Acerca del autor

De la contraportada

De la solapa interior

Fragmento. © Reproducción autorizada. Todos los derechos reservados.

Restless Dead: Encounters between the Living and the Dead in Ancient Greece

University of California Press

No live organism can continue for long to exist sanely under conditions of absolute reality.
Shirley Jackson, The Haunting of Hill House (1959 )

So begins one of the most effective ghost stories of the twentieth century. It is an appropriate overture for a tale that explores how human beings cope not only with incursions by the restless dead but also with the uncertainty of whether what they are experiencing is really the work of ghosts or only the creation of their own imaginations. When the main character, Eleanor, is challenged by the other members of a group investigating a haunted house as to whether she has really seen a ghost, she responds, "I could say 'all three of you are in my imagination; none of this is real'." Eleanor is half joking when she says this, but Dr. Montague, the professor of anthropology who has organized the investigation, gravely replies that if he thought she were serious, he would send her home immediately, for she would be 'venturing too close to the state of mind which would welcome the perils of Hill House with a kind of sisterly embrace."

Dr. Montaguea well-trained academicwishes to keep what he considers real and what he considers imaginary firmly separated. By the end of the story, however, we have learned that for Eleanor (and for many other people as well, Shirley Jackson implies), belief in a world beyond the immediately visible one, however unpleasant that other world

The quotations from The Haunting of Hill House used here are taken from the 1984 Penguin edition, pp. 3, 140.

may be, is absolutely necessary for the expression of otherwise inexpressible fears and desires. Retaining one's sanity, as Jackson's first sentence insists, depends upon occasional vacations from reality.

Conversely, as Jackson also knew very well, a ghost story succeeds only when the narrator has managed to persuade her audience to suspend their disbelief, at least temporarily. Of course, this is one variation on a rule that applies to all fiction: the world constructed by the narrator must make enough sense to the audience for them to be able to enter into it without being constantly distracted by internal contradictions. Even if there is little expectation that a story's occurrences could take place in the real world, therefore, a properly constructed story will provide glimpses into the real world's system of beliefs because it will adhere to rules that resemble those of the real world. For example, although the vast majority of contemporary Americans who watch a vampire film do not believe that vampires really exist, they are able to suspend their disbelief long enough to enjoy watching the story unfold, both because the screenwriter has been careful to construct a fictional universe that follows its own rules and because those rules bear some similarity to rules of the real world. Thus, if a vampire is averted by a crucifix early in the story, then the crucifix must serve as a reliable means of averting vampires throughout the rest of the story, unless some good explanation that nullifies the rule is subsequently offered. Why a crucifix, and not, for instance, a piece of coral, such as some Polynesian cultures use to avert demons? Because the crucifix is a symbol of beneficent power that can be understood by any audience member who has grown up within the predominantly Christian American culture.

Even more interesting are the existential rules of many fictional worlds. As viewers of a vampire movie, we have agreed to believe that there are some corpses that return to life, but not that all corpses do. Vampires may arise from those who die under tragic or abnormal circumstances. This includes suicides, those who are unburied or who are buried improperly, and those who die cursing God. This rule makes a certain kind of sense because the early truncation of a life or the marring of a soul's passage from life to death disrupts what we like to believe is the normal progression from birth to death. People who would laugh at the idea that vampires really exist might still believe that death under such circumstances brings unhappiness to the soul or prevents its postmortem reunion with God. Witness for example the Orthodox Jewish belief that the entire body, including any severed limbs, must be buried properly if the deceased is to enjoy the eventual resurrection promised

to the faithful. Even when they cannot articulate precise reasons that proper burial is necessary, survivors usually feel compelled to provide it; the importance placed on the recovery of bodies from battlefields or accident sitessometimes at great expense and risk to those undertaking the recoveryattests to this. At least one of the rules governing vampire stories, then, indirectly reflects the values of those who listen to them. What would be impossible to accept, even within the artificially constructed confines of a vampire story, is that a pious person who died of natural causes at an advanced age, and whose funeral was conducted properly, could become a vampire.

Effective ghost stories, like effective vampire stories, reflect the values of the culture in which they developed. There are further problems to be considered before we use them as evidence for real beliefs, however, particularly when we are studying a culture like that of ancient Greece, where few people would have understood, much less accepted, Dr. Montague's assumption that a clear line can be drawn between what we call the natural and the supernatural worlds. Although a good narrator will not incorporate into a story elements that his audience will reject as "illogical" or "anachronistic," a good narrator may incorporate elements that mislead ushis distant audiencebecause they provide only part of a bigger picture. Part of our interpretive task, therefore, whenever we use narrative sources as evidence for real belief, is to recreate, as best we can, the situation in which the narrative was originally presented. When we are dealing with narrative presentations of the dead and the afterlife, with ghosts, the journey to the Underworld, its geography, and the rules by which it works, this can become complicated, for the factor that constrains narrative treatments of civic rites such as the Panathenaiarealization that the audience can compare the narrative construction to what they see and hear in real lifeis no longer fully operative. We can probably assume that no one who listened to the story of Odysseus's journey to the Underworld believed that they themselves had also traveled to Hades. Few people who watched the Erinyes pursue their victim in Aeschylus's Eumenides thought that they had ever seen one of these monstrous creatures themselves. The "reality" against which Homer's or Aeschylus's presentations of these phenomena were evaluated by an ancient audience, therefore, consisted of other things that they had heardof other constructions of a world beyond the normal sensory perceptions provided over the course of their lives by their friends, their parents, by other narrators of stories, and by the visual artists who created vase paintings, wall paintings, and temple decor.

The situation is made even more difficult by the fact that beliefs existing under no official societal sanction or control, which includes most of those concerning the afterlife, tend to be fluid, changing easily from time to time, from locale to locale, from neighbor to neighbor, and even from one statement to the next during a single conversation with a given individual. This is particularly so for beliefs about the dead because they arise in response to death itself, a phenomenon that, although inevitable and ubiquitous, is unpredictable, poorly understood, and cloaked in conflicting emotions. As the feeling of grief or guilt about another's death shifts to resignation or relief, as fear concerning one's own inevitable end shifts to hope for postmortem bliss or back again, the ways in which the afterlife and the passage into death are pictured shift as well. A contemporary American man or woman may take flowers to the grave of a loved one, perhaps in the assumption that the departed soul somehow needs or appreciates the gifts of survivors and can receive them at the location where his corpse was deposited. And yet that survivor might simultaneously believe that the departed soul dwells in a Heaven cut off from the physical world, where all needs are met and neither flowers nor anything else of a material nature has any relevance.

Even if the beliefs of an individual are fluid and sometimes contradictory, however, each of them has its place within a range of culturally acceptable beliefs. The example I just gave reflects the fact that contemporary American views of the dead admit both the idea that the soul lingers near the grave and the idea that the soul completely escapes the earthly realm. The Greeks held similarly contradictory views about the disembodied soul, imagining it now in Hades and again at the tomb. Similarly, beliefs about such things as the way the dead look can shift from one extreme to another: the Greeks tended to describe ghosts as being either sooty black or transparently pale. These descriptions reflect, on the one hand, the grim and threatening nature of many ghosts and, on the other, the washed-out, lifeless appearance of a corpse. Independently, either representation works well, even if they do not work well together.1 Christiane Sourvinou-Inwood has discussed a similar phenomenon, namely, the way that new beliefs concerning death and the

See Winkler, 159-65, on ghosts, and compare the eloquent statement made by Bot-tro at the end of his discussion of Mesopotamian ideas about death and the afterlife (p. 286): "The typical aspect of all mythological thought, in contrast to logical thought, is that it provides different answers to the same question, even opposing answers, because the answers are imaginary, exact, and calculated, toties quoties , without concern for coherence."

afterlife can enter into a culture without completely displacing the old ones. As the needs of a situation demand, now the new beliefs and now the old ones are called upon to serve.2

These methodological problems do not imply that we should ignore narrative sources when we study ancient beliefs concerning the dead as noted, narrative texts can in fact be excellent sources of information when handled sensitively. With due caution, let us now proceed on our survey, examining narrative sources grouped chronologically and by genre. At the end of each section, I shall pause to consider what general conclusions might be derived from the evidence. I shall not, however, offer detailed analyses of most of the material; that is the job of later chapters.

Homer

The Homeric poems3 are about the spectacular exploits of vigorous heroes. And yet, they leave us with no doubt that death is the inevitable end to life, except for a few extraordinary individuals such as Menelaus, who escape by virtue of their special relationship to the gods.

What came after this end? Homeric descriptions of funerary cult and mourning imply that the recently dead were able at least to hear the living and receive their offerings. The nekuia of Odyssey 11, however, suggests that in the long run, the dead were capable of very little interaction with the living. Although they looked just as they did while alive, and could be held at bay by Odysseus's sword, they were unable to converse

Sourvinou-Inwood 1995 uses this model passim, applying it to different issues as they arise, but stating it most explicitly in the methodological appendix (e.g., 416-17).

