Windy words in Penelope's joking dream: Odyssey 4.787-841.
As the Telemachia draws to a close, Penelope lies in her chamber, so worried about whether her son will survive the suitors' ambush that she cannot bring herself to eat or drink. When sleep finally overcomes her, Athena sends a dream that assures her of Telemachus's safe return, and Penelope seizes the opportunity to ask whether Odysseus is alive. The apparition refuses to answer for the perplexing reason that "it is bad to babble windy things" ([TEXT NOT REPRODUCIBLE IN ASCII], Od. 4.837). An answer to Penelopes question would be anything but [TEXT NOT REPRODUCIBLE IN ASCII], whether construed as "trivial," "empty," "without purpose," or "ineffectual." Knowing that Odysseus is safe would shatter Penelope's defining uncertainty and change the entire second half of the epic, as the PQ scholion ad 4.796 astutely notes. The apparition might be expressing ignorance, but, as Stephanie West (1988,244, ad 4.837) writes, it is "absurd that the dream-figure should thus allege lack of reliable information as grounds for its refusal." I suggest here that the apparition's response is actually a brilliant example of multilayered metapoetic wordplay. Later Hellenic tradition features dreams whose interpretations depend on solving puns or word-riddles embedded in their content, (1) and such riddles or puns would be perfectly at home in the Odyssey, whose frequent engagement in wordplay is well-known. (2) Although most instances of Odyssean wordplay seem to be serious in tone, (3) some have been considered humorous. (4) The wordplay in Penelope's dream, I suggest, is of the latter sort: its rhetorical mechanisms help the passage develop an emotional rapport between its audience and Penelope which its concluding joke brings to a culmination.
Before considering how this wordplay works, let us examine some of the scene's other puzzling aspects, which cannot be justified by the literary nature of Penelope's dream, since some of them lack any obvious literary purpose while others actually seem to contradict the dream's ostensible literary functions. These aspects appear even odder in light of both Homeric and non-Homeric visitation and dream typology. (5) Perhaps the most striking feature is that Penelope's joy is postponed until after she awakens, with the result that she seems pleased by the apparition's refusal to speak about Odysseus rather by its good news about Telemachus. This is not simply a matter of Penelope needing to wake up before she can react emotionally to her dream, since she cries "though in a dream" ([TEXT NOT REPRODUCIBLE IN ASCII]) at Odyssey 19.541. (6)
The dream-figure itself is no less strange. Only Zeus sends proxies in comparable scenes; in fact, in Homeric poetry, he always uses messengers instead of appearing to mortals personally. Athena, in contrast, always deals with humans personally, as, for instance, when she appears to Nausicaa in a dream that is otherwise strikingly similar to Penelope's. Christine Walde (2001, 47) theorizes that Athena sends the apparition to avoid disturbing Penelope with her own divine presence and thus disrupting the dream's calming effects. Yet if this were the case, Athena could simply have maintained her disguise as she does in Nausicaa's dream. Her absence does not seem to matter anyway, since Penelope surmises that her visitor is divine (Od. 4.831) without causing herself further agitation. The PQ scholion ad 4.796 offers another theory--that Athena sends the apparition "so that she won't be forced to say something about Odysseus" (7)--but surely Athena could have avoided answering Penelope at least as well as the dream-vision does. To imply that Penelope's question would put Athena in an awkward position smacks of the sort of Platonic moralism that is so alien to Homeric gods. Also, if this were Homer's primary concern, he could simply not have had Penelope ask her question in the first place, particularly since the question itself is unparalleled. (8) Dream-dialogue is extremely rare outside Homer, (9) and in Homer the only other person who ever speaks while dreaming is Achilles. He speaks to Patroclus's ghost, but only once, beginning with a question about why Patroclus has come (Il. 23.94-5). Penelope not only asks her dream-figure the same question (Od. 4.810), but she 'one-ups' Achilles by speaking again and asking a second question that introduces an entirely new subject. It is this unique second question that the dream-figure refuses to answer. Nor does it seem plausible that Athena would send an apparition to avoid denigrating herself or Penelope by refusing to answer Penelope's question personally or by manifesting in disguise without being recognized. Athena often influences Penelope even in ways that directly oppose Penelope's desires (e.g., Od. 18.178-97) without disgracing her in any evident way, and although Athena never appears to Penelope, disguised or not, operating unseen approximates the effect of operating in disguise. (10) So there seems to be no obvious reason why Athena does not come herself.
