Century-Storytelling.
Stanford, CA.
Geertz, Armin, and Jeppe Sinding Jensen, eds. 2011. Religious Narrative, Cognition and Culture:
Image and Word in the Mind of Narrative. Sheffield, UK.
George, A. R. 2013. “The Poem of Erra and Ishum: A Babylonian Poet’s View of War.” In Warfare
and Poetry in the Middle East, edited by Hugh Kennedy, 39–71. London.
Gernet, Louis. 1981. The Anthropology of Ancient Greece. Baltimore. (= Gernet, Louis. 1968.
Anthropologie de la Grèce antique. Paris.)
Gibson, J. J. 1979. The Ecological Approach to Visual Perception. Boston.
Giles, David. 2010. “Parasocial Relationships.” In Revisionen 3: Characters in Fictional Worlds:
Understanding Imaginary Beings in Literature, Film, and Other Media, edited by Jens Eder,
Fotis Jannidis, and Ralf Schneider, 442–448. Berlin.
________
. 2002. “Parasocial Interaction: A Review of the Literature and a Model for Future
Research.” Media Psychology 4, no. 3: 279–305.
Glassner, Jean-Jacques. 2004. Mesopotamian Chronicles. Edited and translated by Benjamin R.
Foster. Atlanta. French original published in 1993.
Goff, Barbara. 2004. Citizen Bacchae. Berkeley, CA.
Gow, A. S. F., and A. F. Scholfield, eds. and trans. 1953. Nicander: The Poems and Poetical
Fragments. Cambridge.
Graf, Fritz. 2008. Apollo. London.
________
. 2004. “Trick or Treat? On Collective Epiphanies in Antiquity.” Illinois Classical Studies 29:
111–130.
________
. 2003. “Initiation: A Concept with a Troubled History.” In Initiation in Ancient Greek
Rituals and Narratives, edited by David B. Dodd and Christopher A. Faraone, 3–24. London.
________
. 2000. “The Locrian Maidens.” In Oxford Readings in Greek Religion, edited by R. G. A.
Buxton, 250–270. Oxford. German original published in 1978.
________
. 1997. “Medea, the Enchantress from Afar: Remarks on a Well-Known Myth.” In Medea:
Essays on Medea in Myth, Literature, Philosophy and Art, edited by James J. Clauss and Sarah
Iles Johnston, 21–43. Princeton, NJ.
________
. 1995. “Il mito tra menzogna e ‘Urwahrheit.’ ” In Presenze classiche nelle letterature
occidental: Il mito dall’età antica all’età moderna e contemporanea, edited by Margherita Rossi

Cittidini, 43–56. Perugia.
________
. 1993. Greek Mythology: An Introduction. Baltimore. (= Graf, Fritz. 1987. Greichische
Mythologie. Zurich.)
________
. 1987. “Orpheus, a Poet among Men.” In Interpretations of Greek Mythology, edited by Jan
Bremmer, 80–106. London.
________
. 1980. “Milch, Hönig und Wein: Zum Verstandnis der Libation im griechischen Ritual.” In
Perennitas: Studi in onore di Angelo Brelich, 209–221. Rome.
Graf, Fritz, and Sarah Iles Johnston. 2013. Ritual Texts for the Afterlife: Orpheus and the Bacchic
Gold Tablets. 2nd ed. London.
Green, Peter. 1997. Introduction to The Argonautica, by Apollonios Rhodios, 1–41. Berkeley, CA.
Gregory, Lady Augusta. 1902. Cuchulain of Muirthemne. London.
Greimas, A. J. 1983. Structural Semantics: An Attempt at a Method. Lincoln, NE. (= Greimas, A. J.
1966. Sémantique structural; recherché de method. Paris.)
Gunn, Anna. 2013. “I Have a Character Issue.” New York Times, August 23.
Hall, Edith, ed., trans., and comm. 1996. Aeschylus. Persians. Warminster, UK.
Halliday W. R. 1913. Greek Divination: A Study of Its Methods and Principles. London.
Hallo, William K., and Lawson Younger Jr., eds. 1997. The Context of Scripture. Vol 1. Leiden.
Hamilton, Christine R. E. 2017. “The Function of the Deus ex Machina in Euripidean Drama.” PhD
diss., The Ohio State University.
Hänninen, Kiri. 2017. “Narrating Supernatural Experiences.” In Narrating Religion, edited by Sarah
Iles Johnston, 125–140. New York.
Hansen, William. 2017. The Book of Greek and Roman Folktales, Legends and Myths. Princeton, NJ.
Harari, Yuval. 2015. Sapiens: A Brief History of Humankind. New York.
Harrelson, Walter. 2005. “Myth and Ritual School.” In Encyclopedia of Religion, edited by Lindsay
Jones, 6380–6383. Detroit.
Harrison, Jane Ellen. (1928) 1976. The Myths of Greece and Rome. London. Citations refer to the
1976 edition.
________
. (1925) 1965. “Reminiscences of a Student Life.” Arion 4, no. 2: 312–346. Citations refer to
the 1965 edition.
________
. (1924) 1963. Mythology. New York. Citations refer to the 1963 edition.
________
. 1921. Epilegomena to the Study of Greek Religion. Cambridge.
________
. (1912) 1927. Themis: A Study of the Social Origins of Greek Religion. 2nd ed. Cambridge.
________
. (1903) 1922. Prolegomena to the Study of Greek Religion. 3rd ed. Cambridge. Citations
refer to the 1922 edition.
Harrison, Jane Ellen, and Margaret de G. Verrall. 1890. Mythology and Monuments of Ancient
Athens. London.
Harvey, Graham. 2000. “Fantasy in the Study of Religions: Paganism as Observed and Enhanced by
Terry Pratchett.” Discus: The Journal of the British Association for the Study of Religions 6.
http://www.basr.ac.uk/diskus/diskus1-6/harvey-6.txt.
Hawes, Greta, ed. 2017a. Myths on the Map: The Storied Landscapes of Ancient Greece. Oxford.
________
. 2017b. “Of Myths and Maps.” In Hawes 2017a, 1–13.

Heim, Ricardus. 1892–1893. “Incantamenta magica Graeca Latina.” In Jahrbücher für classische
Philologie, suppl. 19. Leipzig.
Henrichs, Albert. 2010. “What Is a Greek God?” In The Gods of Ancient Greece: Identities and
Transformations, edited by Jan Bremmer and Andrew Erskine, 19–39. Edinburgh.
________
. 1994. “Anonymity and Polarity: Unknown Gods and Nameless Altars at the Areopagus.”
Illinois Classical Studies 19: 27–58.
________
. 1991. “Namenlosigkeit und Euphemismus: Zur Ambivalenz der chthonischen Mächte im
attischen Drama.” In Fragmenta Dramatica: Beiträge zur Interpretation der griechischen
Tragikerfragmenta und ihrer Wirkungsgeschichte, edited by Heinz Hofmann and Annette Harder,
161–201. Göttingen.
Herman, David, Manfred Jahn, and Marie-Laure Ryan, eds. 2005. The Routledge Encyclopedia of
Narrative Theory. London.
HEXRPG, LLC. 2017. “Hogwarts Extreme: The Interactive Hogwarts Experience.”
http://hexrpg.com.
Higbie, Carolyn. 2003. The Lindian Chronicle and the Greek Creation of Their Past. Oxford.
Himmelfarb, Martha. 1983. Tours of Hell: An Apocalyptic Form in Jewish and Christian Literature.
Philadelphia.
Hoffner, Harry A., Jr. 1990. Hittite Myths. Atlanta.
Holmes, Sally. 2014. “Angelina Jolie Says Violent Maleficent Scene Was a Metaphor for Rape.” Elle,
June 12. http://www.elle.com/culture/celebrities/news/a15426/angelina-jolie-maleficent-scene-
metaphor-for-rape.
Hooke, S. H., ed. 1958. Myth, Ritual and Kingship. Oxford.
________
. 1938. The Origins of Early Semitic Ritual. London.
________
, ed. 1935. Labyrinth: Further Studies in the Relation between Myth and Ritual. London.
________
, ed. 1933a. Myth and Ritual. London.
________
. 1933b. “The Myth and Ritual Pattern of the Ancient East.” In Hooke 1933a, 1–14.
Hoorn, Johan F., and Elly A. Konijn. 2003. “Perceiving and Experiencing Fictional Characters: An
Integrative Account.” Japanese Psychological Research 45, no. 4: 250–268.
Hopman, Marianne Govers. 2012. Scylla: Myth, Metaphor, Paradox. Cambridge.
Horton, Donald, and Richard R. Wohl. 1956. “Mass Communication and Parasocial Interaction:
Observations on Intimacy at a Distance.” Psychiatry 19, no. 3: 185–206.
Hufford, David. 1982. “Traditions of Disbelief.” New York Folklore 8: 47–55.
Hutton, Ronald. 1999. The Triumph of the Moon: A History of Modern Pagan Witchcraft. Oxford.
Hyman, Stanley. 1960. Review of Python by Joseph Fontenrose. Carleton Miscellany 1: 124–127.
________
. 1955. “The Ritual View of Myth and the Mythic.” Journal of American Folklore 68: 462–
472.
James, M. R. 2011. Collected Ghost Stories. Edited by Darryl Jones. Oxford.
Janko, Richard. 2013. “The Hexametric Incantations against Witchcraft in the Getty Museum: From
Archetype to Exemplar.” In Faraone and Obbink 2013, 29–53.
Jeanmaire, Henri. 1939. Couroi et Courètes: Essai sur l’éducation spartiate et sur les rites
d’adolescence dans l’antiquité hellénique. Lille.
Jenkins, Henry. 2006. Convergence Culture: Where Old and New Media Collide. New York.
________
. 1992. Textual Poachers: Television Fans and Participatory Culture. London.

Johnston, Sarah Iles. 2014. “Goddesses with Torches in the Getty Hexameters and Alcman fr. 94.”
Zeitschrift für Papyrologie und Epigraphik 191: 32–35.
________
. 2013. “Myth in the Getty Hexameters.” In Faraone and Obbink 2013, 121–156.
________
. 2009. “A New Web for Arachne.” In Antike Mythen: Medien, Transformationen und
Konstruktionen, edited by Ueli Dill and Christine Walde, 1–11. Berlin.
________
. 2008. Ancient Greek Divination. Malden, MA.
________
. 2006. “Antigone’s Other Choice.” Helios 33S: 179–186.
________
. 2005. “Delphi and the Dead.” In Mantike: Studies in Ancient Divination, edited by Sarah
Iles Johnston and Peter T. Struck, 283–306. Leiden.
________
. 2002. “Myth, Cult and Poet: The Homeric Hymn to Hermes and Its Performative Context.”
Classical Philology 97: 109–232.
________
. 1999. Restless Dead: Encounters between the Living and the Dead in Ancient Greece.
Berkeley, CA.
________
. 1995. “The Song of the Iynx: Magic and Rhetoric in Pythian 4.” Transactions of the
American Philological Association 125: 177–206.
________
. 1994. “Penelope and the Erinyes: Odyssey 20.61–82.” Helios 21: 137–159.
Jolly, Alison. 1999. Lucy’s Legacy: Sex and Intelligence in Human Evolution. Cambridge, MA.
Jones, Christopher. 2010. New Heroes in Antiquity: From Achilles to Antinoos. Cambridge, MA.
Judd, Walter S., and Graham A. Judd. 2017. Flora of Middle-Earth: Plants of J. R. R. Tolkien’s
Legendarium. Oxford.
Käppel, Lutz. 1992. Paian: Studien zur Geschichte einer Gattung. Vol. 37 of Untersuchungen zur
antiken Literarur und Geschichte. Berlin.
Katz, Joshua, and Katharina Volk. 2000. “ ‘Mere Bellies’? A New Look at Theogony 26–28.”
Journal of Hellenic Studies 120: 122–131.
Kearns, Emily. 1989. The Heroes of Attica. London.
Kemmerer, Lisa. 2011. Animals and World Religions. Oxford.
Kirk, Geoffrey S. 1974. The Nature of Greek Myths. Harmondsworth, UK.
________
. 1970. Myth: Its Meaning and Functions in Ancient and Other Cultures. Berkeley, CA.
Klein, Jacob, ed., trans., and comm. 1997. “Enki and Ninmah.” In Hallo and Lawson 1997, 516–518.
Klimkiewicz, JoAnn. 2013. “Desperately Seeking Christy.” New York Times Magazine, November
22.
Kluckhohn, Clyde. 1942. “Myths and Rituals: A General Theory.” Harvard Theological Review 35:
45–79.
Kripal, Jeffrey. 2011. Mutants and Mystics: Science Fiction, Superhero Comics, and the Paranormal.
Chicago.
Krummen, Eveline. 1990. Pursos humnon: Festliche und mythisch-rituelle Tradition als
Voraussetzungen einer Pindarinterpretation (Isthmie 4, Pythie 5, Olympie 1 und 3). Berlin.
Kyle, Donald G. 1992. “The Panatheniac Games: Sacred and Civil Athletics.” In Neils 1992a, 77–
102.
Labov, William, and Joshua Waletsky. 1967. “Oral Versions of Personal Experience.” Journal of
Narrative and Life History 7: 3–38.
Lakoff, George, and Mark Johnson. 1980. Metaphors We Live By. Chicago.