I should note that I am in general agreement with most current scholars in assuming that the Homeric poems reached more or less their present form in the mid to late eighth century, after several centuries of development, but that changes continued to be made until the late seventh or early sixth century; see, e.g., Nagy 1992, 52; 1990, 17-18 (but cf. Kirk 1-10; M. L. West 1995). For a good treatment of the relationship between vase paintings and the problem of dating the poems, see Loweustam, who also provides an extensive bibliography of earlier scholarship. For a discussion of the implications of this dating for interpreting the poems and the societal forms that they reflect, see Seaford 1994, 1-10, 14.4-54. I should also note here, however, that it is my view that the absence from the poems of phenomena that are well attested in later sources must be understood to reflect an absence of those phenomena in the societies in which the poems developed, unless other cogent explanations for their absence can be found within the thematic concerns of the poet, for example; this view governs my analysis of Homeric ideas and practices regarding the dead. I shall discuss this approach in some depth at the end of this chapter but for now will proceed on the assumption that if the poems do not mention ideas about the dead that are amply attested in later sources, this is because the ideas were not available at the time that poems underwent their main development.

with him in any meaningful way until they had drunk the blood that he provided. Teiresias does speak briefly to Odysseus before drinking the blood, in order to demand access to it, but he does not speak "knowledgeably" or "clearly" (nemertea ) until afterwards. He later tells Odysseus that the same is true for all of the soulsOdysseus can learn nothing profitable from them until they drink. The souls of Agamemnon and Odysseus's mother, Anticleia, do not even recognize him until after consuming the blood.4

It seems, therefore, that although the dead are not completely senseless in their natural stateafter all, they swarm up to the blood as soon as it is poured, like instinct-driven animalsthey exist in a sort of twilight state, incapable of any meaningful interaction with the living. They are, in a word, aphradeis , lacking all those qualities expressed by that complex notion phrade and its cognates that make converse between intelligent creatures possible: wit, reflection, and complexity of expression.5 It is only by means of the blooda striking emblem of the vigorous life they have left behind foreverthat they temporarily become capable of normal human converse. Even after they have drunk the blood, the souls of the dead remain physically insubstantial, unable to embrace, much less affect, those who are still alive, as Odysseus's futile attempt to hug his mother illustrates; his arms close upon the air. This insubstantialness is also reflected in Homeric descriptions of the dead as "flitting like shadows" and being "smokelike" or "dreamlike."6

The Homeric Underworld, then, is filled with ghosts who must be specially nourished before they can interact with even those members of the living world who arrive at their own doorstep. There is no indication that these ghosts can return to the land of the living. Indeed, Anticleia expressly claims that the opposite is true: she tells her son that terrible rivers form an uncrossable barrier between the two worlds. Odysseus has traveled to the bitter edge of the upper world in order to make his sacrifice and speak with the dead.7 It is only at this special

Sword holds dead at bay: Od . 11.48-50 (cf. the interesting twist on this scene at B. 5.68-84). Teiresias speaks clearly: 11.96. Teiresias explains the system: 11.146-49 (but cf. 10.492-95, where it is said Teiresias can speak clearly because Persephone granted him the special boon of a clear mind even after death). Odysseus's mother: 11.140-44, 152-54. Agamemnon: 11.385-90.

Aphradeis : e.g., Od . 11.476. This idea is expressed as well by Circe's description of all of the dead except for Teiresias as being without intelligence at 10.492-95.

Odysseus attempts to hug his mother: Od . 11.206-24. Shadows, smoke, and dreams: e.g., Od . 10.495; Od . 11.207; Il . 23.100-101.

Od . 11.155-59.

place, carefully designated by the goddess Circe, that any interaction between those who inhabit the upper and lower worlds is possible.8

Homer knows of some members of the dead, however, who are able to interact with the living precisely because they have not yet crossed the river that Anticleia mentions. The dead Patroclus reappears to Achilles and complains that he cannot cross the river and find peace because he has not yet received burial rites. Similarly, the ghost of Odysseus's companion Elpenor, who is among the first to arrive at the pit, and who is able to recognize and speak with Odysseus even without drinking the blood, has not yet been admitted into the Underworld because his body has not yet received funerary rites.9 The myth of Sisyphus, to which Homer alludes, and for which Alcaeus, Theognis, and Pherecydes already offer full details.10 plays with this idea, for it was by instructing his wife not to give his body funeral rites that Sisyphus ensured he would not really "die." His soul, excluded from the Underworld because his body was unburied, was given permission by the gods to return to the upper world long enough to ask for funeral rites, and once there it took advantage of the situation by repossessing his body. Sisyphus, the ultimate trickster, made what most people feared work to his own advantage.

This idea, that the dead are not admitted to the Underworld until their physical remains are ceremonially honored and disposed of in the upper world, is extremely common throughout the world. Many cultures believe that until the body is properly removed from the presence of the living, the soul of a dead person must wander restlessly betwixt and between the two worlds, no longer allowed to share in the society of the living and yet not admitted amongst the dead either.11 This belief

On Circe's significance in this role, see Marinatos.

Il . 23.65-74; Od . 11.71-78. In other passages, however, Homer describes the unburied dead as making a "squeaking" or "hissing" noise (trizo ): Il . 23.101; Od . 24.5, 9. This seems to align with an alternative belief that the dead in general, rather than being completely voiceless, made inarticulate sounds: see Soph. fr. 8.79; D.L. 8.21; Bremmer 1983, 85. It is uncertain whether the phrase "uncanny cry," thespesiei iachei , used of the dead at Od . 11.43, refers to an articulate or inarticulate sound; in Homer, it is used both of divine voices (e.g., Il . 2.600; Od . 12.158) and of the noises made by inanimate things, such as the wind blowing through the trees (Il . 16.769). In some cases, it seems simply to mean "deafening" or "overpowering" (e.g., Il . 8.159), which is probably the meaning intended here.

Od . 11.593-600; Alc. fr. 38; Thgn. 702-12; Pherecyd. 3 F 119. Cf. Gantz, 173-76.

The classical treatments are van Gennep, 146-65, and Hertz, esp. 46; see now also Metcalf and Huntington, passim, but esp. pt. 2. Van Gennep 146 notes that the transitional state, during which the individual is neither fully alive or dead, is the most ceremonially elaborated in many cultures. The dangers inherent in this transitional stage, when the individual is "betwixt and between" normal roles, have been most fatuously explored by Victor Turner in his various works, including 1967, ch. 4.

that the unburied dead are restless gives rise to another very common-idea, which we also find expressed in the Odyssey . Elpenor tells Odysseus that if his funeral rites are not carried out as soon as the men return to Circe's island, he will become "a cause for the gods' wrath" (theon menima ) upon Odysseus. Similarly, in the Iliad , the dying Hector tries to use this threat to persuade Achilles to return his body to the Trojans for burial.12 In ancient Greece, as in many other cultures, souls not yet admitted to the Underworld have the abilityand apparently the desireto compel the gods to bring harm upon the living who have done them wrong. This is an idea that continues throughout Greek history.13

Elpenor's is not the only soul that Odysseus encounters at the border of the Underworld: the souls of brides, unmarried men, virginal girls, men killed in battle who still wear their bloody armor, and "elders who have suffered many things" also wander up out of Erebus en masse as soon as Odysseus pours blood into the pit, "giving forth an uncanny cry."14 The warriors still wearing bloody armor are probably unburied, like Elpenor; no good Greek would allow the corpse of a friend to go to its grave uncleansed and without the proper shroud. It has often been noted that the brides, virgins, and unmarried men match the types of souls that later sources describe as having died "untimely"without having married and had children.15 Although it is not explicitly stated, information from later sources suggests that it is their abnormal status that keeps them from entering the Underworld. This is a topic that I take up in depth in another chapter; here I would note only that, by making these dead the first to rise up to meet Odysseus and by describing him as being afraid of them (in contrast to his fearless conversation with the other, fully dead souls), the poet implies familiarity with the ideas that

Od . 11.72-73; Il . 22.358.

Review of evidence for the Greek belief and analysis at Garland 1985, 101-3; Bremmer 1983, 89-94. For other cultures, see the treatments cited in n. 11.

Od . 11.42-43; cf. n. 9 above.

Johnston 1994; Bremmer 1983, 103; Lattimore 187; Merkelbach 1969, 189; Meuli 1975, 1: 316. Later in book 11, Odysseus meets several souls who should qualify as aoroi and biaiotbanatoi by normal standards, and who thus should be stranded at the border rather than in Hades itself, most notably Epicaste (271-80), Phaedra, Procris, and Ariadne (321-25), Eriphyle (326), Agamemnon (387-464), and Ajax (55o-67). Narrative convenience may have prompted the poet to break the "rules" he had already implicitly laid down by putting some restless dead at the entrance; this shows the ease with which conflicting eschatological models can coexist. And of course some of these individuals appear in passages that some scholars judge to be interpolations.

the abnormal dead lingered between the two worlds and that they were a source of potential trouble for the living. Another hint of this idea occurs at Odyssey 20.61-82, where the daughters of Pandareus are snatched away on the eve of their weddings to wander eternally with the Erinyes, frightful creatures of the Underworld who sometimes harm the living.16 In later sources, we shall hear a lot more about how these unhappy souls returned to the upper world of their own volition, like the unburied. We shall also hear about them being invoked by the living and forced, by means of curse tablets or other special techniques, to accomplish tasks. There is no trace of this latter idea here, however.