Still more peculiar is the fact that the figure Athena sends is called an [TEXT NOT REPRODUCIBLE IN ASCII] rather than (an?) [TEXT NOT REPRODUCIBLE IN ASCII]. Modern scholars generally refer to all ancient dream-figures as [TEXT NOT REPRODUCIBLE IN ASCII], (11) but this use seems to derive from late fifth-and fourth-century discussions of dreams in philosophical treatises on optics and ontology, where [TEXT NOT REPRODUCIBLE IN ASCII] is used as a technical term for all sorts of "images" or "appearances." Beyond such philosophers as Democritus, Plato, and Aristotle, (12) no archaic or classical Greek author seems to refer to a dream-figure or dream as an [TEXT NOT REPRODUCIBLE IN ASCII], without the figure also being a ghost and so warranting the label for that reason. (13) On the other hand, because Penelope describes "dreams" ([TEXT NOT REPRODUCIBLE IN ASCII]) as "un-measured in speech" ([TEXT NOT REPRODUCIBLE IN ASCII], Od. 19.560), a word applied elsewhere to Ihersites and Priam (Il. 2.246 and 2.796), and because [TEXT NOT REPRODUCIBLE IN ASCII] have their own [TEXT NOT REPRODUCIBLE IN ASCII] (defended territory/populace (14)) at Odyssey 24.12, the Odyssey seems to share the Iliad's conception of preexisting, anthropomorphic [TEXT NOT REPRODUCIBLE IN ASCII] that the gods presumably can send. So if Athena is going to send a proxy rather than appear in Penelopes dream herself, why does she not send an [TEXT NOT REPRODUCIBLE IN ASCII], just as Zeus does in Iliad 2?
Whatever other effects and causes these odd elements may have, they cooperate to set up the complex word-game at the scenes end, in which a figura etymologica brings to light a syntactic ambiguity of the sort Aristotle describes as "when the synthesis signifies multifariously, but [each element] isolated [signifies] singly, such as [in the phrase] 'knowing letters'" ([TEXT NOT REPRODUCIBLE IN ASCII], Soph. el. 1.4). Penelope's dream ends as follows (Od. 4.836-40):
[TEXT NOT REPRODUCIBLE IN ASCII] Replying to her, the dim/insubstantial apparition said: "I will not tell you about that [man] from beginning to end, [Whether] he lives or is dead; for it is bad to babble empty things." After speaking thus, it slipped out by the cross-bar of the door-post Into the gusts of the winds.
Some readers, beginning with Eustathius, have noticed that the [TEXT NOT REPRODUCIBLE IN ASCII] displays the "airy" (15) nature that [TEXT NOT REPRODUCIBLE IN ASCII] traditionally exhibit when, for instance, they slip through people's arms "like smoke" ([TEXT NOT REPRODUCIBLE IN ASCII], Il. 23.100). (16) The word "winds" ([TEXT NOT REPRODUCIBLE IN ASCII]), meanwhile, highlights the etymological meaning of [TEXT NOT REPRODUCIBLE IN ASCII] (windy things). Thus, the scene presents an [TEXT NOT REPRODUCIBLE IN ASCII] that deprecates babbling "windy things" to explain why it does not answer a question whose answer would be anything but "windy" immediately before it demonstrates its own airiness by exiting through a crack into the winds. If we reexamine the apparitions language, we would realize that placing the idiom "to babble windy things" ([TEXT NOT REPRODUCIBLE IN ASCII]) in an epexegetic infinitive construction has made the phrase technically ambiguous. With the same words that it uses to deprecate babbling windy things, the vision also says, "It is bad for windy things to babble." This response makes much better sense: the [TEXT NOT REPRODUCIBLE IN ASCII] does not answer Penelope's question because windy things like itself should not babble on after saying what they are supposed to say. (17) If this wordplay were serious in tone, it would only underscore the dream's airiness in a way that would be not only vapid but counterproductive to the scene's explicit aims. If the wordplay is humorous, however, as I will argue it is, its humor would modulate the audience's emotional responses so that they align with Penelope's in a manner that helps explain the odd postponement of her joy until after the [TEXT NOT REPRODUCIBLE IN ASCII] departs.