Landy, Joshua. 2012. How to Do Things with Fictions. Oxford.
Lang, Andrew. 1894. Cock Lane and Common-Sense. London.
Lardinois, André. 1992. “Greek Myths for Athenian Rituals: Religion and Politics in Aeschylus’
Eumenides and Sophocles’ Oedipus Coloneus.” Greek, Roman, and Byzantine Studies 33, no. 4:
313–327.
Laroche, Emmanuel. 1971. Catalogue des textes hittites. Paris. This resource is now available online
and is regularly updated: http://www.hethport.uni-wuerzburg.de/CTH/.
Larson, Jennifer. 1995. Greek Heroine Cults. Madison, WI.
Leach, Edmund. 1954. The Political Systems of Highland Burma. London.
Leitao, David D. 1995. “The Perils of Leukippos: Initiatory Transvestism and Male Gender Ideology
in the Ekdusia of Phaistos.” Classical Antiquity 14: 130–163.
Lepore, Jill. 2011. “Dickens in Eden: Summer Vacation with ‘Great Expectations.’ ” New Yorker,
August 29.
Levaniouk, Olga. 2008. “Penelope and the Pandareids.” Phoenix 62: 5–38.
Lévi-Strauss, Claude. 1958. Anthropologie structural. Paris. (= Lévi-Strauss, Claude. 1972.
Structural Anthropology. London.)
Lewis, C. S. 1966. Of Other Worlds: Essays and Stories. New York.
________
. 1947. “On Stories.” In Essays Presented to Charles Williams, edited by C. S. Lewis, 90–
105. Oxford. Reprinted in Lewis 1966, 3–21.
LiDonnici, Lynn R. 1995. The Epidaurian Miracle Inscriptions. Atlanta.
Lincoln, Bruce. 1999. Theorizing Myth: Narrative, Ideology and Scholarship. Chicago.
Lloyd-Jones, Hugh. 1985. “Pindar and the Afterlife.” In Pindare: Entretiens sur l’antiquité classique,
vol. 31, 245–284. Geneva.
López-Ruiz, Carolina. 2018. Gods, Heroes and Monsters: A Sourcebook of Greek, Roman and Near
Eastern Myths in Translation. 2nd ed. New York.
________
. 2010. When the Gods Were Born: Greek Cosmogonies and the Near East. Cambridge, MA.
Lorenz, Konrad. 1963. Das sogenannte Böse: Zur Naturgeschichte der Aggression. Vienna: G.
Borotha-Schoeler. Translated as On Aggresion by Marjorie Kerr Wilson. New York: Harcourt,
Brace & World, 1966.
Louden, Bruce. 2011. Homer’s “Odyssey” and the Near East. Cambridge.
________
. 2006. The “Iliad”: Structure, Myth and Meaning. Baltimore.
Lowe, N. J. 2007. “Gilbert Murray and Psychic Research.” In Gilbert Murray Reassessed:
Hellenism, Theatre, and International Politics, edited by Christopher Stray, 349–370. Oxford.
________
. 2000. The Classical Plot and the Invention of Western Narrative. Cambridge.
Luhrmann, Tanya. 2013a. “Belief Is the Least Part of Faith.” New York Times, June 30.
________
. 2013b. “Conjuring Up Our Own Gods.” New York Times, October 14.
________
. 2013c. “C. S. Lewis, Evangelical Rock Star.” New York Times, June 25.
________
. 2012. When God Talks Back: Understanding the American Evangelical Relationship with
God. New York.
Lupu, Eran. 2005. Greek Sacred Law: A Collection of New Documents. Leiden.
Lyons, Deborah. 1997. Gender and Immortality: Heroines in Ancient Greek Myth and Cult.
Princeton, NJ.
Maas, Paul. 1942. “The Philinna Papyrus.” Journal of Hellenic Studies 62: 33–38.

Maehler, Herwig. 1997. Die Leider des Bakchylides. Vol. 2. Leiden.
Magliocco, Sabina. 2004. Witching Culture: Folklore and Neopaganism in America. Philadelphia.
Martin, Richard. 2004. “Home Is the Hero: Deixis and Semantics in Pindar Pythian 8.” Arethusa 37,
no. 2: 343–363.
Maspero, Gaston, trans., and Hasan El-Shamy, ed. 2004. Popular Stories of Ancient Egypt. Oxford.
Matthey, Philippe. 2016. “Étudier les mythes en context francophone: À propos de quatre ouvrages
récents.” Kernos 29: 1–15.
Miller, Andrew. 1986. From Delos to Delphi: A Literary Study of the Homeric Hymn to Apollo.
Leiden.
Mills, Mary. 1990. Human Agents of Cosmic Power in Hellenistic Judaism and the Synoptic
Tradition. Sheffield, UK.
Mittell, Jason. 2015. Complex TV: The Poetics of Contemporary Television Storytelling. New York.
________
. 2010. Television and American Culture. New York.
Morgan, Kathryn. 2000. Myth and Philosophy from the Presocratics to Plato. Cambridge.
Most, Glenn W. 1999. “From Logos to Muthos.” In Buxton 1999a, 25–49.
Moyers & Company. 1999. “The Mythology of ‘Star Wars’ with George Lucas.” June 18.
http://billmoyers.com/content/mythology-of-star-wars-george-lucas/.
Murray, Gilbert. 1925. Five Stages of Greek Religion. Oxford.
________
. 1914. Hamlet and Orestes: A Study in Traditional Types. Oxford.
________
. (1912) 1927. “An Excursus on the Ritual Forms Preserved in Greek Tragedy.” In Harrison
(1912) 1927, 341–363.
Murray, Penelope. 1999. “What Is a Muthos for Plato?” In Buxton 1999a, 251–262.
Neils, Jennifer, ed. 1992a. Goddess and Polis: The Panathenaic Festival in Ancient Athens.
Princeton, NJ.
________
. 1992b. “The Panathenia: An Introduction.” In Neils 1992a, 13–28.
Nelson, Victoria. 2001. The Secret Life of Puppets. Cambridge, MA.
Nesbit, Edith. 1912. The Magic World. Kent, UK.
Nilsson, Martin P. (1950) 1961. Geschichte der griechischen Religion. Vol. 2. 2nd ed. Munich.
________
. (1941) 1967. Geschichte der griechischen Religion. Vol. 1. 3rd ed. Munich. Citations refer
to the 1967 edition.
________
. 1932. The Mycenaean Origin of Greek Mythology. Berkeley, CA.
Noel, Ruth. 1980. The Languages of Tolkien’s Middle-Earth: A Complete Guide to All Fourteen of
the Languages That Tolkien Invented. New York.
Noyes, Dorothy. 2016. Humble Theory. Bloomington, IN.
Nussbaum, Emily. 2012. “Tune in Next Week: The Curious Staying Power of the Cliffhanger.” New
Yorker, July 30.
O’Flaherty, Wendy Doniger. 1988. Other People’s Myths: The Cave of Echoes. Chicago.
O’Higgins, Dolores. 1997. “Medea as Muse: Pindar’s Pythian 4.” In Medea: Essays on Medea in
Myth, Literature, Philosophy and Art, edited by James J. Clauss and Sarah Iles Johnston, 103–
126. Princeton, NJ.
O’Sullivan, Sean. 2013. “Serials and Satisfaction.” Romanticism and Victorianism on the Net 63.
http://ravonjournal.org/.
Pardee, Dennis, ed. and trans. 1997. “The Kirta Epic and the Aqhat Legend.” In Hallo and Lawson
1997, 331–355.

Parker, L. P. E., ed. 2016. Iphigenia in Tauris, by Euripides. Oxford.
Parker, Robert. 2007. “Gilbert Murray and Greek Religion.” In Gilbert Murray Reassessed:
Hellenism, Theatre, and International Politics, edited by Christopher Stray, 81–102. Oxford.
Parkinson, Richard, ed., trans., and comm. 1997. The Tale of Sinuhe and Other Ancient Egyptian
Poems, 1940–1640 BCE. Oxford.
Partridge, Chrisopher. 2004. The Re-enchantment of the West. Vol. 1. London.
Paskin, Willa. 2013. Review of the season finale of Homeland. Slate, December 15.
http://www.slate.com/articles/arts/television/2013/12/homeland_recap_the_star_season_3_finale_
reviewed.html.
Pavlou, Maria. 2012. “Bacchylides 17: Singing and Usurping the Paean.” Greek, Roman, and
Byzantine Studies 52: 510–539.
Peacock, Sandra J. 1991. “An Awful Warmth about Her Heart: The Personal in Jane Harrison’s Ideas
on Religion.” In The Cambridge Ritualists Reconsidered, edited by William M. Calder III, 167–
184. Atlanta.
Perlman, Paula J. 1995. “Invocatio and Imprecatio: The Hymn to the Greatest Kouros from
Palaikastro and the Oath in Ancient Crete.” Journal of Hellenic Studies 115: 161–167.
Phelan, James. 2017. Somebody Telling Somebody Else: A Rhetorical Poetics of Narrative.
Columbus, OH.
Pinch, Geraldine. 2002. Egyptian Mythology: A Guide to the Gods, Goddesses and Traditions of
Ancient Egypt. Oxford.
Piraro, Dan. 2012. “Ill-Fated Blind Date: St. Patrick and Medusa.” Bizarro, March 16.
http://bizarro.com/comics/march-16-2012/.
Pirenne-Delforge, Vinciane. 2008. Retour à la source: Pausanias et la religion greque. Kernos,
suppl. 20. Liège.
________
. 1994. L’Aphrodite grecque: Contribution à l’étude de ses cultes et de sa personalité dans le
pantheon archaïque et classique. Liège.
Pirenne-Delforge, Vinciane, and Gabriella Pironti. 2016. L’Héra de Zeus: Ennemie intime, épouse
definitive. Paris.
Platt, Verity. 2010. “Art History in the Temple.” Arethusa 43, no. 2: 197–213.
Pócs, Eva. 2009. “Miracles and Impossibilities in Magic Folk Poetry.” In Roper 2009, 27–53.
Portelli, Alessandro. 1997. “On the Lower Frequencies: Sound and Meaning in Native Son.” In
Critical Essays on Richard Wright’s “Native Son,” edited by Kenneth Kinnamon, 213–230. New
York.
Propp, Vladimir. (1928) 1958. Morphology of the Folktale. Bloomington, IN. Citations refer to the
1958 edition.
Prstojevic, Alexandre. 2017. “Pourquoi la fiction? Entretien avec Jean-Marie Schaeffer, directeur de
recherches au CNRS, auteur de Pourquoi la fiction? (Seuil, 1999).” Vox-Poetica. http://www.vox-
poetica.org/entretiens/intSchaeffer.html.
Raglan, Lord. (1936) 2003. The Hero: A Study in Tradition, Myth and Drama. Mineola, NY.
Reed, Joseph D. 1996. “Antimachus on Adonis?” Hermes 124: 381–383.
________
. 1995. “The Sexuality of Adonis.” Classical Antiquity 14, no. 2: 317–347.
Rehm, Albert, and Richard Harder. 1958. Didyma II: Die Inschriften. Berlin.
Reicher, Maria. 2010. “The Ontology of Fictional Characters.” In Revisionen 3: Characters in
Fictional Worlds: Understanding Imaginary Beings in Literature, Film, and Other Media, edited