There is one more possible trace of the idea that the dead might affect the living to be found in the Homeric poems. We get a glimpse of what looks like hero cult at Iliad 2.547-51, where it is said that the Athenians worship their deceased king Erechtheus alongside Athena in her rich temple, offering yearly sacrifices of bulls and lambs. It is generally agreed among current scholars, on the basis of archaeological evidence, that hero cult began in the eighth century or sojust before or at approximately the same time as most scholars think that the bulk of the Homeric poems were assuming their final forms. If a hero was essentially a dead person who had retained more of his "vitality" after death, or indeed had even become more powerful than he was while alive, then hero cult represents the belief that some very special dead were capable of more than we see them doing in the nekuia of Odyssey 11.17 Granted that they are the exceptions, this passage nonetheless suggests that the notion that some dead might directly affect the living was developing at this time.

Before leaving Homer, we must pause at the issue of how souls were treated in the afterlife. Although this has little direct bearing on the question of whether the dead can interact with the living, there is some connection between the two topics, as discussed in chapter 3; thus it is important to be aware of how such ideas changed. In the nekuia of Odyssey 11, we hear about how Tantalus, Tityus, and Sisyphus suffer great punishments after death. In Odyssey 4, we learn that at least one individualMenelauswill escape death altogether and be allowed to

See further chapters 6 and 7.

For more detailed discussions of the nature of heroes, and discussions of the origin of their cult, see Seaford 1994, passim; Boedeker; Henrichs 1991, 192-93; Antonaccio; de Polignac; Kearns; I. Morris 1988; Snodgrass 1988; Whitley and Fontenrose 1968. Rohde 1925, ch. 4, and Farnell 1921 are still very good collections of information from the ancient sources; see also Nock, 574-602.

dwell forever in the Elysian Fields, an idyllic paradise. Other Homeric passages, such as Iliad 20.232-35, where Zeus's abduction of Ganymede is narrated, similarly describe individuals being carried off alive to enjoy eternal bliss in lovely places.18 The epic Aethiopis tells of Achilles' conveyance to Leuke, the marvelous "White Island," where he is to spend a very pleasurable eternity.19

Some scholars have argued that these passages prove that at the time they were composed, people already believed that a broad span of possible afterlives were available and that one's behavior or station while alive affected one's postmortem treatment.20 This is incorrect for a number of reasons, however. First, these passages concern extraordinary individuals. The crimes of the great sinners were against the gods, and like others who had offended the godsNiobe, for examplethey had to be punished to an extraordinary degree; most especially, they had to be punished for eternity and in unusual ways. It is notable that their punishments take place in the Underworld, but this may be nothing more than a way of making the punishment more odious by situating it in the most unpleasant realm imaginable. Additionally, situating the punishments in the Underworld may be a way of moving them outside of the normal world into the marginal sort of location where fantastic things occur. Niobe's punishment similarly takes place on Mt. Sipylus in Asia Minor, "somewhere among the rocks, in the lonely mountains, near the resting place of the goddess-nymphs," and Prometheus's in the distant Caucasus mountains.21 Those who won paradisical existences were extraordinary as well. Menelaus was rewarded because he was Zeus's son-in-law; Ganymede because he was Zeus's beloved.22 Neither of these groups of peoplethe sinners or the favoredare anything like ordinary people, and neither group, therefore, are meant to serve as models for what might happen to those listening to the poems. It is not until the fifth century, in Pindar, that we find clear evidence for punishment after death, and not until the fourth century, in the context of the gold tablets from southern Italy, that we find ordinary mortals claiming to become anything like a god after death.23

Sufferers: Od . 11.576-600. Menelaus: 4.561-69. Cf. Nagy, 167-71.

See Proclus's summary, lines 26-28; and cf. A. Edwards.

A. Edwards; I. Morris 1989, 309-13; Richardson 1985.

Il . 24.614-17; A. Pr . 2 (called here Scythia).

Od . 4.561-69; Il . 20.232-35.

Pi. O. 2.57-58, frs. 129, 133; h. Cer . 481-82 may hint at punishments as well, although it says only that the uninitiated will have no share in the good things the mysteries provide. Generally on this topic, see Graf 1974a, 90-126. Gold tablets A1.8, A4.4, A5.4; cf. Graf 1993; 1991a. See also the conclusions of Sourvinou-Inwood 1981 and 1995, 10-107, which are similar to mine here.

We also have to wonder whether these extraordinary people were really imagined to be dead , at least in the same sense as ordinary people would be one day. To be snatched away by the gods before life was over is not at all the same thing as dying. Proteus explicitly tells Menelaus that he will be carried away by the immortals instead of dying .24 That is part of the boon that these individuals were granted: they avoid the pain and distress that accompany the passage from life into death. The great sinners probably were not imagined as having died in any traditional sense either. Tityus was the son of Gaia, and Tantalus was the son of Zeus and a minor Titan; neither of them, in other words, were necessarily mortal in the normal sense to begin with.25 Each of these stories makes a point (do not offend the gods; the gods' favor is valuable), but none of them can be used to delineate beliefs about what would happen to real people in the afterlife. The most we can say is that the stories would have helped to pave the way for later beliefs in a system of universal postmortem rewards and punishments, although it must be noted that the punishments expected by ordinary mortals even in those later timesan eternity in muck or dung, for examplebear no resemblance to the spectacular ones suffered by the Homeric sinners.26

Achilles' afterlife requires a bit more comment before we leave this topic. Anthony Edwards has shown that the end described for Achilles in the Aethiopis miraculous translation to the paradisical island of Leukewas the standard version of what happened to Achilles after death at the time that the Odyssey was reaching its final form. What we hear about in the Odyssey Achilles' glum existence in a gloomy Hadesis the poet's innovation, a contradiction of the established

Od . 4.561-64.

Tityus's parentage varies according to source (in Pherecyd. 3FS and Simon. fr. 234, for example, he is the son of Zeus and a minor goddess named Elara), but at Od . 11.576, he is called the son of Gaia alone, and this seems to be the most persistent tradition, with variations throughout antiquity. No parentage for Tantalus is given by Homer, but later tradition usually makes him the son of Zeus. Sisyphus, in contrast, was wholly mortal according to Il . 6.153-54, which makes him the son of the founder-hero Aeolus. Of course, this is absolutely necessary if the myth of his tricking death is to work at all.

The punishment later ascribed to the Danaids, however, is echoed by that threatened for some real people in the Underworld; see Graf 1974a, 107-20. The idyllic afterlife described in some later sources, such as Pi. O. 2.61-77 and fr. 129, does sound something like the paradise promised to Menelausbalmy ocean breezes, warm weatherbut these are such common elements of the good life (see Hes. Op . 111-21, 169-73; Nagy, 167-73; Lincoln, 21-31) that it is hard to argue direct influence.

story. Edwards suggests that the Odyssean fate, which also is intimated in the Iliad , moves Achilles out of the class of heroes and into the class of mortals; this serves certain thematic purposes in the two poems. This is surely correct, but I would emphasize that for the twist to work, the glum existence in Hades must have been understood by the audience as being normal for mortals; this is the dominant paradigm from which the extraordinary, heroic Achilles escapes in the standard version of his story that we find in the Aethiopis . Indeed, if we were to assume, with Anthony Edwards, Ian Morris, and others, that the average listener had some hope of achieving an idyllic afterlife, then Achilles' fate in the Odyssey would have to be read as an exceptionally harsh end for the son of Thetis and a favorite of the all of the gods.27 Surely this could not have been the poet's intention.28

Odyssey 24

Because I agree with many scholars that the final book of the Odyssey was composed later than the rest of the two poems,29 I shall discuss it separately. It presents one significant contrast with what I have just described. Although their bodies still lie unburied, the souls of the suitors are able to meet with and talk to the souls of Achilles, Ajax, Agamemnon, and others who have received proper burial.30 This should be impossible, given the rule that the unburied could not enter Hades. It is not enough just to say that the poet ignored the rule because he wanted to introduce the great heroes of the Trojan War into his scene. In order to get away with this, he needed an audience that would not be bothered by the contradiction. In other words, we must assume that the rule that lack of burial led to exclusion from the Underworld was no longer hard and fast by the time that book 24 was composed. This does not mean that it could not still be invoked as a reason for a soul's anger and subsequent persecution of the living, as we shall see from later evidence, but rather that the causal connection between lack of burial and exclusion had slackened. This is a good example of how seemingly contradictory

A. Edwards, esp. 218-19; I. Morris 1989, 310.

See the analysis of Sourvinou-Inwood 1981 and 1983, which together reach similar conclusions, although by a different route (which I did not know when I composed this section). She summarizes these arguments in 1995, 423-29.

For a recent summary of the arguments regarding the relative date of Odyssey 24, see Sourvinou-Inwood 1995, 94-103 and 94 n. 239, with further bibliography.

Od . 24.98-204.

eschatological beliefs can coexist; the individual or society calls now on one belief, now on the other, as a situation requires.