But there is a daunting obstacle here. How can we tell whether this wordplay is supposed to be funny? Most scholars of Homeric humor examine laughter as a literary device or use laughing characters to identify comic moments. (18) Yet because the humor in Penelope's dream exists mostly in the narrative language, it would be presumptuous (though not impossible) to suppose that her joy suggests that she actually gets the joke. Those scholars who have discussed instances of Homeric humor that are directed exclusively at the audience tend to rely all but entirely on their own subjective responses to identify funny passages, (19) but using our own laughter to determine what is and is not humorous is not a viable methodology. Such subtle humor as I propose here can be difficult for even native audiences to appreciate, and we modern readers face a far greater challenge. Even if we disregard the problems posed by differences in senses of humor and the shift from oral to written presentation, many jokes require that their audiences be so thoroughly steeped in the jokes' native language and culture that they do not need conscious, deliberate thought to get' the jokes' references, implications, and tonal nuance. Such thought may interfere with the comic process. (20) The old adage is true: explaining why a joke is funny kills its humor. So the sort of analysis sometimes required to recognize a foreign joke's humor may prevent the joke from functioning or even being recognized as a joke, and the whole undertaking becomes self-defeating. No matter how hard we try, a joke from another culture may never seem funny to us.
Although no objective test for determining whether or not something is humorous exists, semiolinguistic humor studies may help here, even if they do entail beating an already dead joke with exactly the sort of technical analysis most lethal to laughter. Building upon earlier theories (particularly those of Algirdas Greimas, Patrick Charaudeau, Violette Morin, and Victor Raskin), Salvatore Attardo (1994) posits a descriptive rhetorical model for jokes. His schema cannot prove definitively whether or not any given passage is a joke (that is not the schemas purpose), but it can at least offer a paradigm against which Penelope's dream can be compared. If the passage sets up its concluding wordplay with the same rhetorical operations as Attardo identifies for jokes, it seems reasonable to propose that its wordplay may in fact be humorous, and to see whether reading the passage accordingly helps resolve its various conundrums. Scholarship may never acculturate us sufficiently to laugh, but we may be able to understand why an ancient Greek raised on Homer might have laughed, and perhaps even be able to smirk ourselves.
According to Attardo, jokes generate humor from the sudden switch in how their audience understands an ambiguous lexical element, called the connector, from an original sense to a new, previously unnoticed one. This new sense must be antonymic or antagonistic to the original sense so that a logical incongruity arises between them, and causes the new sense to seem "ridiculous," "absurd," or "nonsensical." Humor is generated when an audience "resolves" the incongruity by temporarily adopting a "local logic" capable of accepting the new, incongruous sense of the connector (1994, 143-5). (21) The process occurs over three narrative functions, which often take the form of discrete phases in the joke's exposition. The first function establishes the context and uses logical norms to direct the audience along one particular interpretive path so that when they encounter the ambiguous connector they will initially notice only one of its possible interpretations. The second function "creates expectations'; it introduces the need for a resolution in the story ... in verbal humor, it often contains the connector" and usually takes a dialogical or "non-narrative form" (88-9). The third phase "resolves the process in an unexpected, non-standard fashion that goes against the expectations set up in [function 1] and [function 2] and often is nonsensical" (91). At the heart of this phase lies the disjunctor, which is "the element that causes the passage to switch from the serious to the humorous sense, and hence is responsible for the humorous effect itself" (89). It compels the audience suddenly to perceive the previously unperceived second sense of the connector and to invert the primacy of the two senses so that the newly recognized one dominates. The disjunctor often takes the form of a "punch-line."
Let us now analyze the rhetoric of Penelope's dream to see how well it accords with Attardo's model. I will address the effects of its humor afterwards.
The Rhetorical Mechanics of Penelope's Dream
The phrase [TEXT NOT REPRODUCIBLE IN ASCII] (Od. 4.837) appears to have been an idiom (so Eustathius 1.194.24; cf. 1.428.6). Odysseus demonstrates its use at Iliad 4.355 when he responds to Agamemnon's insults by saying, "You are babbling empty things here" ([TEXT NOT REPRODUCIBLE IN ASCII]). Agamemnon plays upon Odysseus's diction in his reply: "May the gods set my whole speech among the winds" ([TEXT NOT REPRODUCIBLE IN ASCII], Il. 4.363). Such linguistic or semiotic games may traditionally have accompanied this phrase, in which case the audience would have anticipated something similar in Penelope's dream. Anticipation of humor, though not necessary for a joke's success, can certainly contribute to it. (22)
Odysseus also uses the idiom in the Odyssean Nekyia when he responds to the ghost of Agamemnon's question about Orestes' welfare by saying: "I don't know at all [whether] he lives or has died; and it is bad to babble empty things" ([TEXT NOT REPRODUCIBLE IN ASCII], Od. 11.463-4). Because Odyssey 11.464 is the same as Odyssey 4.837, Odysseus's syntax is technically ambiguous, and some environmental features do resonate with its other possible meaning--Hades is associated with [TEXT NOT REPRODUCIBLE IN ASCII], at least in the allegorical literature; (23) ghosts are physically "empty"; and Agamemnon is called a [TEXT NOT REPRODUCIBLE IN ASCII] (Od. 11.387), a term that, at its most basic level, means "breath" (24)--but context clearly disambiguates the verse. Odysseus could hardly be expected to disparage "windy things" speaking, when he has come specifically in order to hear a shade speak, and he is the one who would babble [TEXT NOT REPRODUCIBLE IN ASCII], not Agamemnon's ghost. Thus, while there is again the potential for wordplay, it goes unrealized. Only an overly fussy pedant who has spent far too much time dissecting the passage would even notice it here.