by Jens Eder, Fotis Jannidis, and Ralf Schneider, 111–133. Berlin.
Rennie, Bryan S. 1996. Reconstructing Eliade: Making Sense of Religion. Albany, NY.
Richardson, Brian. 2010. “Transtextual Characters.” In Revisionen 3: Characters in Fictional
Worlds: Understanding Imaginary Beings in Literature, Film, and Other Media, edited by Jens
Eder, Fotis Jannidis, and Ralf Schneider, 527–541. Berlin.
Ritner, Robert K. 1998. “The Wives of Horus and the Philinna Papyrus (PGM XX).” In Egyptian
Religion: The Last Thousand Years; Studies Dedicated to the Memory of Jan Quaegebeur, edited
by Willy Clarysse, Antoon Schoors, and Harco Willems, 1027–1041. Leuven.
________
, ed. and trans. 1997. “The Apophis Battle.” In Hallo and Lawson 1997, 32.
Roper, Jonathan, ed. 2009. Charms, Charmers and Charming: International Research on Verbal
Magic. New York.
________
, ed. 2004a. Charms and Charming in Europe. New York.
________
. 2004b. Introduction to Roper 2004a, 1–6.
Rowe, Christopher. 1999. “Myth, History and Dialectic in Plato’s Republic and Timaeus-Critias.” In
Buxton 1999a, 263–78.
Rule, Sheila. 1989. “Sherlock Holmes’s Mail: Not Too Mysterious.” New York Times, November 5.
http://www.nytimes.com/1989/11/05/world/sherlock-holmes-s-mail-not-too-mysterious.html.
Rutherford, Ian. 2001. Pindar’s “Paeans”: A Reading of the Fragments with a Survey of the Genre.
Oxford.
Ryan, Marie-Laure. 2012. “Narration in Various Media.” In The Living Handbook of Narratology.
Hamburg. http://wikis.sub.uni-hamburg.de/lhn/index.php/Narration_in_Various_Media.
Saler, Michael. 2012. As If: Modern Enchantment and the Literary Prehistory of Virtual Reality.
Oxford.
Scarborough, John. 1979. “Nicander’s Toxicology II: Spiders, Scorpions, Insects and Myriapods.”
Pharmacy in History 21, no. 1: 3–34.
Schaeffer, Jean-Marie. 2010. Why Fiction? Lincoln, NE. (= Schaeffer, Jean-Marie. 1999. Pourquoi la
fiction? Paris.)
Scheid, John, and Jesper Svenbro. 1996. The Craft of Zeus. Cambridge, MA. (= Scheid, John, and
Jesper Svenbro. 1994. Le métier du Zeus: Mythe du tissage et du tissu dans le monde gréco-
romaine. Paris.)
Schjoedt, Uffe, Hans Stødkilde-Jørgensen, Armin W. Geertz, and Andreas Roepstorff. 2009. “Highly
Religious Participants Recruit Areas of Social Cognition in Personal Prayer.” Scan 4: 199–207.
Schmid, Hannah, and Christoph Klimmt. 2011. “A Magically Nice Guy: Parasocial Relationships
with Harry Potter across Different Cultures.” International Communication Gazette 73, no. 3:
252–269.
________
. 2010. “Goodbye Harry? Audience Reactions to the Ends of Parasocial Relationships: The
Case of Harry Potter.” Paper presented at the annual meeting of the International Communication
Association, Suntec Singapore International Convention & Exhibition Centre, Suntec City,
Singapore, June 22. http://www.allacademic.com/meta/p404037_index.html.
Schüle, Andreas. 2009. “The Divine-Human Marriages (Genesis 6:1–4) and the Greek Framing of
the Primeval History.” Theologische Zeitschrift 65, no. 2: 116–128.
Scullion, Scott. 2002. “ ‘Nothing to Do with Dionysus’: Tragedy Misconceived as Ritual.” Classical
Quarterly 52, no. 1: 102–137.

________
. 1999. “Tradition and Invention in Euripidean Aitiology.” Illinois Classical Studies 24–25:
217–233.
Seaford, Richard. 2009. “Aitiologies of Cult in Euripides: A Response to Scott Scullion.” In The
Play of Texts and Fragments: Essays in Honor of Martin Cropp, edited by J. R. C. Cousland and
James R. Hume, 219–234. Leiden.
________
. 1988. “The Eleventh Ode of Bacchylides: Hera, Artemis and the Absence of Dionysus.”
Journal of Hellenic Studies 108: 118–136.
________
. 1980. “Black Zeus in Sophocles’ Inachos.” Classical Quarterly 30: 23–29.
Segal, Robert. 1998a. Introduction to Segal 1998b, 1–14.
________
. ed. 1998b. The Myth and Ritual Theory: An Anthology. Oxford.
Shapiro, Jordan. 2014. “Why Disney’s ‘Maleficent’ Matters.” Forbes, June 5.
Sjöblom, Thomas. 2011. “Pumping Intuitions: Religious Narratives and Emotional Communication.”
In Geertz and Jensen 2011, 163–175.
Smith, Mark S. 2014. Poetic Heroes: Literary Commemorations of Warriors and Warrior Culture in
the Early Biblical World. Grand Rapids, MI.
Smith, Murray. 1995. Engaging Characters: Fiction, Emotion, and the Cinema. New York.
Smith, R. Scott. 2014. “Bundling Myth, Bungling Myth: The Flood Myth in Ancient and Modern
Handbooks of Myth.” Archiv für Religionsgeschichte 16: 239–258.
Smith, William Robertson. 1889. Lectures on the Religion of the Semites. Edinburgh.
Sourvinou-Inwood, Christiane. 1987. “Myth as History: The Previous Owners of the Delphic
Oracle.” In Interpretations of Greek Mythology, edited by Jan Bremmer, 215–241. London.
________
. 1979. Theseus as Son and Stepson: A Tentative Illustration of Greek Mythological Identity.
London.
Struck, Peter. 2004. Birth of the Symbol. Princeton, NJ.
Tatar, Mary, ed. 2002. The Annotated Classic Fairy Tales. New York.
Tolkien, J. R. R. 1947. “On Fairy-Stories.” In Essays Presented to Charles Williams, edited by C. S.
Lewis, 38–39. Oxford.
Updike, John. 1963. The Centaur. New York.
Valladolis, Julio, and Frederique Apffel-Marglin. 2001. “Andean Cosmovision and the Nurturing of
Biodiversity.” In Indigenous Traditions and Ecology: The Interbeing of Cosmology and
Community, edited by John A. Grim, 639–670. Cambridge, MA.
Vanstiphout, Herman. 2003. Epics of Sumerian Kings: The Matter of Aratta. Atlanta.
Vernant, Jean-Pierre. 1974. Mythe & société en Grèce ancienne. Paris. (= Vernant, Jean-Pierre. 1980.
Myth and Society in Ancient Greece. Baltimore.)
Versnel, Hendrik S. 1993. Transition and Reversal in Myth and Ritual. Leiden.
________
. 1987. “What Did Ancient Man See When He Saw a God? Some Reflections of Greco-
Roman Epiphany.” In Effigies Dei: Essays on the History of Religions, edited by Dirk van der
Plas, 42–55. Leiden.
Veyne, Paul. 1983. Did the Greeks Believe in Their Myths? An Essay on the Constitutive
Imagination. Chicago. (= Veyne, Paul. 1979. Les Grecs ont-ils cru à leurs mythes? Paris.)
Vidal, Gore. 1977a. “On Rereading the Oz Books.” New York Review of Books 24, no. 16.
________
. 1977b. “The Wizard of the ‘Wizard.’ ” New York Review of Books 24, no. 15.

Von Staden, Heinrich. 1993. “Spiderwoman and the Chaste Tree: The Semantics of Matter.”
Configurations 1, no. 1: 23–56.
Waller, J. D. 2015. “Echo and Historiola: Theorizing Narrative Incantation.” Archiv für
Religiongeschichte 16: 263–280.
Walsh, Richard. 2007. The Rhetoric of Fictionality: Narrative Theory and the Idea of Fiction.
Columbus, OH.
Watkins, Calvert. 2001. How to Kill a Dragon. Oxford.
West, M. L., ed. and trans. 2003. Homeric Hymns. Homeric Apocrypha. Lives of Homer. Cambridge,
MA.
________
. 1999. The East Face of Helicon: West Asiatic Elements in Greek Poetry and Myth. Oxford.
________
, ed. and comm.1966. Hesiod. Theogony. Oxford.
Westenholz, Joan Goodnick, ed., trans., and comm. 1997. Legends of the Kings of Akkade: The Texts.
Winona Lake, IN.
Weston, Jesse. (1920) 1993. From Ritual to Romance. Princeton, NJ. Citations refer to the 1993
edition.
Wheeler, E. L. 1982. “Hoplomachia and Greek Dances in Arms.” Greek, Roman, and Byzantine
Studies 23: 223–233.
Wolf, Mark J. P. 2012. Building Imaginary Worlds: The Theory and History of Subcreation. London.
Wolf, Werner. 2007. “Description as a Transmedial Mode of Representation: General Features and
Possibilities of Realization in Painting, Fiction and Music.” In Description in Literature and
Other Media, edited by Werner Wolf and Walter Bernhart, 1–90. Amsterdam.
________
. 2005. “Intermediality.” In Herman, Jahn, and Ryan, 252–256.
Wooffitt, Robin. 1992. Telling Tales of the Unexpected: The Organization of Factual Discourse.
Hemel Hempstead, UK.