It is also notable that book 24 is our earliest portrayal of Hermes as a psychopompos (lines 1-10). This is our first indication that the gods have any control over the movement of souls between the two worlds. Although it would be risky to conclude from the absence of Hermes as psychopompos in other parts of the poems that the role developed only after their composition, the absence is nonetheless striking. Other Homeric descriptions of passages to the Underworld portray souls as simply flying away from their bodies, suggesting that in the view of this poet, transition to death was swift and simple, requiring no divine aid.31 As we shall see, the need for psychopompoi not only persisted but apparently grew as time went on: by the time of the epic Minyas , Charon had joined Hermes in this role.32

The geography of the passage to the Underworld is given in some detail in book 24. The poet mentions dank pathways, the Oceans stream, the White Rock, the Gates of Helios, the Country of Dreams, and a Meadow of Asphodel where the souls congregate.33 Sourvinou-Inwood has hypothesized that this geographical detail, as well as the introduction of Hermes as a psychompompic god, reflects a growing concern at the time of book 24's composition with the physical boundaries between life and death.34 The passage from book 24, however, is not the only place in which Underworld geography is given in detail. The Oceans stream and the Meadow of Asphodel are mentioned in books 10 and 11 during descriptions of the Underworld, as is a rock that lies at the entrance to the Underworld (although it is not specifically called a white rock).35 When Circe tells Odysseus how to get to the border between the upper and lower worlds, she also mentions woods, groves of Persephone, tall black poplars, sterile willows, and four separate, named rivers of the Underworld. We hear about the Cimmerians, a race of people who live close to the border, in eternal darkness, too.36 Such geographic and ethnographic details would be at home in tales of heroic descents to the Underworld, which predate the Odyssey as we know it and which do

E.g., Il . 16.856, 22.362; Od . 11.222. Cf. Sourvinou-Inwood 1995, 56-59.

Minyas fr. 1 Davies (= Paus. 10.28.1). Cf. Sourvinou-Inwood 1995, 94-106, 303-61.

Od . 11.13-22.

Sourvinou-Inwood 1995, 103-7.

Stream of Ocean: Od . 10.508, 511; 11.21, 639; 12.1, 20. Meadow of Asphodel: 11.538-39. Rock: 10.515.

Cimmerians: Od . 11.13-19.

not, I would add, necessarily have any connection with beliefs concerning the average person's travels back and forth.37 The interest is analogous to that shown in the details of the Phaeacians' island or Aeolus's palace: descriptions of exotic, distant lands are always fascinating. The interest in Underworld geography persists even as other eschatological beliefs change, as we shall see. Alone, therefore, it is not a good barometer by which to measure those changes.

Other Material From the Epic Cycle; Hesiod; the Hymns

Fragments of the epic cycle38 have little to say about our topic. The only substantial mention of interaction between the dead and the living comes from the Nostoi , where the ghost (eidolon ) of Achilles appears to the Greeks leaving Troy and tries to prevent their departure by foretelling the doom that awaits them.39 This might be taken to indicate that even the ghosts of the properly buried could return to the upper world, although as a hero, Achilles could also be understood as an exception to the rules that govern the ordinary deadwe have already seen that different epic traditions had different ideas about the fate of Achilles' soul. The passage is also our first indication that the dead might give advice to the living, a role they continue to play throughout antiquity either of their own volition or, later, when asked to do so through rituals.40

Hesiod,41 in lines 121-23 and 126 of his Works and Days , tells about how the privileged dead of the Golden Race return to earth to protect the living and bestow wealth upon them. Some scholars have interpreted

Heracles' descent is referred to already at Od . 11.622-26.

I should note that I find it difficult to accept any attempts to date other poems in the epic cycle relative to the Odyssey and Iliad there seems no secure means of resolving this issue. For my purposes, however, relative dating of the poems is unimportant, because the other cyclical poems do not offer information that contradicts that of the Iliad and Odyssey indeed, they scarcely offer information about the dead at all.

Proclus's summary, lines 15-17.

I do not consider Teiresias's prophecies to Odysseus to be an earlier instance of the dead prophesying, as Teiresias was able to foretell the future even before he died. It may seem that in the Iliad , the dying warrior twice is able to predict the death of the one killing him: Patroclus tells Hector he will be killed by Achilles at 16.852-54, and Hector tells Achilles that he will be killed by Paris and Apollo at 22.358-60. The former instance, however, seems like nothing more than a threati.e., Patroclus is sure that his friend will avenge his death and wishes to frighten Hector with that certaintyand the latter reflects a well-known and apparently long-standing prediction about Achilles' death, which we already have heard about at 21.275-78 (and cf. 18.96, 19.416-17).

I agree with most current scholars (as expressed. e.g., in M. L. West's "Hesiod" in the OCD , 3d ed.) in placing Hesiod's floruit around 700 B.C.E ., i.e., after the Homeric poems had undergone most of their development (see n. 3 above).

a later passage (252-55) as indicating that these souls of the Golden Race also play a role in punishing the misbehavior of the living. It describes the 30,000 deathless guardians of mortals who "keep a watch over lawsuits and wicked acts, wandering over all the earth, clothed in mist." The latter two lines of this passage are also inserted by some manuscripts after line 123, in the middle of Hesiod's description of the Golden Race, which would serve to equate the souls of the Golden Race with the 30,000 deathless guardians.42 Martin West objects to including these lines in the earlier passage because in his view it is inappropriate for those who bestow gifts on mortals to serve as a "secret police" as well, but the objection does not hold water: the preservation of justice is just as great a boon as wealth or any other benefit that the Golden Race souls might bring.43 Indeed, the importance of justice for preserving any other good that might befall mortals is one of the pervasive themes of the Works and Days .

But however that issue might be settled, we must take note of the fact that the souls of the Golden Race are an extraordinary type of dead, elevated by Zeus to something very near the status of gods. Like the Phaeacians of the Odyssey , they are described as "dear to the gods" and as daimones , a term that in Hesiod's day still served primarily as a synonym for theoi . Neither the honored dead of the Silver Racesecond only to the Golden in perfectionnor those of the Heroic Race are said to interact with the living. The former are blessed after death but dwell under the ground; the latter dwell in bliss at the ends of the Earth, as Proteus says that Menelaus will, and as Achilles is said to do in the Aethiopis .44 Nothing at all is said about what happens to souls of our own age. Hesiod, then, makes no reference to the possibility that the ordinary dead might return to interact with the living. This silence is particularly striking given that the poem ends with a list of warnings about unlucky acts and dangerous situations that the listener must avoid. Surely, if fear of the returning dead were rampant, we would find it reflected here.45

Discussion of the problem at M. L. West 1978, 183, and cf. 219-20. Plutarch accepts the equation of the souls of the Golden Race with the Watchers (De def. or . 431b-e); Proclus does not (p. 87, 15 Pertusi).

Cf. the remarks of the heroic dead in Ar. fr. 322, who promise to punish thieves and robbers by sending various illnesses against them (quoted on pp. 153-54 below).

Menelaus: Od . 4.561-69; Achilles: Proclus's summary of the Aethiopis , lines 26-28.

The closest Hesiod comes is to advise against conceiving children after a funeral (Op . 735) and against allowing boys to sit on "unmoveable objects," which may refer to tombstones (akinetoisi , 750, see M. L. West 1978 ad loc.). The first may reflect any number of ideas, including the general impurity that participants in a funeral incur or the idea that a fetus is affected by its mother's state of mind. The latter similarly may reflect the pollution generally associated with death.

One other work composed at about this time must be considered: the Homeric Hymn to Demeter .46 It makes no mention of the dead returning to interact with the living, but it does demonstrate that the boundary between the upper world and the Underworld was permeable; if Persephone could pass back and forth, perhaps others could as well. Hecate's role as her propolos and opaon , her companion and guide, during this journey is especially interesting in this respect because it seems to reflect Hecate's role as the mistress of the dead and the goddess who could lead them forth into the upper world or restrain them in the Underworld as she wished. As we shall see in chapter 6, Hecate's association with Persephone, a paradigmatic virgin, seems also to articulate the reason that Hecate took on the role of mistress of ghosts in the first place.47 Thus in this, one of Hecate's earliest appearances in Greek literature, we find her already fulfilling the duties for which she was most highly valued in later times.

The Hymn is also important because it introduces the idea that all individuals will be punished or rewarded after death for their behavior during life. We also find this idea in Pindar, for example, and it becomes quite common during the classical period. According to some ancient texts, including the myth that concluded Plato's Republic , the choice between reward and punishment depended on the individual's conduct during life; other texts, including the Hymn and a fragment of Pindar, promised that by undertaking special rites while alive, anyone might win postmortem rewardsperhaps even an afterlife that included sunlight, feasting, and beautiful surroundings, similar to the paradisical existence promised to heroes in earlier works.48 Thus were introduced not only the possibility of a better afterlife but the necessity of worrying about one's afterlife while still alive and of wondering about the condi-

For date, see Richardson 1974, 5-11, who suggests that we can date its composition no more precisely than to say between 675 and 550thus, surely no earlier than Hesiod and probably later. (Some of Richardson's arguments, depending as they do on the assumption that the poet was Attic or at least intimately familiar with Attica, have been called into question by Clinton 1986 and 1992, 28-37, but this will not affect Richard-sons main lines of argumentation.)

Cf. the analysis of Clay, 202-62. Hecate as propolos and opaon : lines 438-40.

Pl. R. 614b2-621d2; h. Cer . 480-82; Pi. fr. 137. Cf. Graf 1974a, 98-103; Lloyd-Jones 1984.

tion of other people who had died. Death and the dead became objects of greater concern precisely because variation had been introduced.