Penelope's dream evokes this conversation in such a way that Odysseus's unambiguous statement at Odyssey 11.464 predisposes the audience initially to overlook the ambiguity with which the [TEXT NOT REPRODUCIBLE IN ASCII] utters the same verse at Odyssey 4.837. By thus drawing upon its audience's Hellenic enculturation and intimate familiarity with Homeric poetry, the passage fulfills function 1 of Attardo's model for jokes. Both Penelope's dream and the Nekyia scene feature an anxious discussion between a living person and a phantom about the safety of a son, which ends with the same verse being spoken in answer to an enquiry about a family member's welfare. Orestes, the object of Agamemnon's concern, is held up throughout the Odyssey as a model for Telemachus, who is the primary object of Penelope's anxiety, and two of the places Agamemnon suggests Orestes might be (Pylos and Sparta) are the two places Telemachus visits during the expedition that leads to Penelope's dream. Both scenes also feature uxorial fidelity as a prominent theme, and Odysseus as the explicit (and surprising) subject of immediate concern in Penelope's dream and as the addressee in the Nekyia.
Various facets of Penelope's dream attune the audience to recognize these thematic parallels by cooperating to generate a sense of proximity to the underworld. Dreams were closely associated with the dead and the underworld throughout Greek thought. (25) R. G. A. Van Lieshout (1980, 34) notes that "supernatural powers [including Olympian gods] involved in man's dreams are preferably chthonically conceived," while Eric Dodds (1951, 110-1) writes that enstatic dreams such as Penelope's were typical of incubation "practiced at the shrines of heroes--whether dead men or chthonic daemons--and at certain chasms reputed to be entrances to the world of the dead." In postclassical Greek dream-literature, the term [TEXT NOT REPRODUCIBLE IN ASCII] was used for both ghosts and dream-figures, (26) but in non-philosophical classical literature, it was applied only to dream-figures when they are ghosts, such as the [TEXT NOT REPRODUCIBLE IN ASCII] of Periander's wife that appears during an incubation in a necromanteion (Herodotus 5.92.4). Homer uses the term with an even narrower association with the underworld, reserving it all but exclusively for ghosts of the dead. (27) The only exception (apart from Penelope's visitor) is the [TEXT NOT REPRODUCIBLE IN ASCII] of Aeneas in Iliad 5.449 and 5.451, but this instance is remarkable enough that Plutarch, for instance, writes that Aeneas's [TEXT NOT REPRODUCIBLE IN ASCII] exists "as though [Aeneas] were dead" ([TEXT NOT REPRODUCIBLE IN ASCII], Mor. 1105F). Both Eustathius (1.193.10) and Joachim Hundt (1935,67-8) consequently remark that the [TEXT NOT REPRODUCIBLE IN ASCII] in Penelope's dream resembles a ghost.