Acknowledgments
THIS BOOK HAS benefited from the insights and critiques of my colleagues,
friends, and relatives.
I am lucky enough to teach at a university where there is a large and
engaging group of faculty and graduate students working on ancient
Mediterranean religions and myths. They have served as test audiences for
my ideas and helped me learn more about the myths of ancient cultures
other than Greece. I particularly thank Michael Beshay, Michael
Biggerstaff, David Brakke, Katie Caliva, Hanne Eisenfeld, Julia Nelson
Hawkins, Tom Hawkins, Warren Huard, Anthony Kaldellis, Matt Maynard,
Tina Sessa, Michael Swartz, Jimmy Wolfe, and Marcus Ziemann. My
greatest debts, however, are owed to Carolina López-Ruiz, who for many
years now has been teaching me about the intercultural transmission of
myths within the ancient Mediterranean world, and Tim McNiven, who, as
he has so many other times in the past, helped me better understand the
artistic representations of Greek myths and their relationships to literary
texts.
Other Ohio State colleagues made vital contributions to my work, as
well. I am grateful to Dorry Noyes and Jim Phelan, whose research gave me
useful models to think with and who talked me through some of those
models’ more challenging aspects. Team-teaching a seminar on narrative
with Jim in Autumn 2016 was a wonderful experience. Merrill Kaplan
welcomed me as an auditor into her course on Scandinavian myths, which
gave me a wider comparative outlook.
I am also indebted to scholars outside of my home university. Kathryn
Morgan has for almost thirty years been my ideal go-to person when I want
to talk about Greek myths and many other kinds of entrancing narratives,
even if we must usually do it at a distance. A semester of fellowship at the
Lichtenberg-Kolleg at the Universität Göttingen in 2012 introduced me to
Regina Bendix, Jason Mittell, and Annette Zgoll, all of whom opened my
eyes to new ways of approaching myths. The Lichtenberg-Kolleg also
reintroduced me, happily, to Antoine Cavigneaux, who shared some ancient

Near Eastern myths that were new, at least to me. Heinz-Günther
Nesselrath, one of my hosts in Göttingen, helped me with some of the
literary sources. I thank Jeff Kripal, Tanya Luhrmann, and Mike Murphy for
conversations we’ve had both inside and outside of colloquia at the Center
for Theory and Research at Esalen. I also thank Toni Bierl, Philippe
Borgeaud, Jack Emmert, David T. M. Frankfurter, Jessica Lamont, David
Leitao and John Furman, Sabina Magliocco, Elaine Pagels, Joy Reed, and
Antonia Syson. I am grateful to my hosts and audiences at the University of
Crete, Université de Gèneve, Columbia University, the University of
California–Los Angeles, the University of Washington, the University of
Minnesota, Wright State University, San Francisco State University, the
University of California–San Diego, Yale University, the Midwestern
Consortium on Ancient Religions, and the Universität Basel for their
questions and comments following presentations in which I tried out early
versions of my ideas.
The referees who read my manuscript for Harvard University Press
pushed me to think differently about some issues and thereby made this
book better than it would have been. Katie Caliva excellently served as
research assistant throughout the final year of the book’s development.
Some of what is in this book springs from discussions not with my
academic colleagues, however, but with my friends and family (not that
these groups are mutually exclusive, of course). It was conversations with
my brothers, Bob and James Iles, my brothers-in-law Wayne Johnson and
Eric Rudolph, and my son Tristan Johnston that helped me appreciate more
fully the similarities between ancient myths and contemporary television
shows. Kelly Allan, Alice Conklin, Barbara Haeger, and Geoffrey Parker,
with whom I have watched many a movie and Netflix episode, helped me
think through some of the arguments that I make in Chapters 3 and 5. My
son Pelham conceptualized a way of visually presenting an idea that I
introduce in Chapter 4 and created the charts that do so. My son-in-law
Daniel Gloor, an arachnologist, taught me a lot about spiders, which in turn
helped me understand poor Arachne a little better. My niece Rebecca Graf
helped me during an early stage of Chapter 3’s development. My
grandchildren, to whom this book is dedicated, have been, and continue to
be, audiences that compel me to think more carefully about how stories are
told.

There are also those who are no longer amongst the living but whose
presence I continued to feel as I finished this book—somewhat like the
heroes of Chapter 7, I suppose. As I worked, I often thought of Walter
Burkert and Martin West, each of whom contributed so much to the study of
Greek myths within a broader Mediterranean context, and each of whom
was always generous with help to a younger scholar. I was amused to
realize, upon finishing my manuscript, that Walter was invoked on both the
first page of the first chapter and the final page of the final chapter,
bookending everything else I said.
I will never be able to sufficiently express my gratitude for the early and
continuous support of my interest in myths by my late parents, Robert and
Phyllis Iles. Before I could read for myself, my mother gamely met my
repeated requests to read aloud Nathanial Hawthorne’s version of the story
of Pandora, which was included in an anthology of children’s stories that
we owned. As I grew older, my father steered me away from books about
Greek myths that he thought dumbed them down. I longed for a brightly
colored copy of D’Aulaires’ Book of Greek Myths, but (rightly) judging it to
be feeble, he instead put into my hands Padraic Colum’s The Children’s
Homer and The Golden Fleece and the Heroes Who Lived before Achilles,
with Willy Pogány’s marvelous line drawings. Later, he gave me Isaac
Asimov’s Words from the Myths and, later still, W. H. D. Rouse’s Gods,
Heroes and Men of Ancient Greece. All of these would probably seem
terribly out-of-date and unglamorous now, to generations lucky enough to
have been reared on Usborne’s eye-catching and excellently narrated series
of books of Greek myths, but for me, the books my father gave me were
wonderful. My parents’ willingness to occasionally contradict my teachers
was important as well. For career day in third grade, I did a project in which
I stated that I planned to become a mythologist. The teacher told me that
there were no such things as mythologists and made me do the project over
(the second time, I chose “mother,” which the teacher accepted). My
parents, however, told me to ignore what she had said about mythologists
and encouraged me to pursue both careers. (Are you reading this, Mrs.
Harmon? I’m two for two.)
Finally, my husband, Fritz Graf (who is colleague, friend, and relative all
rolled into one), was an indispensable sounding board during the years
when this book was developing. I don’t know what I would have done
without him.

Earlier versions of some discussions in this book have appeared in print
elsewhere. Portions of Chapters 3 and 5 were included in “Narrating Myths:
Story and Belief in Ancient Greece,” in Arethusa 48, no. 2 (2015): 173–
217, © 2015 by Johns Hopkins University Press. An earlier version of
Chapter 4 appeared as “The Greek Mythic Storyworld,” in Arethusa 48, no.
3 (2015): 283–311, © 2015 by Johns Hopkins University Press. A longer
version of the section on Arachne in Chapter 6 was published as “A New
Web for Arachne,” in Antike Mythen: Medien, Transformationen und
Konstructionen, ed. Ueli Dil and Christine Walde (Berlin, 2009), 1–22.

Index of Names and Terms
Abaris, 91, 214

Accretive characters, 27–29, 156–161, 170, 245–246, 284
Achilles, 40, 77, 171, 192, 218, 228, 248, 252, 255, 257, 258, 279; marries Medea and Helen on the
White Island, 132–133; narrates myths, 77, 212, 230
Actaeon, 203–204, 207, 212, 216
Admetus, 143, 165
Adonis, 36, 132–133, 218, 249
Aeacus, 229
Aedon, 188–191, 202, 207, 213
Aeetes, 3, 143, 259
Aegeus, 4, 120
Aegisthus, 174–175
Aegypius, 208
Aeneas, 242
Affordances, 31, 195–201, 203
Agamedes, 268–269
Agamemnon, 144, 196, 279
Agenor, 173
Aglaurus, 199–200
Agriania / Agrionia, 295n20
Aitia / aitiological approach, 22, 41, 48–49, 58–64, 120, 121, 133, 175, 184, 189–190, 194, 203, 208,
211, 217
Ajax, 162–163, 174, 258, 280
Alcestis, 243
Alcinous, 114
Alcyone, 177, 203, 213
Amazons, 145, 259
Amphiaraus, 146, 229
Amphitrite, 74, 112, 260, 272–273
Anat, 233, 236
Anchises, 107–109
Antaeus, 79, 252
Anthesteria, 48–49, 174–175
Anthos, 185, 213
Antigone, 94, 171
Antiochus, 225–226
Aphrodite, 132, 133, 154, 165, 199–200, 217, 249, 250, 273; and Anchises, 107–109, 269
Apollo, 17, 59–61, 78, 107–109, 115–116, 120, 127, 132, 142, 143, 146, 165, 166, 169, 180, 202,
209, 215, 221, 224, 229, 230, 238, 250, 253–254, 257; and Python, 79, 131, 261, 266–270

Apollonius of Tyana, 214
Apophis, 260–261
Appu, 234
Aqhat, 233, 236
Arachne, 31, 132–133, 177, 183, 187, 193–202, 207, 213, 216, 318n1
Arcas, 186, 208
Ares, 142–143, 145, 165, 166, 250, 320n31
Argo / Argonautica / Argonauts, 79, 80, 129, 134, 139, 227, 253, 254, 255, 274
Argos, 3, 118, 207, 227
Argus (dog), 221
Argus (herdsman), 132, 205, 261
Arion, 221
Aristeas, 91, 214
Aristodemus, 226
Arrhephoroi / Arrhephoria, 51, 199–200
Artemis, 132, 133, 146, 162, 180, 186, 204, 209, 230, 249, 250, 276
Arthur (king), 45, 241–242
Ascabulus, 318n1
Asclepius, 19, 31, 103–104, 146, 209, 303n135; Asclepius’s Rock, 118
Atalanta, 134, 173, 203, 218
Athena, 42, 110–111, 130, 132, 142, 143, 145, 146, 149, 165, 166, 194, 198–201, 209, 222, 229, 231,
240, 243, 244, 258, 259, 263, 266, 267, 269, 272–275
Athirat, 233
Atrahasis, 216, 278
Atreus, House of, 63, 140, 197
Audience attention, focused and unfocused, 13–17
Augeas, stables of, 254
Autolycus, 135, 306n42
Azi Dahaka, 241
Baal, 232, 236

Battus, 203, 302n106
Bellerophon, 115, 126–127, 129, 131, 132, 136, 173, 233, 235, 248, 253, 259, 265, 270, 272, 275,
277
Beowulf, 241
Biblical heroes, 216, 231, 237, 240, 242–243, 278
Bizarro Comics, 178
Bond, James, 148, 150, 151, 155, 309n10
Boreas, 61–62, 145
Bovary, Madame, 156
Brauron, 62, 186
Breaking Bad, 88
Bremmer, Jan, 51
Brisson, Luc, 57
Bulls (monstrous), 117–118, 126, 184, 185, 241, 254, 259, 270

Buphonia, 49, 62, 217
Burkert, Walter, 1–2, 4, 9, 22, 45, 50–55, 200, 277–280
Busiris, 253
Cadmus, 129, 131, 132, 142–143, 145, 212, 240, 253, 254, 255, 268, 269, 270
Calame, Claude, 20–21, 57–58, 67, 76–80
Callirhoe, 30, 182
Callisto, 179, 180, 186, 203, 208, 212, 216
Calvin and Hobbes, 247
Calydonian Boar / Boarhunt, 126, 134, 139, 184, 265, 276
Calypso, 96, 254, 255
Capulet, Juliet, 148, 151
Cassandra, 190, 228
Castor and Polydeuces, 165, 229
Centaurs, 115, 126, 127, 160, 181, 251. See also Chiron; Nessus
Cephalus, 249
Cerberus, 111, 115, 126, 131, 146, 181, 182, 221, 243, 244, 252, 265, 270, 272, 277
Cerynitian Hind, 115, 267, 270
Ceto, 181, 184, 265, 270–271
Ceyx, 177, 203
Charybdis, 112–113, 126, 247
Chelidon, 188–191, 202, 207
Chimaera, 30, 111, 115, 126–127, 131, 132, 177, 181, 182, 185, 259, 265, 267, 270, 272, 275, 277
Chiron, 17, 142, 229, 254
Chrysaor, 182
Circe, 3, 30, 96, 111–114, 159, 180
Clytemnestra, 16, 174–175, 196
Cognitive approaches to narrative, 10, 18, 28, 166–170, 281–282
Conceptual metaphor, 67, 73–75
Cornford, Francis, 38, 40, 47
Coronis, 115, 169
Cosmogonic myths, 31–32, 260–270
Crataeis, 30
Crazy Ones, 251
Creusa, 143, 173–174, 184
Croesus, 228–229, 288n32
Crommyonian Sow, 126, 184, 254, 258, 270
Cronus, 5, 154, 261, 263
Crossover characters, 137–143, 284
Cuchulain, 241
Cycnus, 253
Cyrene, 79, 80, 142, 171, 229
Danel, 233, 236