One more important idea introduced during the later archaic age was that of metempsychosis. Our earliest extant references to it are from the first half of the fifth century, in Pindar and Empedocles, but ancient sources insist on crediting it to Pherecydes and Pythagoras, who lived about a century earlier, as well.49 Pindar's mention of it is particularly important because his poetry circulated widely throughout Greece during the fifth century. By the turn of that century, the idea of metempsychosis was familiar enough to well-educated Athenians to be used in Plato's dialogues without further explanation. Judging from comments made by Plato and Aristotle, metempsychosis was taught in association with the Orphic mysteries, which might imply that an even wider segment of the population knew the concept (although we cannot be sure that all rituals or beliefs called "Orphic" by ancient authors were necessarily part of all "Orphic" initiations).50 Metempsychosis, like belief in a system of postmortem rewards and punishments, assumes an expectation that souls will be treated as individuals after death, and it therefore also indicates, again, that we have moved quite a bit away from the Homeric picture of an afterlife in which all are treated equally. Of similar import is the belief that the souls of the living can temporarily separate themselves from their bodies to wander abroad for periods of time, which also shows up first in the late archaic age.51

One more important idea can be found in Empedocles' poetry. In fragment 101, he boasts that he will teach his students to lead souls back out of Hades. This claim is confirmed by comments made by his pupil Gorgias and by the remarks of later authors such as Diogenes Laertius.52 This art, properly called either psycbagogia or goeteia , is analyzed in depth in chapter 3, but I should note here that Empedocles' poem is one

Pi O. 2.68-77 and fr. 133; Emp. frs. 107, 108 Wr. = 31 D 117, 115 D-K, and see discussion at Wright 63-76, 270-76. Pherecydes: evidence is collected and analyzed in M. L. West 1971, 25. Pythagoras: evidence is collected and discussed at Burkert 1972, 120-65.

Pl. Lg . 870d4-e2; cf. Cra . 400c and Arist. de An . 410b29 = Orph. fr . 27. Discussion at Burkett 1985, 297-301.

Bremmer 1983, 25-38; Meuli 1975, 2: 817-79; Burkett 1967, 147-49; Bolton; Dodds 1951, 135-78.

Fr. 101 Wr. (= 31 B 111 D-K); line 9 = . Cf. the source of the fragment, D.L. 8.59 (quoting Gorgias) and discussion in chapter 3 below.

of the earliest mentions we have of the very important idea that the dead not only were capable of returning on their own but could be made to return by actions performed by the living.

Lyric Poetry, Epinician Poetry

It is not surprising that we derive little information about the status and activity of the dead from this body of literature, for most of it is concerned with the immediacies of lifelove, war, athletic glory, and the pleasures of the symposium. When death is mentioned, it usually is by way of contrast. Thus, for example, Stesichorus, fragment 244:

... for it is futile and pointless to weep for the dead.

and Sappho, fragment 55:

[After death you will be] forgotten,
and there will never be any longing for you, because you have no share
     of the roses
of Pieria. Unseen in the house of Hades,
flown from our midst, you will wander amongst the shadowy dead
     [phoitases ped' amauron nekuon ].

The poets choose to emphasize the permanence and irreversibility of death, rather than any possibility of contact between the living and the dead. Death is an emptiness, an absence of life.53

Nonetheless, the poets occasionally reveal familiarity with the possibility of something else. Simonides' ninth epigram seeks to emphasize the idealistic notion (very useful to the emerging city-state) that one's military valor (arete ) survives death:

These men adorned their dear country in imperishable fame
when they cloaked themselves in the dark cloud of death.
They died but they are not dead: their valor bestows glory upon them here
     above
and leads them up from the house of Hades.

The verb anago , found here in the phrase "leads up from the House of Hades" (anagei domatos ex Aideo ), is frequently used of the invocation of souls in the classical period. Although Simonides uses the idea of invoking the dead only symbolically, to give expression to the common-

Cf. also Sol. fr. 21; Simon., epigrams 84 and 85 = AP 7.516 and 7.77.

place idea that glory survives death, the phrase suggests that his audience was familiar with the sort of ritualized psychagogia that we see acted out a few years later in Aeschylus's Persians , for example. If so, this would, with Empedocles' poem, be one of our earliest literary attestations of the idea that the dead could be made to rise up from the Underworld (the epigram is thought to refer to the Spartans who fell at Plataea, and thus can be dated to shortly after 479).

In Pythian 4.159-64, Pindar tells of how Pelias persuaded Jason to travel to Colchis by describing a dream in which Phrixus, their shared ancestor, said that his wrath could be assuaged if his soul (psyche ) were brought home to Thessaly. The way in which this could be done, according to the ghost, was to retrieve the Golden Fleece.54 In many ways, this scene echoes the much earlier encounter between Achilles and Patroclus's ghost in Iliad 23, for in both cases, the dead appear to the living in order to ask them to perform rituals that will allow them to rest more easily in death. And yet there is a significant difference. Patroclus wants only the rites that will admit him to Hades; behind Phrixus's plea lies the assumption that a soul can be transferred by means of ritual from one place to another here in the upper world. This is the very essence of the sort of ritual that is described in stories, discussed in chapter 3, in which expert practitioners lead ghosts away from one spot to another. Although detailed accounts come only from later sources such as Plutarch, an allusion to such a process is made by Thucydides, and Euripides mentions the psychagogos , the professional "leader of souls,"55 which suggests that techniques for transferring ghosts from one place to another were available at Pindar's time. Phrixus's plea also confirms that the soul of a dead person, however much it may be imagined to dwell in Hades, has some attachment to a physical place in the upper world as well. A few decades later, Herodotus's story of Periander and Melissa's ghost rests on the same assumption, as do a number of scenes from tragedy.

Pindar's story adds one more detail that provides an important insight: Pelias claims that the Oracle at Delphi had confirmed and approved of his dream and its message. This not only reminds us that appearances of the dead in dreams were not automatically assumed to

Cf. pp. 154-55 below.

Th. 1.134.4-35.1; E. Alc . 1127-28. Cf. pp. 61-62 below on the ritual to which Thucydides alludes.

be valid (a double check might be necessary) but also indicates that the Oracle, one of Greece's most esteemed religious institutions, both upheld the possibility that the dead might convey information in dreams and advocated the use of rituals through which the living could interact with the dead. The ritual to which Thucydides referred similarly was commanded by Delphi, and we shall hear more about this Oracle, as well as the Oracle of Zeus at Dodona, giving instructions for such rituals later in chapters 2 and 3. This indicates that interaction with the dead and the experts who specialized in it were not anathema to mainstream Greek culture and religion. Indeed, Pindar's story suggests that even an old Greek hero like Jason could be imagined as engaging in psychagogic activities. Another old Greek hero, Odysseus, also interacted with the dead, of course. The Merlinesque figure of the magiciandark and mysterious, dwelling in the woods at the fringe of civilization does not work well here. However frightening and potentially dangerous the dead themselves might be, the individual who could control and use them was welcome and sometimes even esteemed.

We find intimations of a darker form of postmortem activity in Sappho. Fragment 178 mentions Gello, who is "fond of children." This refers to a demonic creature whom we know well from later sources, the soul of a dead virgin who wandered around in the world of the living, enviously killing babies and pregnant women. Already in Homer, there are allusions to the belief that the unmarried dead were excluded from the Underworld and might harm the living, but it is in this fragment of Sappho that we first find reference to one by name, or any intimation of the specific form that this harm might take.

From lyric, elegiac, and epinician poetry, then, we have been able to glean several useful pieces of information. We find our earliest hints that the dead might be called out of Hades by the living (Simonides, cf. Empedocles) and the idea that the living could affect the circumstances of the dead through rituals (Pindar). We also get another glimpse of an unhappy soul returning to attack the living (Sappho). Finally, in Pindar, as in Empedocles and the Homeric Hymn to Demeter , we meet the idea that each individual soul survives and experiences an afterlife that is exclusively tailored to it, whether it take the form of punishment, reward, or metempsychosis.56

In the following section, I occasionally refer to evidence from Herodotus, comic drama, and Plato's dialogues in the text or notes. None of these sources has enough to say about popular beliefs in the dead to merit a sections of its own, but all of them offer informarion that confirms the picture found in tragedy, with a few interesting additions.

Tragedy and Contempoary Literature

Heraclitus said that Dionysus and Hades were really one and the same.57 Students of tragedy should find this easy to believe, for nowhere do we meet the dead more often than in the plays of Aeschylus, Sophocles, and Euripides; the boundary that separates them from the living seems less secure than ever before. Quite a few of the dead transgress it, moving out of Hades either in flesh or in spirit;58 two tragedies are introduced by a ghost; in one of them, a ghost speaks alone on stage for 58 Frees before anyone living appears. In at least two tragedies, the dead rise in response to rituals performed by the living, which confirms that psychagogic techniques such as those alluded to by Empedocles, Simonides, and Pindar were becoming well known by this tune. In the Alcestis , Heracles refers to professionals who made a living from such rituals: one of the names they use is. "psychagogos ," literally, a "leader of the soul."59 In still other cases, where the dead themselves cannot rise in flesh or spirit, they send forth their agents, the Erinyes, to do their bidding.60

A number of other tragic characters, falsely believed to be dead, miraculously "return" from Hades to make their presence felt among the living in various ways.61 Pretended death is central to the plots of several plays, either as a treacherous overture to real death (it is the "dead" Orestes who kills his own mother in Aeschylus's and Sophocles' versions of the story) or as a stratagem to escape from death (Helen rescues Menelaus by pretending to perform his funeral rites in Euripides' Helen ). Even when not central to the plot, scenes in which survivors mistakenly mourn those who are still living make good dramatic fodder. Particularly interesting, because particularly perverse, are those in which a purportedly deceased person, like Tom Sawyer, watches his survivors

Fr. 22 B 15 D-K.