The epithet [TEXT NOT REPRODUCIBLE IN ASCII], which modifies the [TEXT NOT REPRODUCIBLE IN ASCII] twice during Penelope's dream (Od. 4.824 and 4.835) and occurs nowhere else in Homeric poetry, enhances this resemblance. Although the root's precise definition is unclear and its etymology unknown, (28) its attestations suggest two different, though related, meanings: 'difficult to see/dark' and 'feeble/insubstantial.' (29) In archaic literature the root often serves adjectivally as a poeticism for 'dead' and verbally as 'to kill,' often with the sense of 'ruin,' like other words with similar meanings. (30) While the root does not modify dreams again in the extant archaic, classical, or Hellenistic literature (so far as I can determine), it is used not uncommonly in contexts involving the dead and the underworld by, for instance, Sappho (Fr. 55.4 Lobel-Page), Aeschylus (Pers. 223 and Fr. 273.6 Radt), and Euripides (Fr. 781.64 Nauck). The root has another, perhaps more important function, as I shall discuss later. The [TEXT NOT REPRODUCIBLE IN ASCII] Penelope sees is not a ghost; it is a divine fabrication, akin to Aeneas's [TEXT NOT REPRODUCIBLE IN ASCII] in Iliad 5, which happens to be seen in a dream. Nonetheless, its unique presence in a dream and its epithet [TEXT NOT REPRODUCIBLE IN ASCII] cause it to resemble a ghost closely enough to concentrate in Penelope's dream the underworldly atmosphere that Hellenic dreams often have--however nebulous that atmosphere may be in most other Homeric dream-scenes.
Two other Homeric dream-scenes with clear underworldly overtones bear very close structural resemblance to Penelope's dream. According to some scholars, Penelope's dream resembles Agamemnon's in Iliad 2, (31) but its resemblance to Achilles' dream in Iliad 23 is far more substantive. (32) While all three scenes share the typological elements that the dream-figure stands over its visitee's head ([TEXT NOT REPRODUCIBLE IN ASCII], Od. 4.803; Il. 2.20,2.59; Il. 23.68), begins its speech with "You are asleep" ([TEXT NOT REPRODUCIBLE IN ASCII], Od. 4.804; Il. 2.21,2.60; Il. 23.69), and addresses its visitee by name, (33) Penelope's and Achilles' dreams also have other, non-typological similarities: both dreamers refuse water (Achilles for washing; Penelope for drinking), find food abhorrent (though Achilles does eat), isolate themselves, are beset by cares for a loved one, are shortly thereafter overcome by a "loosening" (Il. 23.62, Od. 4.794) and [TEXT NOT REPRODUCIBLE IN ASCII] (sweet: Il. 23.63, Od. 4.793) sleep, and are visited at dusk or soon thereafter (34) by dream-figures that are identified directly (Penelope's apparition) or indirectly (Patroclus's ghost) as [TEXT NOT REPRODUCIBLE IN ASCII] (35) and that demonstrate their incorporeality when they leave. Most idiosyncratic to these two scenes, however, is that Achilles and Penelope are the only Homeric dreamers who engage their dream-figures in dialogue. While dialogue is unique in Homeric dreams to these two scenes, and very rare in non-Homeric dreams, it is perfectly standard for encounters with Homeric ghosts. Not only is Patroclus's [TEXT NOT REPRODUCIBLE IN ASCII] explicitly a ghost, it is identified as an [TEXT NOT REPRODUCIBLE IN ASCII] only because it is a ghost, not because it is a dream. (36) The next most substantive comparandum for Penelope's dream seems to be Priam's vision of Hermes in Iliad 24.677-88, (37) although it is not clear whether it is a dream- or visitation-scene due to the overlapping of these two scene-types. (38) Priam's entire expedition to Achilles' tent is well-known for being a figurative katabasis (39) in which Hermes serves as psychopomp, to which his role as "guide of dreams" ([TEXT NOT REPRODUCIBLE IN ASCII], Hymn. Hom. Herm. 4.14) is related. (40)
Certain elements of Penelope's dream suggest more specifically a metaphoric katabasis colored by necromantic incubation or ekstasis. C. J. Mackie (1999, 499-500) concludes that Achilles' abstinence from food, drink, and companionship in Iliad 18 and 19--an abstinence that Achilles reiterates through his reluctance to engage in these things before he sees Patroclus's ghost in Iliad 23--distances Achilles from life in preparation for his katabasis. Such abstinences, particularly fasting, were also part of the preparatory rituals for oracular incubation at necromantic cult centers such as Charon's Cave in Asia Minor. (41) Moreover, those who ventured to or into the underworld often experienced a metaphoric death that was anticipated by or even took the form of sleep. (42) Penelope's own abstinences resemble these poetic and ritual practices, as they prepare her for a sleep that is depicted by its prefatory simile as a figurative death (Od. 4.789-94): (43)
[TEXT NOT REPRODUCIBLE IN ASCII]. [Penelope lay] turning over in her mind whether her skillful son would escape death Or whether he would be killed by the overweening suitors. As much as a lion dithers amidst of a mass of men, Terrified, when they draw their crafty circle around him, While was fretting that much, sweet sleep came upon her; She fell asleep after lying down, for all her joints were loosened.