Daphne, 180, 183, 202

Deianira, 59, 120, 127, 135, 140, 251, 257
Deiphobus, 133
Delphic Oracle, 59, 60, 61, 109, 127, 221, 224, 261, 267–270, 295n20
Demeter, 11, 49, 51, 72–73, 107, 109, 133, 145–146, 165, 171, 187, 209, 269, 295–296n25, 320n31
Demophon, 109
Detienne, Marcel, 56–57, 187
Deucalion, 217
Dictys, 255
Diomedes (king of Thrace), 126, 184, 270, 324n92
Diomedes (son of Tydeus), 165, 166, 230, 248
Dionysus, 36, 39–40, 59, 107–108, 120, 162, 174, 180, 211, 225, 228, 230, 262, 272, 295n20, 307–
308n64; mysteries of, 7, 26, 133, 140
Dioscuri. See Castor and Polydeuces
Doctor Who, 140, 172
Downton Abbey, 92–94, 96, 246
Dracula, 42, 124
Dragon / snake, 126, 131, 142, 182, 185, 209–210, 212, 218, 234, 241, 242, 259, 260–264, 265, 270,
273–274. See also Apophis; Echidna; Fafnir; Hydra; Illuyanka; Ophion; Python; Typhon
Dumuzi, 232–233
Durkheim, Émile, 38–39, 41
Echidna, 181, 182, 265, 271

Eileithyia, 165
El, 232
Electra, 94
Eleusis / Eleusinian mysteries, 49, 75, 107–109, 140, 145, 238, 269, 295–296n25, 320n31
Elizabeth II (queen), 26, 151–153, 155, 252
Ender’s Game, 172
Eniautos-Daimon, 39–40
Enki, 232, 278
Enkidu, 218, 232, 241, 242
Eos, 165, 249
Ephialtes, 276
Epidaurus, 19, 103–105, 118, 209, 301n95, 305n18
Episodic narration, 24, 32, 68, 91–96, 134–136, 146, 246–260
Erechtheus, 80, 173
Ereshkigal, 232
Erichthonius, 134, 143, 200
Erigone, 140, 173–175
Erinyes, 41, 141, 191–192
Erra, 232, 277–278
Erythmanthian Boar, 126, 184, 252, 270
Eumenides, 49, 143
Euphemus, 80
Europa, 129, 131, 132, 144

Eurycleia, 5
Exorcist, The, 84, 284
Fafnir (Mimir), 241

Fairy tales, 1–3, 7, 12, 29, 30, 53, 122, 172, 173, 208, 214, 242
Fan fiction, 29, 172–173
Flood, 216–217, 232, 277–278
Fontenrose, Joseph, 46–48, 50, 54
Formative fiction, 23, 82–85, 119, 283–284
Forsyte Saga / Forsyte Chronicles, 3–4, 7, 135–136
44 Scotland St., 149–151
Frankenstein’s monster, 158, 160
Frazer, Sir James, 22, 35–40, 42, 47, 53, 65
Gaea, 135, 261, 264, 266, 277

Gaiman, Neil, 84, 128
Ganymede, 221
General Hospital, 135
Geryon, 30, 181–182, 227, 265, 270, 305n21
Getty Hexameters, 71–75, 77, 276
Gigantes, 261–264, 269
Gilgamesh, 232, 236, 239, 241, 242
Glaucus, 131, 230, 248
Golden Fleece, 209, 255, 259, 273–274
Gorgons, 38, 126, 131, 143, 181, 222, 255, 259, 263, 265, 271, 275. See also Medusa
Graeae, 222, 265, 271
Graf, Fritz, 51, 57
Grendel, 241
Griffins, 126, 262
Hades (god), 11, 126, 164–166, 171, 217, 222, 243, 250, 272
Hades (Underworld), 9, 40, 61, 75, 115, 131, 133, 140, 146, 160, 165, 204, 213–214, 233, 235, 238,
239, 242–244, 250, 258, 272, 277, 279
Haemon, 94
Hannahanna, 232
Harpies, 126
Harrison, Jane Ellen, 22, 38–45, 47, 49, 50, 53, 61
Harry Potter series, 20, 88, 172
Hecabe, 31, 191–193, 203
Hedammu, 232, 260
Helen, 96, 131, 133, 144, 145, 155, 163, 224, 228
Helenus, 191
Helios, 130, 164, 165, 247, 272

Hera, 62, 65, 132, 162, 164, 165, 169, 171, 180, 181, 182, 185–186, 253, 266, 272, 276, 279–280,
299n69
Heracleidae, 145, 226
Heracles, 3, 18, 25, 40, 45, 57, 62, 76, 79, 80, 117, 122, 126, 129–130, 131, 137, 145, 146, 149, 155,
163, 165, 173, 182, 219, 223, 225–229, 235, 238, 241, 248, 251, 252–270, 272, 274–277, 324n92;
apotheosis of, 240, 243; fetches trees from Hyperborea, 61, 162, 251; gigantomachy, 262–265, 269;
journey to Underworld, 61, 131, 133, 135, 140, 146, 242–244, 258, 272, 277; killed by Deianira,
59, 120, 131, 135, 140, 142, 250–251; releases Prometheus, 136–137
Herder, Johann Gottfried, 6, 22
Hermes, 27, 130, 135, 222, 243, 244, 253–254; cattle thief, 60, 132, 166, 221, 302n106; kills Argus,
132, 205, 261; plurimedial character, 154; special sandals, 126, 165–166, 222, 272
Herse, 199
Hesiod and the Muses, 102, 105–106
Hesperides, 129–130
Heyne, Christian Gottlob, 5–6, 8
Hippodamia, 133
Hippolytus, 40, 117–119, 132, 133, 235, 249
Historiola, 66–67, 68–75
Hitchcock, Alfred, 95
Holmes, Sherlock, 24, 26, 88, 139, 141–142, 148, 150–151, 153, 155, 247, 250–251, 298n57,
324n98; as plurimedial character, 155–158, 160
Homeland, 92, 96, 246
Hooke, Samuel Henry, 44
Horus, 68–70, 72–73
Humbaba, 241
Hupasiya, 234, 263
Hyacinthus, 203
Hybrids, 177–183
Hydra, 18, 181, 182, 221, 250–251, 252, 265, 270, 275, 276
Hyman, Stanley, 45–47
Hyperborea, 59, 61, 115, 129, 162, 213, 247, 251, 255
Hyperseriality, 134–139, 284
Hypnos, 253
I, Claudius, 135
Icarius, 174
Idas, 228
Illuyanka, 232, 234, 260, 263
Inanna, 232–233, 241
Inara, 232, 234, 260, 263
Initiation paradigm, 39–40, 50–51, 53
Ino / Leucothea, 324n87
Io, 186, 203–205, 207, 212, 216, 228
Iolaus, 117
Ion, 173–174, 223, 257

Iops, 223, 235
Iphigenia, 62, 129
Ishtar, 232–233, 241
Ishum, 232, 277–278
Isis, 68–72, 74, 186, 212
Isyllus, 209
Iteration, 252–260
Itys / Itylus, 188
Ixion, 115, 181
Iynx, 185, 187, 193
James, M. R., 98–101, 104–105, 119

Jason, 3, 95, 129, 131, 136, 142–143, 223, 227, 253, 254, 255, 257, 259, 270, 272–274
Jeeves, 150
Jesus Christ, 82–83, 178, 231, 240, 242
Kataphatic prayer, 85–86

Kennings, 80–81
Keret, 233
Kessi, 234
King, Stephen, 135, 139, 300n84
Kirk, Geoffrey, 48–49
Kluckhohn, Clyde, 45–46
Kore. See Persephone
Kouretes / kouros, 39–40
Kumarbi, 232
Labyrinth, 272
Laestrygonians, 261
Laius / House of Laius, 63, 197, 251
Lamia, 276
Leach, Edmund, 46
League of Extraordinary Gentlemen, The, 139
Leda, 277
LeGuin, Ursula K., 84
Lemminkäinen, 242
L’Engle, Madeleine, 84
Leto, 165, 180, 186, 208
Leucippus, 173, 208
Lévi-Strauss, Claude, 55–56
Lot’s wife, 177, 179, 186
Lovecraft, H. P., 143; Dunwich Horror, 184
Lugalbanda, 232, 236, 239
Lycaon, 146, 164, 183, 185, 193, 202, 203, 209, 216

Lycurgus, 230, 248
Mad Men, 93
“Making believe” (feintise ludique), 20–21
Malinowski, Bronislaw, 45–46
Mallarmé, Stéphane, 82
Marduk, 44, 233, 260, 278
Mares of Diomedes. See Diomedes (king of Thrace)
Maximal and submaximal characters, 150–152
Medea, 3, 4, 29, 62, 81, 127, 131, 133, 136, 144, 171–172, 173
Medusa, 115, 134, 136, 178, 181, 182, 222–223, 265, 270, 271, 272
Megara (city), 80
Megara (Heracles’s wife), 257
Melampus, 162, 295n20
Meleager, 131, 133, 135, 140, 142, 228, 241, 248, 258, 265, 276, 306n28
Meleagrides, 208
Memorate, 97–119, 283, 301n95
Menelaus, 109, 133, 224, 226, 228, 235, 240
Menoetes, 182
Mentor, 110
Mestra, 214
Metanira, 107, 212
Middle-Earth, 125
Middlemarch, 92, 95, 96, 246
Milk, 69–70, 72–75
Mimir. See Fafnir
Minos, 17, 144, 145, 160, 162, 258, 272
Minotaur, 17, 30, 126, 131, 134, 160, 181–183, 255, 258, 270, 272
Mintha, 187
Minyas, daughters of, 208
Moirai, 143
Monster fighting, 46, 128, 184–185; by gods, 32, 127, 260–270; by heroes, 32, 111–115, 126–127,
241–242, 252–255, 263–265, 270–277
Mormo, 276
Moses, 45, 105, 153, 231, 233
Murray, Gilbert, 38, 40, 42, 45, 47
Muses, 20, 21, 102, 105–106, 211, 302n105
Myrrha, 179
Names, importance of, 2–3, 8, 28–29, 170–176, 183–187, 189, 207–208, 236, 242, 255
Naram-Sin, 233, 236
Narnia, 86, 136, 141
Nausicaa, 209
Nemean Lion, 126, 182, 184, 221, 254, 265, 267, 270
Neoptolemus, 40, 49, 59–60, 248

Nephele, 209
Nephthys, 69–70
Nereus, 215
Nergal, 232
Nesbit, E., 141, 307n49
Nessus, 127, 251
Network, mythic, 2–3, 8, 131–138
Nietzsche, Friedrich, 38, 41
Ninsun, 239
Niobe, 77, 132, 179, 204, 206, 212
Nisus, 80
Noah, 216, 278
Noyes, Dorothy, 12–17, 98, 145, 282–283
Oceanus, 132, 271

Odin, 239, 241, 323n73
Odysseus, 3, 5, 9, 37, 51, 93–94, 96, 129, 134, 135, 164, 166, 182, 192, 209, 213, 221, 223, 241, 250,
253, 261, 276; battles Scylla, 30, 112–115, 247, 265, 270; journey to Underworld, 131, 243, 277; as
plurimedial character, 159; resourcefulness, 16, 114, 252, 254–255, 270, 279; storyteller, 9, 109–
115, 127
Oedipus, 45, 117, 139–140, 141, 223, 227, 251, 253, 270
Oenomaus, 60
Old Curiosity Shop, The, 87–88, 92, 246
Ontological templates, 166–170
Ophion, 261, 263
Oreithyia, 62, 145
Orestes, 45, 49, 171, 174–175, 251, 252
Orpheus, 133, 135, 227, 228, 243
Orthrus, 181, 182, 265, 270
Oschophoria, 160–161
Ouranos, 154, 266
Oz, 25, 82, 124–125; Wicked Witch of the West, 95
Pan, 102–103, 146, 180