Clytemnestra in the Eumenides , Alcestis in the Alcestis , Heracles and Theseus in the Heracles . The ghost of Polydorus speaks the prologue of Euripides' Hecuba , and the ghost of Achilles apparently spoke the prologue of Sophocles' Polyxena (fr. 523). As North, 51-52, points out, the development of a device known as "Charon's steps" in the middle of the orchestra of many Greek theaters suggests how frequently ghosts must have appeared in plays.

In the Persians , Darius rises in response to rituals performed by the Persian elders (623-842). Aeschylus's lost Psychagogoi dealt with Odysseus's invocation of ghosts and thus must have presented them on stage. E. Alc . 1127-28.

E.g., A. Eu . 94; S. El . 112. See further chapters 4 and 7 below.

Heracles in the Heracles , Ion in the Ion , Iphigenia in the Iphigenia in Tauris . Cf. Heracles' statement at Tr . 1160-63 that Nessus has returned from the dead to kill him.

mourn.62 Death seems rather weak in tragedy, no longer able to hold on to those it has snatched. It is punctuated by a question mark rather than a period.

Nor do the living stay in their proper place. Great heroes such as Heracles, Theseus, and Orpheus are said to have actually visited the land of the dead,63 and many other tragic characters move into it symbolically. For example, the condemned Antigone is described as a corpse lingering among the living. The Erinyes promise to make the living Orestes a "bloodless shadow" (anaimatos skia ), and Philoctetes says that he is already dead from his illness, nothing more than a vaporous shadow or ghost (kapnou skia, eidolon ). Electra cries out that in murdering Clytemnestra, her children have become like the dead themselves (isonekues ) When Heracles returns alive and victorious from Hades, he finds his family macabrely dressed in their funeral garments, awaiting death and praying that his ghost might aid them. Of course, it will be Heracles himself, this amazingly resurrected ghost, who kills them in the end, proving their costumes all too appropriate.64

The living and the dead, then, constantly exchange places in fifth-century drama. Their worlds are dreadfully close; they move between them with a disarming ease. Unspoken questions about the nature of life and deathand challenges to their assumed dichotomylinger over the tragic stage, seeming to reflect the arguments made by Empedocles and others of the time, who said that they were but two sides to the same coin: as we die we enter into a new sort of life that eventually brings us to birth again; as we are born, we begin a sort of death. We see more fully developed forms of what we only glimpsed before: a belief that death is not the end of everything at all, but only the beginning of a new sort of existence, for better or for worse.

We might be tempted to dismiss tragedy's fascination with death and the dead as something that came naturally to the genre. Tragedy loves both paradoxes and sudden reversals of plot, which are readily provided by situations in which, for example, those still living join the world of the dead (Antigone walking into her own tomb), or the dead turn out to be living (Orestes "rising" from Hades to rescue Iphigenia). Tragedy,

E. IT 144-77; S. El . 1098-1229. Cf. Alc . 1037-1126, where the "dead" Alcestis listens to Admetus extol her virtues to Heracles.

E.g., E. HF 516-18, 610-21; Heracl . 218-19; Alc . 357-62. Cf. Pl. Phd . 68a2-7 and Ar. Ra . throughout the first half of the play.

S. Ant . 850-52; cf. 559-60, 810-12, 1070-71; A. Eu . 302; S. Ph . 946-47; E. Or . 194-207; E. HF 442-43.

moreover, is by its very nature a genre that challenges the boundary between reality and illusion, both in its performative aspects and by bringing mythic figures onto the stage. The ghostthe eidolon , the skia , the phasma , the thing that is here in front of our eyes and yet not really hereemblematizes quite nicely the slippage between reality and illusion that tragedy loved.

And yet, these observations alone cannot account for tragedy's fascination with the dead and their world, for there were other subjects that could lend themselves to paradox and reversal and other subjects through which the issue of illusion and reality could be explored. Tragedy's subject matter, however nicely it may generate plot twists and provocative paradoxes, always grapples with issues of contemporary concern; to borrow the words of one of the genre's foremost scholars, tragedy is "the epistomological genre par excellence , which continually calls into question what we know and how we think we know it."65 We have to assume, therefore, that tragedy's constant references to the dead, their world, and the permeability of the boundary that separates it from ours reflect a genuine interest in and uncertainty about these matters in the fifth century. Of course, in some of the cases I have cited in the past few paragraphs, these topics are relatively peripheral to the main subject matter and action of the plays, but this actually helps to make my point: if even in those plays where they do not occupy a central role, death and the dead persistently manifest themselves, then these were indeed topics that longed for expression and resolution in the fifth century.

Richard Seaford's thesis, which finds allusions to eschatologically oriented mysteries of Dionysus in the Bacchae and suggests that tragedy's precursors, dithyramb and satyr-play, offered veiled dramatizations of Dionysiac initiations, is also important here, for it provides a reason that tragedy, of all the genres available in the fifth century, became the popular medium through which the topics of death and the dead were explored.66 Seaford believes he must concede, however, that the Bacchae is unique among tragedies in retaining its precursors' interests, for in most other tragic plots we find little or no reference to eschatological mysteries, direct or symbolic. But I would suggest that it is precisely in tragedy's obsession with the dead, and particularly in its constant consideration of the possibility that the living and the dead might trade places, that we find tragedy carrying on this tradition. As the genre had continued to

Zeitlin 1990, 78.

Seaford 1981, and cf. also Seaford 1976 and 1977-78.

develop and become more important to civic identity, its subject matter naturally had expanded to address issues of contemporary concern that had "nothing to do with Dionysus" in his role as a god of the mysteries. The genre's traditional link to eschatological mysteries therefore came to be expressed in subtler, more allusive ways that enabled tragedy to expand its thematic horizons; tragedy chose to provoke thought about these matters rather than to encrypt the patterns of one single answer, the answer provided by Dionysiac mysteries.

Comedy's most extended commentary on tragedyAristophanes' Frogs confirms this conclusion. Not only is the second half of the play set in Hades' palaceit is here that both tragedy and all of Athens either will be saved or be lost, like a mystery initiate facing his postmortem questioningbut, more important, the first half of the play comprises a journey into the Underworld. Our hapless guide on this journey is Dionysus himself, the very god of tragedy but also the very god who promised that those who were initiated into his mysteries would receive perfect knowledge of how their souls might travel safely to the best parts of the Underworld. Dionysus's complete ignorance regarding these matters in the Frogs surely is a spoof of his real role as a god of eschatological mysteries, for his ignorance specifically takes the form of needing Heracles to tell him about the best paths to travel by and the best places to visit in the Underworld. Apparently, these were precisely the things that the Dionysiac mysteries promised to teach the initiates, for the gold tablets that were buried with them reminded the initiate of the particular Underworld roads that they should choose, the landmarks that they should look for, and the phrases that they should pronounce to officials whom they encountered.67 Setting the second half of his play in the Underworld gave Aristophanes an interesting way to make his main point. Tragedy was dead; if Dionysus wanted it back he would quite literally have to revive it by reviving Aeschylus, leading him back to the upper world like one of the psychagogoi who were familiar in Athens by then. But by including Dionysus's trip into the Underworld, Aristophanes not only was making fun of the god's fame as a draftsman of infernal road maps but perhaps also parodying what he viewed as tragedy's fascination with the interaction between the two worlds. By

On the gold tablets, see Graf 1991a and 1993. There are also allusions to the Eleusinian mysteries in the first half of the Frogs , as many scholars have noted, but comic poets are under no obligation to be consistent, and when representing the mysteries on stage it was probably best to be imprecise anyway.

the time he reaches Hades, Dionysus has poked his nose into all manner of mortuary, eschatological, and psychagogic matters.

The commingling of the upper world and Underworld that tragedy evinces, needless to say, would have been somewhat disquieting for the average person. For one thing, there would have been the feeling that we, the living, were being watched by numerous shadesnot just Hesiod's beneficent daimones of the Golden Race, but potentially the spirit of every individual who had died. The many cracks in Hades' crumbling walls meant that more attention had to be paid to keeping these spirits happy; tragedy had much to say about this as well. The dead demanded libations, tears, dedications of hair and clothing, and even human sacrifices upon occasion. They liked to be greeted by the living who passed by.68 Of course, their first heed was to be buried, a need around which the plots of Sophocles' Antigone and Ajax and Euripides' Suppliant Women revolve. Those who mistreated the dead, either while they were living or after death, would suffer dreadfully: illness, madness, and other, unspecified evils befell them, sometimes brought directly by haunting ghosts. "Those beneath the earth blame and are angered against their murderers," warn the Chorus of the Choephoroi .69 Plato, tragedy's younger contemporary, also discussed the fact that most people feared the wrath of the dead and the possibility that they might return to wreak vengeance. Such discussions arose in the context of other issues; indeed, popular belief in the return of the dead was used to clarify or illustrate other points, which implies that it was well understood and accepted. In one famous passage, for example, Socrates says:

The soul that is tainted by [the corporeal] is weighed down and dragged back into the visible world, through fear, as they say, of Hades or the invisible, and it hovers about tombs and graveyards. The shadowy apparitions that have actually been seen there are the ghosts of those souls that have not got clear away, but still retain some portion of the visible, which is why they can be seen.... Of course, these are not the souls of the good, but of the wicked, and they are compelled to wander about these places as a punishment for their bad wantonness in the past.70

E.g., A. Pers . 219-23; Ch . 4-9, 13-15; S. El . 448-52; E. IT 159-77; El . 90-93, 323-25, 508-15; Hec . 108-43, 530-41; Hel . 1165-66; cf. Tr . 1246-50. Cf. Hdt. 5.92h , the story of Periander's consultation of his dead wife Melissa at the nekuomanteion of Ephyra (quoted on p. vii above). Melissa demands that she be given clothing to keep herself warm in Hades before she answers the question; the dead exist in a state not too different from that of the living, in which they may have needs and desires.