The standard reading of this passage holds that it extols Penelope's resourcefulness and anticipates her defeat of the suitors because "the lion can be expected to turn on its hunters and destroy them" (Podlecki 1971, 206), (44) but this is not the case. Penelope is explicitly not plotting how to defeat the suitors here. Instead, the "indirect deliberation of the either X or Y type" that Penelope's anxieties provoke is one of only two in all of Homeric poetry in which the subject contemplates possible events outside of her or his control rather than feasible actions. (45) By retaining the structure's form while altering its significance in this way, the poetry underscores the fact that Penelope does not have any options comparable to those the motif normally affords its subjects. Nothing she can do might help Telemachus escape. The word [TEXT NOT REPRODUCIBLE IN ASCII] expresses the urgent energy that would normally drive her to act, but here can only roil internally because efficacious action is impossible. (46)
The lion simile re-creates Penelope's mental state imagistically through a parallel rhetorical inversion. Lions may symbolize virile heroic qualities (47)--strength, aggression, courage--and be typical of Odysseus in the Odyssey, but the victory of Penelope's lion is anything but assured. Even in the Iliad, "lions and men are relatively equal antagonists," (48) and Penelope's lion simile is a unique formulation whose effect resembles that of Odysseus's at Odyssey 6.130-6, which strips its lion of heroic sublimity through suggesting that the lion's behavior is determined by its susceptibility to biological and environmental forces rather than by any heroic virtues. (49) Penelope's lion is afraid. In Iliad 12.41-8, a lion or boar confronts its hunters because "it is not afraid" ([TEXT NOT REPRODUCIBLE IN ASCII], Il. 12.46) and is killed (so we cannot assume Penelope's lion will kill its hunters!), while the lion in Iliad 17.109-11 is frightened by its human assailants and flees. The lion's fear in Penelope's simile therefore suggests that [TEXT NOT REPRODUCIBLE IN ASCII] does not mean that it is "deliberating" or "strategizing" how to defeat its hunters (50) but that it is "fretting" or "dithering" (51) in a reflection of Penelope's own "feckless anxiety." (52) Penelope's lion, like the one in Iliad 17.109-11, would escape if it could, but it, like Penelope, has no options. Penelope's trickiness, which had enabled her to hold off her suitors with the ruse of Laertes' shroud ([TEXT NOT REPRODUCIBLE IN ASCII]: Od. 2.93, 2.106, 19.137, 24.128, 24.141), belongs here to the hunters' "crafty [[TEXT NOT REPRODUCIBLE IN ASCII]] circle" rather than to the lion that represents Penelope--a transposition that leaves the lion (and Penelope) resourceless. Since it is still a lion, we can conjecture that it retains the potential to "turn the tables" on its hunters, but at this moment it can only roil impotently in its encircling trap, much like Penelope's mind, exacerbating the emotional turmoil both of them feel at their impending dangers.
At the moment when the hunters would attack, the narrative shifts back to Penelope as sleep "attacks" ([TEXT NOT REPRODUCIBLE IN ASCII]) (53) her and "loosens her limbs," an expression that Homer uses for various types of enervations but most frequently for death. (54) Odyssey 4.794 recurs only at Odyssey 18.189, where it again describes a "sweet sleep" ([TEXT NOT REPRODUCIBLE IN ASCII]) which overcomes Penelope. There, she redefines that sleep as a [TEXT NOT REPRODUCIBLE IN ASCII], a deathlike catalepsy that she then explicitly compares to death by wishing that Artemis would send her "a death as gentle" ([TEXT NOT REPRODUCIBLE IN ASCII], Od. 18.202) as her "gentle catalepsy" ([TEXT NOT REPRODUCIBLE IN ASCII], Od. 18.201). (55) The overtones of death that her sleep thereby acquires from these echoes would have seemed completely natural to Homer's audience, for whom sleep and death (and dreams) were so closely joined that sleep commonly served as a euphemism for death. (56)
Penelope takes the next step after "dying" when she first responds to the [TEXT NOT REPRODUCIBLE IN ASCII] at Odyssey 4.808. Her response here is the only time the extremely common (76 times) "then X responded to Y" formula does not immediately precede the speech it introduces. This unique postponement puts tremendous emphasis on the bizarre verse that intervenes: "Very sweetly slumbering in the gates of dreams" ([TEXT NOT REPRODUCIBLE IN ASCII], Od. 4.809). The verses first half clarifies that its second half is metaphorical, for Penelope has not actually left her chamber. Still, the very process of metaphor requires first contemplating the image literally, which transforms Penelope's enstatic dream into an ecstatic one, at least figuratively. (57) Ecstatic dreams are very rare in Greco-Roman literature and most follow the same basic pattern: the dreamer enters a deathlike state and then travels psychically to some chthonic or after-life place where he or