Panathenaia, 199–201
Pandareus, 188, 249–250
Pandava brothers, 239–240
Pandion, 188–189, 208
Pandora, 56
Pandrosus, 199–200
Parasocial interaction and parasocial relationships, 24, 87–91, 284
Paris, 131, 133, 218
Pasiphae, 30, 181–183, 228
Pegasus, 30, 115, 126, 131, 132, 136, 137, 177, 181, 182, 221, 272
Peleus, 171, 218, 240

Pelops, 17, 40, 60, 80, 117, 133, 162, 165, 240
Penelope, 9, 16, 188, 190
Periclymenus, 214
Persephone (Kore), 10–11, 26, 49, 72–73, 75–76, 78, 131, 132, 133, 140, 164–165, 171, 209, 217,
229, 230, 238, 250, 253, 295n25
Perseus, 3, 115, 126, 129, 131, 133, 134, 173, 222–223, 226, 227, 235, 247, 253, 255, 259, 263, 265,
270, 272
Phaeacia / Phaeacians, 9, 109, 111, 114, 129, 166, 209
Phaedra, 221, 303n135
Phaedrus, 61–62, 145
Phaethon, 213, 249–250
Phalanx, 194–202, 216
Phanes, 135
Phidippides, 102–104, 146, 298n63
Philoctetes, 95, 248, 257
Philomela, 145, 189–190, 202–203, 207–208, 216
Phoenix, 248
Phorcys, 181, 184, 265, 270–271
Pirithous, 131, 243, 253
Pittheus, 249
Plurimediality, 27–28, 156–163, 170, 245–246, 284, 309n12
Polydeuces. See Castor
Polyidus, 218
Polymestor, 191
Polyphemus, 16, 37, 51, 93–94, 134, 164, 182, 252, 254, 255, 303n118
Poseidon, 81, 118, 132, 145, 162, 164, 166, 181, 182, 214, 240, 249, 253, 271–272, 275, 276, 279
Pragmatic speech, 21–22, 58, 67, 76–78
Prajapati, 178
Priam, 77, 193, 230
Primary and Secondary Worlds, 26, 122–125, 128–130
Procne, 145, 189–190, 202–203, 207–208, 216
Proerosia, 62
Proetus, 126; daughters of, 51, 162, 228
Prometheus, 136–137, 154, 164, 212, 217, 221, 253
Propp, Vladimir, 52, 55
Proteus, 109, 215
Protogonos, 135
Pyrrha, 217
Pythagoras, 163, 213–214
Python, 79, 126, 131, 182, 184, 261, 266–270
Questing Beast, 241
Ra, 235, 260–261

Raglan, Lord, 45, 47

Rama, 241, 323n74
Ravana, 241
Rhetorical poetics, 68, 96–97
Robert-Houdin, Jean-Eugéne, 23, 82–83, 87, 90–91, 283
Rp’u / Rephaim, 236
Rudra, 178
Rustam, 241, 318n49
Sacrifice, aitia for, 217–218

Sarapis, 212
Sargon, 233, 236, 241
Sarpedon, 165
Sciron, 118, 126, 184, 258
Scylla, 30, 111–115, 126–127, 179–180, 207, 247, 262, 265, 267, 270, 276
Selinus, 72, 295n13
Semnai Theai, 117
Series and serials, 32, 87–88, 93–96, 246–260, 285
Setna, 235
Shape-shifting, 30, 214–215
Shipwrecked Sailor, 234
Sigi, 239
Sigurd, 239, 241
Sinuhe, 234
Sirens, 112, 113, 126, 247, 262, 270
Sita, 241
Sleeping Beauty, 1–3, 7, 172
Smith, W. Robertson, 22, 35, 38, 41, 44, 47
Society for Psychical Research, 42
Socrates, 61–62, 145
Solymoi, 115, 259
Sought and unsought narrative experiences, 12–16, 97–98, 283
Sourvinou-Inwood, Christiane, 51
Sphinx, 126, 181, 223, 265, 267, 270
Spiders, 183–184, 187, 193–202, 207, 216
Star Trek, 124, 172
Star Wars, 3–4, 124, 309n12
St. George, 241
Strangers and Brothers series, 135–136
Stranger Things, 93, 96
Stygian nymphs, 222
Stymphalian Birds, 270, 274
Superheroes, 83–84, 165, 252
Symplegades, 126, 129
Syrinx, 180

Tammuz, 233
Tantalus, 17, 162, 193, 249
Tarhun, 260
Tartarus, 265, 266
Tecmessa, 257
Telemachus, 110–111
Telephus, 218
Telipinu, 232
Telphousa, 165
Tereus, 189–190, 216
Teshub, 234, 260, 263
Tethys, 132, 271
Teumessian Vixen, 126, 184, 254, 270
Theological correctness, 166–169
Theseus, 3, 4, 40, 62, 81, 131–134, 155, 173, 199, 223, 227–228, 237, 241, 249, 253, 254, 255; father
of Hippolytus, 117–119, 132, 160; journey from Troezen to Athens, 120, 159, 258, 270; journey to
Underworld 131, 160, 243, 306n28, 306n38; kills Minotaur, 131, 134, 160; as plurimedial
character, 28, 159–163, 245–246; undersea visit, 17, 78, 115, 159, 162, 258, 260, 272–273, 289n35;
voyage to Crete and back, 17, 95, 120, 129, 159–61, 162, 258
Thesmophoria, 11, 75–76, 230
Thetis, 215, 249, 277, 320n26
Thoricus, 223
Thraetaona, 241
Thymoetes, 223, 225, 235
Tiamat, 260
Tiresias, 94, 317n131
Titans, 229, 264, 266
Tithonus, 165
Tolkien, J. R. R., 26, 84, 122–125, 128, 129–130. See also Middle-Earth; Primary and Secondary
Worlds
Torchwood, 140
Trophonius, 268–269
Typhon, 111–112, 126, 127, 131, 181, 182, 261, 263–271
Ullikummi, 232, 260

Updike, John, 178, 207
Utnapishtim, 216, 242, 278
Vernant, Jean-Pierre, 55–56

Verrall, Arthur and Margaret, 42
Versnel, Hendrik, 51, 53–54
Vineyard Church, 85–89, 284
Weaving, 187–194, 196–201, 202, 213

Weston, Jessie, 45
Willing suspension of disbelief, 123
Wolf Hall, 82
Xuthus, 184
X/Y format, 98–119, 300n84
Zagreus, 39–40
Zal, 318n149
Zeus, 18, 29, 32, 57, 115–116, 131, 132, 137, 149, 154, 155, 165, 171, 180, 181, 184, 185–186, 193,
204, 217, 221, 222, 224, 226, 229, 240, 249, 250, 276, 277, 320n26; attacks and succeeds his
father, 5, 175–176; and Cretan Hymn to the Kouros, 39–40; seduction by Hera, 132, 164, 169, 253,
279–280; Typhon fought by Zeus, 127, 261, 263–264, 265, 266, 269–271
Ziusudra, 216, 278

Index Locorum
Note: Passages are indexed according to where they appear in the notes, and also, in many cases,
where they appear, are discussed, or are alluded to in the text.
GREEK AND LATIN TEXTS
Acusilaus (FGrH 2)
F26: 316n105
F27: 316n105
F28: 162, 309n20
F33: 316n105
Aelian
NA
1.21: 314n76, 315n79
6.57: 315n79
17.11: 315n83
VH
1.2: 315n79
12.20: 188, 313n35
12.42: 318n148
13.1: 318n148
Aeschylus
Ag.
1–20: 257, 325n116
160–176: 175–176, 230–231, 320n33
611–612: 16–17, 288n33
1144–1145: 313n38
1234: 127, 305n15
1492: 196, 315n81
1516: 196, 315n81
Daughters of Phorcys (fr. 261 TrGF)
319n23
Eu.
228
1–8: 327n160
727–728: 143, 307n54
Pers.
78

Pr.
167–195: 164
673–674: 212–213, 317n124
687–692: 203, 316n103
907–917: 164, 310n28
944–953: 164, 310n28
Prometheus Unbound
319n23
Suppl.
58–65: 189, 313n38
291–299: 186, 312n21
565–570: 203, 316n103
887: 196, 315n81
Th.
541: 267, 327n154
773: 223, 319n7
777: 267, 327n154
Aëtius Medicus (Corpus medicorum Graecorum 8.1)

13.20: 315n83
Alcaeus
Hymn to Hermes
60, 293n78
Alciphron
4.16: 288n30
Alcman
fr. 21: 306n28
fr. 34: 317n117
fr. 307: 320n29
Amphis
fr. 47: 312n22
Andron
fr. 8: 318n141
Androtion (FGrH 324)

F16: 62, 293n86
Anthologia Palatina
6.39.3: 314n78
9.372: 314n78, 315n81, 315n87

Anticlides (FGrH 140)
F17: 318n150
Antigonus
Mirabilia
87: 315n82
Antiphanes
fr. 189: 78, 296n28
Antoninus Liberalis
7: 312n20
11: 313n37, 313n39
17: 208, 317n112
23: 302n106
24: 318n1
35: 312n23
Apollodorus
1.5.1: 320n32
1.7.2: 318n141
1.8.3: 225–226
1.9.16: 306n40
1.80–83: 306n40, 317n114
2.1.3–4: 312n25
2.4.3: 327n163
2.5.5: 325n107
2.5.6: 328n172
2.7.4: 318n148
2.77–80: 267, 327n154
3.3.1–2: 318n151
3.8.2: 312n26
3.9.2: 318n148
3.14.4: 306n30
3.14.8: 313n38, 313n42
Epit. 3.7: 325n110
Epit. 5.23: 313n56
Apollonius Rhodius
1.367–393: 325n108
1.503–506: 326n138
1.526–527: 328n169
Apostolius, ap. Leutsch Paroim. Gr.

2.497–498: 328n173

Apuleius
Golden Ass.
317n131
Aristides
schol. p. 41 Dindorf: 328n173
Aristonous
Paean to Apollo
59, 293n76
Aristophanes
Lys.
199
Ra.
228
1314: 315n87
fr. 567 Edmonds: 192, 314n63
fr. 594 Edmonds: 309n16
schol. Pax 758: 328n173
Aristotle
HA
542a12–17: 316n100
555a23–26: 197, 315n82
555b10–15: 197, 315n82
622b23: 314n76
622b28–623b1: 315n79
623a8: 314n78
Ph.
199a20–22: 315n79
Po.
63
1451a37–1451a38: 297n47
1460a12–1461b25: 297n47
Arnobius
5.5–7: 320n31
Athenaeus
101e: 288n30
128b: 288n30
253a: 288n30
577d–f: 288n30

Bacchylides
1
228–229, 319n24
3
228–229, 319n24
54–59: 115, 303n124
5
60–61, 133, 135, 228–229, 258, 306n28, 319n24, 325n122
11
228–229, 319n24
53–58: 162, 309n20
92–95: 162, 309n20
13
228–229, 258, 319n24, 325n122
15
227–228, 319n19
16
59, 120, 227–228, 319n19
17
78, 95, 159, 227–228, 258, 272, 289n35, 309n18, 319n19, 325n122
117–118: 115, 259–260, 303n124
18
120, 159, 227–228, 258, 319n19, 325n122
19
227–228, 316n104, 319n19
20
227–228, 319n19
23
227–228, 319n19
25
227–228, 319n19
26
227–228, 306n27, 312n7, 319n19
27
227–228, 319n19
28
227–228, 319n19
fr. 4.69–70: 314n77
fr. 10: 306n27
fr. 20D:316n104
Callimachus
Aet.
fr. 43: 293n83
fr.178: 293n83