A. Ch . 39-41. To mention just a few other examples: A. Pr . 568; A. 1186-93, 269-96; S. OC 621-22; El . 1418-19; E. Or . 34-45, 253-77; IT 282-308.

Pl. Phd . 81c10-d9 (H. Tredennick's translation, adapted).

Of similar interest is a passage from the Laws :

But let [the good man] take heed not to despise what the old and venerable myth teaches us. It tells us how he who is done to death with violence ... has his wrath kindled against the author of the deed in the days while it is still fresh, how he is filled with fear and horror at his bloody fate, how he is aghast to see his murderer traveling streets that were once familiar to him, and how in its own turmoil [the disembodied soul] joins forces with the very memory of the murderer to bring all possible distraction upon him and all his works.71

Xenophon's Cyrus makes a similar point:

Have you never yet noticed what terror the souls of those who have been foully dealt with strike into the hearts of those who have shed their blood, and what avenging powers they send upon the track of the wicked?72

The fact that at least three comedies about ghosts were written during the late classical period also confirms that a belief in ghosts was widespreadfor how can one make comic something that is not taken at least a little seriously?73

We also hear more about some specialized forms of ghosts in comedy. Midway through Dionysus's and Xanthias's descent to Hades in the Frogs , Empousa appears to the frightened travelers. She is a horrible, hybrid, demonic ghost who tries to frighten and impede them on their way to the Underworld.74 Elsewhere in Aristophanes, we hear about Lamia and Mormo, two other female ghosts who, like Gello, specialized in attacking children and pregnant women. Mormo also is referred to by Plato and Xenophon.75 As noted in later chapters, all of these creatures were said to be unhappy souls who vented their frustration by causing trouble for the living. Well-educated, upper-class men such as Plato may have laughed, but the existence of spells and amulets against such threatening creatures guarantees that not everyone found them to be so funny.

The dead might be frightening and vengeful, but they were also expected to provide help to the living who treated them well, or to those with whom they had a link based on affection, particularly a familial

Pl. Lg . 865d5-e6 (trans. A. E. Taylor); cf. Hp. Ma . 282a7-8.

X. Cyr . 8.7.18-19 (trans. Walter Miller, adapted).

Comedies: in addition to Menander's Phasma , we have fragments of a Phasma by Theognetus (PCG VII 697) and a Phasma by Philemon (PCG VII 272-73).

Ar. Ra . 286-305. Further on Empousa, see chapter 4 below.

Ar. Eq . 693; V . 1035 (= Pax 758), 1177; also schol. on Ar. V . 1035; schol. Ar. Ach . 582; schol. Ar. Pax 474 and 748 and schol. Ar. Ec . 77. Pl. Phd . 77e; X. HG . 4.4.17. See further chapter 5 below.

link. As Plato observes: "People should be properly fearful of ... the departed souls, whose natural inclination is to watch over their own children, to be kind to those who are kind to their children but cruel to those who are cruel to them."76 Often, requests for help are made in the course of offering libations and mourning the dead; the Choephoroi provides the most famous instance of this, where Orestes and Electra beseech Agamemnon for aid at his grave.77 The Persians goes even further, presenting us with a scene in which a dead man is literally called to the upper world in order to give advice to his family.78 Herodotus gives evidence that by the time he wrote, and probably at least a century earlier, various nekuomanteia (oracles of the dead) were doing booming business, promising the living face-to-face encounters with the dead, during which questions might be asked.79 By the fifth century, we are also hearing a lot about hero cult, in which the dead were expected to help individuals or entire cities.80 A darker view is presented by some remarks in Plato, which allude to the practice of ritually manipulating the dead in order to injure other living individuals. The same specialists who did this, apparently, also offered to get rid of problems caused by the dead.81 All of this is quite different from the Homeric picture, in which, as we saw, the world of the dead is by and large cut off from that of the living.

If tragedy and contemporary sources portrayed a reciprocity between the living and the dead, in which each depended upon the other, they also showed a heightened interest in the way that the world of the dead worked in itself. We hear far more than we did before about punish-

Pl. Lg . 927b1-4. General requests for aid: S. El . 1066-81; E. HF 490-95; Hel . 961-68; El . 678-81; Or . 1225-45.

Asking for aid while giving gifts: A. Pets . 219-25; Ch . 123-51, 479-509; cf. Eu . 598; E. Or . 112-25.

Jouan usefully discusses this scene and summarizes previous scholarship. I discuss it myself in some depth in chapter 3.

Hdt. 5.92h (Periander consults the Ephyran oracle of the dead); 1.46.2-3, 49 (Croesus consults the oracles of Trophonius and Amphiaraus); 8.134.1 (Mys consults the oracles of Trophonius and Amphiaraus). If we believe these tales, the nekuomanteion at Ephyra was well known at least by the late seventh century, and the nekuomanteia of Trophonius and Amphiaraus were known as far away as Lydia by the mid sixth century. Cf. also S. fr. 478 and Ar. Av . 1553-64 on the general existence of nekuomanteia in the fifth century. For a detailed description of the way a nekuomanteion might work, see Plu. Cons. Apollo . 109b-d.

Hero cult: A. Eu . 762-74; S. OC 359-90, 457-60; E. Alc . 996-1005; Heracl . 1026-44; Ar. fr. 322 (on which see pp. 153-54 below); cf. Hdt. 6.69.1-4, for example, where the Spartan hero Astrabacus leaves his shrine one night to sire the future king Demaratus on a local woman. Further briefly on hero cult, see pp. 153-55 below and the authorities cited in n. 17 above.

R . 364b5-365a3; Lg . 933a2-b5; cf. 909a8-b6; these passages, and their connection to the curse tablets, are discussed in depth in chapter 3.

ments and rewards, although there is no certainty as to whether it is virtue or merely initiation into mysteries that guarantees a happy afterlife. Both the rewards and the punishmentsbut especially the punishmentsare described with relish.82 We also hear about how the dead treat one another: according to some remarks, they were both socially stratified and ethically judgmental of one another.83 And we hear more about how one got into the Underworld in the first place; psychopompic figures, including Charon, Hermes, and Hades himself, are mentioned several times, as is Cerberus.84 Underworld geography is described in detail.85 None of these features are completely new, but there is a greater fascination with them in tragedy than in previous literature.

Conclusions and Some Methodological Issues

Some conclusions can be reached, but to support them we must revisit the Homeric poems in order to ask some methodological questions that could not be raised until all of our materials had been laid forth. I shall apply Ockham's razor, seeking explanations that require the fewest or least complex suppositions. First, the proposed conclusions:

Like many peoples, the Greeks believed from a very early time that some special dead, such as the unburied and those who had died young, were angry with the living and had some means of harming them. In our earliest narrative sources, the dead do so only through the agency of divine entities such as the Erinyes. Correlative to this is the fact that from very early times, the Greeks, like many other peoples, believed that the dead benefited fromindeed demandedattention from the living. Most of the dead, however, are presented in these earliest sources as being weak, unable to affect the living and, as Anticleia told Odysseus

Punishments and rewards: A. Eu . 273, 320, 388; E. Alc . 744-46; Hel . 1013-14; Ar. Ra . 145-57, 448-78; Pl. Phd . 69c2-d1, 81a4-c3, 113d1-114c8; Grg . 523a5-524a7; Phdr . 248e6-249b1 R . 330d4-331a1, 363c3-d7, 365a1-3, 614b8-615c4.

Socially stratified: A. Pers . 681; Ch . 354; S. El . 836-39 Ethically judgmental: A. Eu . 94-95; cf. Pl. Phd . 108b2-c5.

Psychopompoi : S. Aj . 831-32; OC 1556-78; E. Alc . 253-54, 361, 439-41, 743-44; HF 431-34; Ar. Ra . 139-41, 268-70; Ar. Lys . 605-607. Cerberus: E. HF 610, 1276. We also see Hermes, Persephone, Hades, and even Earth take on the newer role of leading the souls back up again: A. Pers . 629-30; Ch . 124 483-90. In Ar. Ra . 464-78, Aeacus, elsewhere a judge of the Underworld, seems to take on the role of infernal porter. In Plato (Phd . 107d5) each departing soul is given its own guide.

Underworld geography: E. Alc . 252; Hec . 1; HF 1102; Ar. Ra . 186; 194; Pl. R . 614b8-621b1; Phd . 111c5-113c8.

when he visited the Underworld, separated from the living by an uncrossable river.

In narrative sources from the early classical period, we begin to see evidence for two new beliefs. First, the dead themselves can be a threat; they need not always work through agents such as the Erinyes, although they still do this as well.

Second, the dead can be called back into action by the living in a variety of ways, including some that we might label "magical." They can be used by the living both as sources of help, as the Persian queen uses Darius, for example, or as means of harming others.