she receives secret knowledge. (58) Penelope's metaphoric ecstasy brings her specifically to the Gates of Dreams that she herself describes in Odyssey 19.562-7. We cannot presume that Homer imagines that the Gates were situated at the edge of the underworld, where Vergil puts them, since although the Gates are well-attested in post-Homeric literature, (59) the Aeneid is the earliest extant source to give them a location. We cannot even presume that Homer conceives of them as physical structures rather than as purely allegorical. (60) Nonetheless, Eustathius's assessment of the verse is apt: "Sleeping in the Gates of Dreams makes the one so sleeping like a dead person" (1.193.37-8). If the audience were looking to track Penelope's path after her metaphoric death, as seems reasonable, the suitors' journey to the underworld provides the most immediate model. The last two landmarks they pass before entering the meadows of the dead are the "Gates of the Sun and the district of Dreams": ' [TEXT NOT REPRODUCIBLE IN ASCII] (Od. 24.12). There are no Gates of Dreams here, but the words [TEXT NOT REPRODUCIBLE IN ASCII] occur in such close proximity that I find it difficult to think an audience steeped in Homer and already prepared by other factors, both poetic and extrapoetic, to think of the underworld would not have recalled Odyssey 24.12 and the "district of Dreams" when they heard the phrase [TEXT NOT REPRODUCIBLE IN ASCII] at Odyssey 4.809, even if they did not imagine that the Gates of Dreams were located there already. Penelope herself even implicitly associates the Gates of Dreams with the underworld when she calls them the "Gates of Insubstantial/Fleeting (61) [[TEXT NOT REPRODUCIBLE IN ASCII]] Dreams" at Odyssey 19.562. Elsewhere in the Odyssey, [TEXT NOT REPRODUCIBLE IN ASCII] only describes the "insubstantial/fleeting heads of the dead" ([TEXT NOT REPRODUCIBLE IN ASCII], Od. 10.521, 10.536, 11.29, 11.49) in the underworld; its sole Iliadic occurrence (Il. 5.887) describes the deathlike existence Ares would have suffered had he not escaped from Diomedes. (62)
Once the more general thematic and atmospheric parallels have been well established, Penelope's second question tightens the connection to Odysseus's conversation with Agamemnon's ghost in the underworld. Her question and Agamemnon's parallel each other in subject (the welfare of a family member), in Odysseus's prominence, in rhetorical structure (three successive verses beginning [TEXT NOT REPRODUCIBLE IN ASCII] [Od. 4.832-4, 11.458-60], and in demanding an answer with the imperative [TEXT NOT REPRODUCIBLE IN ASCII] (4.832, 11.457).
Yet, while Penelope's second question aligns her dream with the Nekyia passage and so anticipates the same response Odysseus gives Agamemnon, it also begins the process of function 2 in Attardo's model by setting up a situation in which that response cannot possibly mean what it initially appears to. In the Nekyia, Odysseus introduces "It is bad to babble windy things" by explaining that he does not know anything about Orestes' welfare: "I don't know at all / [whether] he is alive or dead" ([TEXT NOT REPRODUCIBLE IN ASCII], Od. 11.463-4). Odysseus therefore uses [TEXT NOT REPRODUCIBLE IN ASCII] as a deprecatory term for the lies he would need to concoct were he to answer Agamemnon's question despite his ignorance. Penelope, on the other hand, sets the precondition that the [TEXT NOT REPRODUCIBLE IN ASCII] should only answer if it is a god or if it has heard the voice of a god with a phrase ([TEXT NOT REPRODUCIBLE IN ASCII], Od. 4.831) which expresses her conviction that the [TEXT NOT REPRODUCIBLE IN ASCII] meets her criteria (63) and so should know about Odysseus's welfare. If the [TEXT NOT REPRODUCIBLE IN ASCII] does not meet her criteria, why does it not profess its ignorance with the excuse she gives it readymade and say exactly what Odysseus does? Instead, the [TEXT NOT REPRODUCIBLE IN ASCII] replaces Odysseus's [TEXT NOT REPRODUCIBLE IN ASCII] (I do not know) with [TEXT NOT REPRODUCIBLE IN ASCII] (I will not proclaim), suggesting that unlike Odysseus it knows the truth but will not divulge it. Thus, [TEXT NOT REPRODUCIBLE IN ASCII] in Penelope's dream cannot mean what it does in the Nekyia. Nor can it mean "to speak a lie crafted despite knowledge," which is how Odysseus uses it at Iliad 4.355. The sequence "I will not tell you [[TEXT NOT REPRODUCIBLE IN ASCII]] ... and/for [[TEXT NOT REPRODUCIBLE IN ASCII]] it is bad to lie" would only make sense if the verb [TEXT NOT REPRODUCIBLE IN ASCII] meant either 'to speak the truth,' which it does not (e.g., Od. 13.327), or 'to speak a lie,' which it also does not. (64) The second exchange between Penelope and the [TEXT NOT REPRODUCIBLE IN ASCII], therefore, sets up the phrase [TEXT NOT REPRODUCIBLE IN ASCII] to be, if not completely nonsensical, certainly odd enough to pique the audience's curiosity, create "expectations," and introduce "the need for a resolution in the story," exactly as function 2 does for the connector.