Epigr. 57
312n 21
Hec.
fr. 232–233: 306n38
Hymns
4.171–184: 308n65, 317n116
6.1–6: 320n32
fr. 233: 306n29
fr. 632: 312n26
fr. 685: 312n20
Charon of Lampsacus (FGrH 262)

F12: 316n110
Cicero
Div.
1.37: 308n65, 317n116
Clement of Alexandria
Protr.
2.17: 318n145
Strom.
1.16.78.5: 319n17
Corinna (PMG)
fr. 665: 316n104
fr. 671: 228, 319n21
fr. 672: 319n7
Cypria
fr. 1 W: 277, 328n179
Deilochus (FGrH 471)

F6: 316n105
Demon (FGrH 327)

F1: 319n4
Demosthenes
60 (the Funeral Oration)
6–8: 145, 307n62
28: 145, 307n62
30: 190, 208, 313n48, 317n111
31: 225–226, 319n13

Diodorus Atheniensis Periegeta (FGrH 372)
F28: 319n4
Diodorus Siculus
4.13.3: 325n107
4.37: 225–226, 319n13
4.42.1: 267
5.4.7: 11, 288n26, 320n32
20.3–5: 276, 328n173
20.41: 328n173
22.9: 308n65, 317n116
Diogenes Laertius
59: 211, 317n119
Dioscorides
De iis
4: 315n83
De Materia Medica
5.99: 295n17
Duris (FGrH 76)

F17: 328n173
Epicharmus
fr. 113: 318n141
Epimenides (DK)
fr. 3: 327n167
frs. 27–29: 126, 304n12
Eratosthenes
Cat.
1: 312n22
8: 310n26, 312n18
Eumelus
fr. 21 West: 325n109, 325n123
Euripides
Alc.
324n92
77–212: 257, 325n114
1578–1612: 117, 303n130
Andr.

252
Ba.
228
1388–1392: 169, 311n42
Cretans
fr. 81 Aus.: 182, 312n13
Elec.
299n69
Erechtheus
77–80
Hec.
324n90
1259–1273: 191, 313n50
Hel.
96
1301–1366: 230
Heracl.
844–866: 117
HF
263
1–86: 257
375: 267
Hipp.
10–59: 324n94
117–119: 160, 221
732: 317n125
887–898: 118
1173–1254: 117–118
1416–1422: 306n30, 324n94
Ion
319n8
102–184: 257
184–218: 293n83, 305n18
985–1010: 143, 230–231, 263
IT
252: 324n90
1089–1092: 317n126
Med.
1–88: 257
62: 293n85
1359: 127
Meleager
306n40
Phoen.
307n47
Suppl.

160
fr. 369 Nauck: 314n77
fr. 930 Nauck: 212, 317n123
fr. 968 Nauck: 192, 314n62
schol. Hec. 3: 191
Eustathius
Od. 1875.15–27: 190
Eutecnius Sophista
Paraph. Nic. Ther. 59
315n83
Hecataeus (FGrH 1)
F6: 327n154
Helladius
Ap. Phot. Bibl. 531
313n37, 313n40
Hellanicus (FGrH 4)
F*38: 62, 293n86, 319n4
F51: 306n27
F117a: 318n141
F168a: 306n28
Herodotus
144
1.1: 312n21
1.23–24: 319n2
1.46–51: 310n26
2.44: 163
3.102–105: 304n11
4.105: 304n11, 317n133
6.52: 319n14
6.105–106: 102–103, 298n 63, 308n65
7.204: 319n14
8.44: 319n8
8.131: 319n14
Hesiod
Cat. (ed. Merkelbach and West)
159, 203–204
fr. 10a.20–24: 319n8
fr. 25: 306n28

fr. 33a: 317n132
fr. 35: 317n132
frs. 37.10–15: 162, 309n20
fr. 43a: 263, 272, 304n14, 317n132
fr. 68: 317n114
fr. 124: 306n32
fr. 126: 306n32
frs. 131–133: 162
fr. 135: 319n3
frs. 140–141: 306n27
fr. 145: 306n27
frs. 161–168: 185, 310n26, 312n19
frs. 190–191: 133, 306n35
frs. 195.28–29: 263
fr. 204.98–101: 277, 328n180
fr. 256: 302n106
fr. 275: 317n131
fr. 280: 306n28
MW
fr. 312: 188, 313n35
Op.
156–165: 328n179
475: 314n77
568: 188, 313n33
777: 314n76
Sc.
216–237: 272
220: 310n35
Th.
269
22–34: 105–106, 301nn98–99
188–192: 154, 309n9
238: 271, 327n161
270–273: 271, 305n21, 327n161
270–336: 181, 312n6, 312n16, 326n145
277–278: 181, 271, 327n162
280–283: 131, 265, 306n26, 319n3
288–289: 312n10, 327n162
289–294: 265, 326n146
297–298, 271, 327n162
304–325: 131, 265, 304n14, 306n25, 312n10, 312n15, 327n162
325–329: 265, 267, 312n15, 319n7
332–333: 265, 327n162
523–525: 319n2
535–564: 217, 318n142
820–880: 112, 182, 265, 303n117, 312n14

954: 263, 326n143
979: 312n10
1011–1018: 96, 299n75
Hesychius
Kallynteria: 316n95
Plynteria: 316n95
protonion: 316n95
Thoricus: 319n6
Hippocrates
De morbis mulierum
1.120: 295n17
Homer
Il.
96
1.1–7: 279, 311
2.308–320: 316n103
2.450–452: 310n36
2.780: 265
2.782–783: 302n15
3.143–144: 306n38
3.242: 133
5.330–352: 165, 324n95
5.370: 154
5.392–404: 165, 324n95
5.439–442: 165, 324n95
5.844–863: 310n36, 324n95
6.130–141: 230, 248
6.145–211: 248
6.171–183: 111, 126–127, 131, 265, 302n15
6.197–211: 277
8.362–369: 265, 302n15
9.529–655: 248, 276–277, 296n27
12.292: 306n27
14: 164
14.153–360: 306n31
14.153–377: 253
14.321–322: 306n27
16.326–329: 267, 302n15, 304n13
16.431–461: 165
19.418–424: 255–257
24.212–213: 192
24.334–335: 154
24.339–345: 310n35

24.599–562: 77
24.602–609: 230
24.617: 212
Od.
109–11
1.1–9: 279
1.25–30: 164
1.96–98: 310n35
3.371–373: 110–111, 316n103
4.214: 287n11
4.561–569: 319n9
5.43–49: 310n35
5.204–224: 255–257
5.234–262: 254
5.333–375: 245
6.229–231: 209
8.280: 314n78
9: 94–95
9–12: 111
9–13: 261–262
9.62–233: 303n118
9.318–330: 254
9.526–535: 164, 182
10.118: 261
10.133–188: 303n118
11: 94
11.488–503: 255–257
11.601–627: 265, 272, 277, 302n15, 306n28
11.630–631: 306n28
12: 247
12.69–72: 3, 272
12.85–99: 112
12.124–126: 267, 312n12
12.235–260: 113–114
13.1–6: 114
16.35: 314n77
17.290–327: 319n2
19.394–398: 135
19.502: 287n11
19.518–523: 118, 313n36, 313n41, 313n46
19.570–575: 16
20.17–18: 314n66
23.187–204: 254
23.301–341: 9
schol. Il.
1.5–6: 328n179

2.629: 325n107
10.334: 318n148
11.700: 325n107
schol. Od.
7.59: 326n141
19.518: 189
Homeric Hymns
107–109, 229, 320n28
HHAp.
79: 221, 266–270, 324n93
88–125: 266, 310n30, 326n150
97–101: 162, 310n30
186–206: 250, 324n97
207–215: 60–61, 267, 293n82, 327n155
216–228: 268, 327n156
254–274: 310n30, 327n159
294–300: 268, 327n159
302–354: 266, 326n149, 326n152, 327n153
351: 182, 312n15
352: 266, 326n152
353–354: 131, 306n25
355–356: 266, 327n153
362–363: 79
364–366: 266, 327n153
368: 131, 304n14, 306n25
375–381: 162, 310n30
382–385: 327n159
388–447: 107–108, 302n108
400–450: 215, 317n134
514–419: 59, 293n76
HHAphr.
77, 109
75–184: 108, 302n109
202–217: 318n2
218–238: 165, 310n32
HHDem.
293n78
15–18: 164–165, 310n29
22–25: 164–165, 310n29
27–29: 164–165, 310n29
54–58: 164–165, 310n29
74–81: 164–165, 310n29
98–295: 107, 302n107
235: 109, 302n111, 317n120
243–249: 212, 317n120

305–313: 212, 317n120
371–374: 164–165, 310n29
405–413: 164–165, 310n29
HHDion.
211–212
HHHerm.
221, 293n78, 306n31, 324n93
75–86: 310n36
87–93: 302n106
138–140: 310n36
145–147: 166, 310n37
187–211: 253–254, 302n106, 310n31
190–200: 165, 310n31
Hymn 7
1–53: 108, 302n110
Hymn 19
324n93
Hymn 27
317n117
Hymn 28
324n93
Hyginus
Astr.
2.7.3: 306n30
Fab.
30: 267, 327n154
45: 313n45
95: 325n110
252: 318n148
Ibycus (PMG)
fr. 291: 306n29
Inscriptions
Getty Hexameters (ed. Faraone and Obbink)
71–75
fr. i, Side B, 45–46: 276, 328n175
Gold Tablets (ed. Graf and Johnston)
3, 5, and 26a–b: 295n17
Inscriptiones Graecae
3
I
976–980 (Archedamus inscription): 298n63
2
IV : 1

126 (a healing miracle): 301n95
128 (Isyllus inscription): 209, 317n117
XI.4
1299 (Sarapis miracle from Delos): 212, 317n121
Lindos II. Inscriptions, ed. Chr. Blinkenberg
2 (The Lindian Chronicle): 145, 307n61
The Parian Chronicle (FGrH 239)
145, 320n31
3
Sylloge Inscriptionum Graecarum 398 (invasion of Delphi by Gauls)
308n65, 317n116
Supplementum Epigraphicum Graecum xxxiii 147 (the hero Thoricus)
319n6
Ion of Chios
fr. 1 Page: 299n69
Isocrates
Paneg.
4.27–31: 62, 293n84
Lucian
Asin.
317n131
VH
2: 295n17
schol. 275–276 (Rabe): 318n145
Lycophron
143: 133, 306n34
152–155: 310n32
174: 133, 306n29, 306n34
333: 313n52
798: 131–132, 306n29, 306n34
1174–1188: 192, 313n53
1391–1396: 317n132
Menander Rhetor
iii.340: 320n29
Menecrates (FGrH 769)

F2: 312n23, 316n105
Musaeus (Bernabé)
fr. 100: 306n27

Myrtis (PMG)
fr. 716: 228
Mythographi Vaticani
I 4: 313n45
II 261: 313n45
Naupactica (West)
fr. 10: 303n135
Nicander (Gow and Schofield)

Ther.
484: 318n1
frs. 43–62 (Met.): 208
fr. 62: 313n51
schol. Alex. 375: 313n28
schol. Ther. 12a: 194
schol. Ther. 715a: 315n82
schol. Ther. 721–724: 315n83
Nicolaus Damascenus (FGrH 90)

F48: 319n4
Oppian
H. 3.486: 313n28
Ovid
Am. 1.14.7
315n87
Fast. 2.177–178
312n26
Met.
1.111: 295n17
1.319: 318n141
1.416–421: 217, 318n142
2.476–488: 186, 212, 213, 312n26
2.485–494: 216, 318n139
3.200–201: 212, 317n122
3.203: 213, 317n127
4.285–388: 312n5
4.790–803: 327n163
5.446: 318n1
6.1–145: 198, 312n26, 318n1
6.143: 315n87