In sources from the early classical period, particularly but not only tragedy, we generally see a great deal of interest in a variety of subjects connected with the dead, and especially in the possibility that the boundary between the Underworld and the upper world could be transgressed (in either direction). This may reflect, in part, tragedy's connection to Dionysus, but it also seems to reflect the fact that these topics were the object of significant interest and reflection at the time.

In general, all of these observations suggest a hypothesis: Greek beliefs evolved from a system in which the dead were relatively weak and unlikely to affect the world of the living, except under very special circumstances and then of their own volition, into a system in which the dead were an active force in the world of the living and could be called into action when the living chose.

It might be objected, of course, that our narrative sources may be defective informants, and that we are therefore constructing a faulty chronology and seeing "development" where none existed. Perhaps a belief in the active dead, and in the power of the individuals to put them to work, existed at the time when the Homeric poems reached their current form, but the poems' composer(s) either had no reason to mention them or did not wish to mention them. I grant that this is possible, bat I find it unlikely for several reasons.

Let us take up the question of opportunity first. As we shall see in the next chapter, one of the contexts in which the dead are most frequently called into use by the living is the agonistic context. That is, the dead are used in situations in which two or more people are competing and only one can win, such as a chariot race, a court case or a love affair. This is precisely the sort of situation that occurs frequently in the Iliad and occasionally in the Odyssey , and yet there is no mention, ever, of using the

dead to gain a competitive edge. There are other incidents, too, where we might expect to hear about the active dead if a belief in them existed, such as the story of Althaea's cursing of Meleager after the death of her brother. But rather than calling on the ghost of her brother himself to avenge his murder (as a ghost might do in later sources) Althaea calls on the gods of the Underworld and they send up an Erinys, a divine agent, to work her curse. Similarly, Homer shows us an Oedipus who is haunted not by the ghost of his mother herself but by an Erinys to whom the ghost has delegated the task. To take another instance, the grieving Achilles makes no attempt to call up the ghost of Patroclus, as Gilgamesh calls up the ghost of Enkidu in another heroic epic. The ghost appears, but only of its own volition and precisely because it falls into the exceptional category of the unburied dead.86

Which brings us to a second possible objection. Perhaps we do not see evidence for belief in a more active dead, and especially for belief in the possibility of the living invoking the dead, in these and other episodes where we might expect to find it because the poet thought it was inappropriate for his characters to hold such beliefs. This is one manifestation of a broader view that used to be fairly popular, namely that the "heroic" ethos of the Homeric poems, and particularly of the Iliad , rejects any taint of the supernatural because admirable men such as Achilles or Hector would have had no truck with it. Even E. R. Dodds, one of the first and most sensitive scholars of Greek beliefs in "irrational" phenomena such as ghosts, assumed that the "seemliness and epic dignity" of Homer's world would have been impaired if the poet had admitted such things. Under this approach, incidents such as Achilles' talking horse, the appearance of Patroclus's ghost to Achilles or the proleptic appearance of the suitors' ghosts to Theoclymenus are explained away in various fashions in order to preserve a Homeric world that keeps nicely to our own, modern standards of reality. Or they are grudgingly accepted but it is emphasized that such scenes are relatively uncommon in the poems, and probably accidental imports from other, contemporary but "less admirable" epics, such as the Thebais or Argonautica .87

Althaea: Il . 9.566-72; Epicaste: Od . 11.279-80; Patroclus's ghost: Il . 23.65-74; Gilgamesh: Tablet XII.

Dodds 1951, 70. Also on the assumption that Homer avoided mentioning the supernatural, see, e.g., Luck, 242; Janko, 53; M. Edwards, 283; Griffin, 40-42; Eitrem 1928. See also discussion in chapter 7 below and Johnston 1992b on Achilles' talking horse in particular.

Interestingly, scholars who hold dear this view of the Homeric world never have any problem accepting the constant interference of the gods on the battlefieldeven when it takes the spectacular form of the gods snatching their favorites away from danger in the middle of a cloud or the appearance of miracles wrought by the gods, such as the tears of blood that preceded Sarpedon's death and the wondrous conveyance of his body from the battlefield and home to Lycia afterwards. Nor do these scholars seem to have any problem accepting, for example, Hera's use of Aphrodite's magical kestos , Odysseus's use of the magical herb moly to avert the enchantress Circe (much less Hermes' prior possession of this herb), or the use of incantations by Odysseus's uncles to stanch the blood in his wound.88 We seem to glimpse behind this very carefully constructed worldview two unspoken rules:

1. Fantastic occurrences are acceptable so long as they are wrought by the gods. This reflects an old attempt among historians of religion earlier in this century to divide "miracle" from "magic" on the basis of who performed it: gods (or rather, God) in the case of the former, mortals or demons in the case of the latter. Implicit of course is the assumption that the former is "good" and marks the hero who benefits from it as one of the "chosen" but the latter is "bad" and must be rejected. This presumed dichotomy between miracle and magic was rejected several decades ago by subsequent historians of religion after further research showed that it was inapplicable to almost any religious system other than Christianity and perhaps Judaism; those who continue to apply it to systems other than these reveal a completely inappropriate (and probably unconscious) Christianocentric bias. It must be emphasized that until very late times, there is nothing in our sources that suggests that the ordinary Greek made any such division or rejected the mortal use of miraculous powers as "unethical" or "demonic."

2. Fantastic occurrences are acceptable so long as they are wrought by females (Aphrodite, Hera, Circe). I hope that I need to say little about the problems in accepting this premiseit should be clear that it reflects, again, a presumption inherited from earlier generations of scholars, according to which most forms of magic, and especially disreputable, deceptive types of magic, were the business of women. As we shall see in chapter 3, the most important form of Greek magic and the

Sarpedon's death: Il . 16.454-61, 667-83; Aphrodite's kestos: Il . 14.153-351; Hermes' moly: Od . 10.287-88; Odysseus's uncles: Od . 19.455-58.

one that relied on interaction with the dead (goeteia ) was exclusively the province of men until the early imperial period.

I note that neither of these unspoken rules takes care of the scene in which Odysseus's uncles sing incantationsthis is usually passed over silently by the scholars who cling to an anti-magical view of Homer. But let us go back to the specific issue. Once we accept a more catholic Homeric worldview, in which such things as ghosts and magic are comfortably at home, is there still reason to assume that interaction with the dead, in particular, would have been rejected as somehow unheroic and thereby have been omitted from the poems deliberately, despite the existence of such a belief at the time when the poems were being composed? Did Homer or his audience, in other words, consider it to be intrinsically unheroic to call upon the dead?

The first thing we must remember in considering this possibility is the fact that there are plenty of characters in the Homeric poems, such as Althaea and Epicaste, who have the opportunity to make use of the active dead if they wish and who are not bound by any heroic standards that might prevent them from doing so. If belief in an active dead, much less a belief in the power of the living to invoke them, was current at the time of the poems' composition, it is hard to imagine that the poet could so successfully suppress all awareness of it in cases such as these (as well as throughout the extensive nekuia of Odyssey 11), particularly given the slippages that occur elsewhere in the poemsthe contradictions, the anachronisms, the minor inconsistencies of plot. It is far easier to assume that he simply did not know of any such belief.

Secondly, even if it were considered unheroic to call upon the dead, the heroic stature of any given character could scarcely be affected by the unsolicited appearance of a dead person at the dead person's own initiative. In fact, Patroclus's ghost does exactly this, without any apparent damage to Achilles' reputation. If we were to try to construct some rule to explain this episode while preserving the general rule that Homer suppresses belief in the return of the dead, we would end up with an absurd premise such as "the unburied dead make honorable ghosts and are allowed to appear so long as they articulate heroic values in their interaction with the living, but all other ghosts would be unheroic no matter what they said." Nor does Theoclymenus's proleptic vision of the ghosts of the newly dead suitors, on their way down to the Underworld, cast any unheroic shadow on Theoclymenus. His own audience laughs and

calls him senseless, but we, the omniscient audience of the poem itself, are meant to accept and respect his vision.

Finally, to return to an observation made earlier, when in later times belief in a more active dead does become common, and the dead begin to be used against competitors in agonistic situations, there is no evidence that any censure is attached to the process. Nor does there seem to be any censure attached to consulting the dead at oracular shrines, for example. The dead were understandably approached with a hesitancy born of uncertainty concerning the mood they might be in, but not with any hesitancy born of shame. If we are to assume that it was unheroic for the Homeric warrior to use or even interact with the dead, we must also assume either that standards of acceptable behavior changed quite a bit between the composition of the poems and the late archaic age or that the Homeric hero was always an extraordinary character, held to standards that not even members of the earliest audiences of the poems would have embraced for themselves. Again, I think it far easier to assume that belief in an active dead was absent at the time when the poems were being composed.

As I said earlier, I believe it is best to look for the simplest explanation for Homer's omissions. I trust that this brief discussion has demonstrated that any approach other than assuming that Homer was unacquainted with the beliefs we see well attested in later sources requires elaborate and unlikely presumptions. Thus, the conclusions I proposed at the beginning of this section seem to offer a good working model of how Greek beliefs evolved. Some further support will be added to this model in the next chapter, where I examine non-narrative evidence and attempt to show, in several cases, how rituals in which the living interacted with the dead developed and changed. In chapter 3 I discuss the model as a whole and offer some reasons that the beliefs developed as they did.

Excerpted from Restless Dead: Encounters between the Living and the Dead in Ancient Greece by Sarah Iles Johnston Copyright 1999 by Sarah Iles Johnston. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.