6.148–149: 132–133, 306n33
6.339–381: 186, 312n23
6.403–411: 165, 310n32
6.424–676: 313n38, 313n42
8.620–724: 318n1
8.738–878: 317n132
10.728: 313n28
13.565–575: 313n54
15.158: 213
15.167: 213
15.172: 213
15.175: 213
Palaephatus (Festus)
23: 317n132
Papyri
P. Amherst II col. ii (A)
294n12
P. Berol. 7504
294n12
Papyri Graecae Magicae
IV.94–153: 70, 294n9
VI.1–47: 294n8, 294n12
P. Oxy 56.3830:
318n150
Parthenius
11: 313n32
Paulus Aegineta Medicus

Epitomae medicae libri septem
5.6t1: 315n83
Pausanias
1.4: 308n65, 317n116
1.5.2: 225–226, 319n13
1.5.4: 313n38, 313n42
1.27.3: 200, 316n93
1.28.7: 139, 307n46
1.39.1: 145–146
1.41.8: 313n38, 313n42
2.4.3: 225–226, 319n13
2.18.9: 319n4
2.27.2: 305n18

2.27.3: 301n94
2.35.11: 145–146
3.12.5: 223, 319n5
6.26.7: 314n78
7.17.10–12: 320n31
8.2.1–4: 145–146, 307n64
8.3.6–7: 316n110
8.34: 324n101
8.38.7: 312n19, 317n133
9.5.9: 313n36
9.31.5: 306n28
10.23–24: 308n65, 317n116
10.28.3: 306n28
Pherecrates
fr. 49: 309n16
fr. 124: 313n36
fr. 142: 314n77
Pherecydes of Athens (FGrH 3)

11, 130, 142–143, 162, 305n22
F11: 327n165
F16–18: 305n21
F22: 142–143, 306n29, 325n109, 327n157
F30: 325n109
F31: 306n29, 325n109
F88: 306n27, 306n29
F89: 306n27
F90: 316n105
F98–99: 317n114
F112: 306n29, 325n109
F114: 162
F120: 306n42
F124: 316n105
F156: 310n26
Pherecydes of Syros (DK)

fr. 7B4: 326n138
Philochorus (FGrH 328)

320n31
F7: 308n64
F18: 309n17
Philostratus
Im.

2.28.1: 314n78
2.28.2: 314n77
28.3–4: 315n81
Philumenus Medicus (Corpus Medicorum Graecorum 10.1.1)

Medicus de Venenatis Animalibus
15.6: 315n83
Photius
Lamia: 323n173
Mintha: 313n8
protonion: 316n95
Plynteria: 316n95
Pindar
Dith.
228
I.
4: 76, 79–80
4.35–38: 162
4.49–53: 76, 80, 303n123, 303n126
8: 320n26
8.26a–45a: 310n 28
8.59: 303n126
N.
1: 229
4.56a: 185
5.46: 80
7: 292n 48
O.
1: 60, 79–80
1.23–89: 17, 117, 162, 287n14, 303n129
1.303n128: 310n 24, 310n32, 314n68
2: 59–61
3: 60–61, 251, 309n18
3.16–33: 30n22
6.44–47: 318n150
7: 229
7.35–38: 229
7.39–48: 303n126, 310n32
7.54–76: 320n27
9.44–53: 318n141
10.28: 325n107
13.63–92: 115, 127, 259, 306n32, 327n166
13.97–100: 259
P.

2: 320n26
2.42–43: 181
2.49–51: 115
3: 320n26
3.11–12: 115–116
3.27–29: 169
4: 77, 79–80, 95, 172, 288n34, 295n23
4.25–27: 325n109
4.216–217: 328n169
4.220–242: 259, 306n29, 325n109
4.250: 259
5: 320n26
7: 320n26
8: 229
9: 17, 229
9.26–65: 142
9.30–50: 254
9.91: 296n37
10.29–30: 305n22
10.31–48: 245, 327n165
10.46–48: 259
10.48–50: 115, 259
12: 320n26
12.16: 271
13.83–84: 259
fr. 34 SM: 229, 320n27
fr. 55 SM: 327n160
Pa.
6: 60, 292n48
8: 58, 327n160
12: 59
Pisander (PEG)
fr. 5: 327n164
Plato
Cri.
113e2–6: 311n 43
Euthd.
290a: 315n87
Grg.
456d: 316n96
La.
181e–182a: 316n96
Lg.
833d–e: 316n96

887c5–e1: 9, 288n21, 320n30
Phdr.
229b4–c3: 145
Plt.
199
R.
159
2.381d1–e6: 9, 320n30
377a3–d6: 5
565d: 312n19, 317n133
Plato Comicus (Kock)

fr. 22.2: 315n80
Pliny
HN
8.81: 312n19, 317n133
11.79–82: 314n78, 315n79
11.85: 197, 315n82
24.61–63: 315n83
34.79: 318n2
Plutarch
Conv. sept. sap.
157d: 304n12
De Is. et Os.
358f: 315n80
Dem.
27: 288n30
De soll. an.
966e–f: 314n78, 315n79
Thes.
1: 163
28.2: 303n135
36.4: 160–161
Tim.
8: 317n116
Poetae Melici Graeci

965: 191, 313n58
Pompeius Trogus
24.8: 308n65, 317n116
Quintus Symrnaeus

14.347: 313n55
Sappho (Lobel-Page)
135–136: 188, 313n33
140: 306n30
206: 306n27
Semonides
fr. 7.32–36 West: 314n66
Seneca
Ep. 121.23
315n79
Servius
Ecl.
6.78: 313n45
Simonides (PMG):
frs. 509–510: 229, 320n25
fr. 543: 227, 319n18
frs. 544–548: 227, 319n18
frs. 550–551: 160, 227, 319n18
Sophocles
Aj.
492–525: 257, 325n121
Aletes
299n69
OC
139, 141, 159, 307n47
1586–1666: 117, 303n130
OT
324n100
Ph.
95, 159, 247–248, 324n90
254–316: 257, 325n118
Tr.
28–35: 257, 325n120
555–581: 250–251, 324n99
573–574: 127, 305n15
1141–1172: 250–251, 324n99
fr. 89: 318n148
fr. 264: 314n77
fr. 585: 313n38, 313n42

schol. Ant. 1350: 299n69
Stesichorus (PMG)
fr. 191: 306n28
frs. 192–193 (the Palinode): 163
frs. 210–219 (the Oresteia): 227, 319n17
fr. 220: 276
fr. 239: 227
Ger. (Campbell)
227, 319n17
frs. 10–13: 182
Strabo
8.3.14: 313n28
Suda
Lamia: 328n173
protonion: 316n95
Theocritus
3.46–48: 306n30
16.96: 314n77
schol. 2.17: 312n20
Thucydides
62
1.3: 141, 307n50
1.8.4: 144, 307n58
1.9.4: 144, 307n58
1.21: 307n58
1.22: 152, 308n6
2.29: 307n62, 313n38, 313n42
Vergil
Aen. 11
242
Ecl.
4: 318n152
G.
1.132: 295n17
4.246–247: 207, 316n108
Xenophon
Mem.

3.11.6: 315n81, 315n87
ANCIENT NEAR EASTERN TEXTS
Angim
231–232
Anzu
231
Aqhatu Legend
321n54
Atrahasis
216, 278
Tablet III: 318n140
Birth Legend of Sargon

233
Catalogue des textes hittites (CTH)

7.1: 234
7.360: 234
7.363: 234
321 (Song of Illuyanka): 232, 234, 260, 263, 326n144
322 (Telipinu and the Daughter of the Sea God): 232
324 (The Disappearance of Telipinu): 232
334 (The Disappearance of Hannahanna): 232
344 (Kumarbi Cycle): 232
345 (Song of Ullikummi): 260
366 (Inara): 232
428 (Song of Hedammu): 232, 260
457.1 (Kamrusepa): 232
789 (Song of Release): 234
Chronicle of Early Kings

Tablet A.20–23: 233
Curse of Agade
233
Descent of Ishtar (Inanna)

232

Egyptian Incantations (ed. Borghouts)
34: 69, 294n4
43: 294n8
121: 294n8
Enki and Ninhursag

232
Enki and Ninmah

232
Enmerkar and Ensuhgirana

232
Enmerkar and the Lord of Aratta

232
Enuma Elish
44, 63, 231
Tablets I–V: 260, 325n129, 326n135
Envoys of Akka
322n63
Epic of Gilgamesh
322n63
Tablet I, i–ii: 232, 239, 321n45
Tablets III–VI: 241, 323n76
Tablets X–XII: 242, 318n140, 324n84
Erra and Ishum

232, 277
Erra and Naram-Sin

233
Great Bull Is Lying Down

322n63
Inanna and Enki

232
Kirta Epic
321n54

Lugalbanda in the Wilderness
232
Lugul-e
231–232
Nergal and Ereshkigal

232
Res Gestae Sargonis

233
Return of Lugalbanda
232
Sargon Legend
Segment B: 233
Tale of King Cheops’ Court

322n61
Tale of Sinuhe
234, 322n63
Tale of the Herdsman

322n61
Tale of the Shipwrecked Sailor

234, 322n63
Tale of Two Brothers

322n63
Theogony of Dunnu
232
HEBREW BIBLE, NEW TESTAMENT, AND APOCRYPHA
Apocalypse of Paul
324n85
Apocalypse of Peter
324n85

Ev. Luc.
17:32: 310n36
Ev. Marc.
82–84
Ev. Nicodemus
324n85
Gen.
6–9: 216, 318n140
19:26: 177–179, 186, 311n1, 312n24
Hos.
5.14: 297n51
Is.
27.1: 326n134
Ps.
74: 326n134
Rev.
5.5: 297n51
OTHER TEXTS
Aitareya Brahmana
3.33–34: 311n1
Beowulf
241, 323n81
Gesta Danorum
Book 5: 323n73
Kalevala
Song 15: 242, 324n86

Mahabharata
I (Adi Parva) Part 6: 239–240, 323n74
Ovide Moralisé
179, 311n3
Prose Merlin
323n83
Puranas
323n74
Matsya Purana
1.11–34: 318n140
2.1–19: 318n140
Ramayana
241, 323n74
6.108 (Yuddha Kanda): 241
Shahnameh
318n149
1.5.1: 241
Shatapatha Brahmana
218, 318n147
Snorri Sturluson
Prose Edda: 323n73
Volsungsaga
Chap. 1: 239, 323n73
Chap. 18: 241, 323n82

The Story of Myth

Uploaded by

Copyright:

Available Formats

The Story of Myth

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Story of Myth

Uploaded by

Copyright:

Available Formats

THE STORY OF MYTH

Sarah Iles Johnston

******ebook converter DEMO Watermarks*******

All rights reserved

Cover design by Lisa Roberts

978-0-674-18507-4 (alk. paper)

The Library of Congress has cataloged the printed edition as follows:

******ebook converter DEMO Watermarks*******

Alex and Damian

For what good is a storyteller, without an audience?

******ebook converter DEMO Watermarks*******

NOTE ON TRANSLITERATIONS AND ABBREVIATIONS

1 The Story of Myth

******ebook converter DEMO Watermarks*******

I HAVE USED A Latinate method of transliterating Greek names in most cases

******ebook converter DEMO Watermarks*******

THE WORD “myth” is a notoriously slippery beast. Even if we stick to

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

The Venues of Greek Myths

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

In ancient Greece, the cultural forms that most pronouncedly belonged in

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

A Few Preparatory Remarks

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

The Chapters of This Book

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

Smith’s book drew considerable attention within the scholarly

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

******ebook converter DEMO Watermarks*******

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*

ebook converter DEMO Watermarks*