The [TEXT NOT REPRODUCIBLE IN ASCII] exit "into the gusts of the winds" ([TEXT NOT REPRODUCIBLE IN ASCII], Od. 4.839) then triggers the wordplay in [TEXT NOT REPRODUCIBLE IN ASCII], just as the disjunctor of a joke does. Given the underworldly atmosphere, that the [TEXT NOT REPRODUCIBLE IN ASCII] "dissolves again into thin air," as Stephanie West (1988, 203) puts it, might remind the more skillful audience members of Patroclus's ghost slipping through Achilles' embrace "like smoke" ([TEXT NOT REPRODUCIBLE IN ASCII], Il. 23.100) or Anticlea's slipping through Odysseus's arms "like a shadow or even a dream" ([TEXT NOT REPRODUCIBLE IN ASCII], Od. 11.207) and then explaining that ghosts ([TEXT NOT REPRODUCIBLE IN ASCII]) fly "like a dream" ([TEXT NOT REPRODUCIBLE IN ASCII], Od. 11.222). Such parallels facilitate the inversion of the connector from the confusing "it is bad to babble empty things" to "it is bad for empty [windy] things [such as the [TEXT NOT REPRODUCIBLE IN ASCII]] to babble." Though far more intelligible in context, this meaning is both antagonistic to the phrase's initial sense and distorts Homeric idiom so that the audience must adopt a new logic for how Homeric language works in order to comprehend it.
Before I discuss how the humor helps resolve the passage's difficulties, I would point out that jokes often contain elements that at first seem irrelevant, but when viewed in retrospect enhance the humor by anticipating the new sense that the disjunctor gives the connector. (65) The phrase [TEXT NOT REPRODUCIBLE IN ASCII] can now be recognized as just such a hint. Two scholia on Odyssey 4.824 which attempt to define [TEXT NOT REPRODUCIBLE IN ASCII] offer "swift" among other possibilities and gloss the definition with [TEXT NOT REPRODUCIBLE IN ASCII] (something carried with winds). However patently erroneous this folk etymology may be, it brings to light a secondary pun, recognized at least by Hellenistic readers, which could explain why [TEXT NOT REPRODUCIBLE IN ASCII] appears twice in this passage despite occurring nowhere else in Homer: the phonological play of [TEXT NOT REPRODUCIBLE IN ASCII] with [TEXT NOT REPRODUCIBLE IN ASCII] anticipates the disjunctor's figura etymologica in both form (roughly) and meaning.
The wordplay in [TEXT NOT REPRODUCIBLE IN ASCII], therefore, does not simply occur naturally in the course of the poetry, as other, non-humorous instances of Homeric wordplay do. It is set up by what precedes it to be a surprising resolution to a disconcerting, bewildering situation, just like the disjunctor of a joke. Because Penelope's dream-scene operates according to the rhetorical model for jokes outlined by Attardo, we may reasonably hypothesize that it may be a joke and, accordingly, analyze the potential effects of its humor.
Effects of the Humor
The dream's underworldly atmosphere sets up its "punch-line" surreptitiously (as jokes do (66)). Its dominant function, as the context and introductory simile suggest, is to externalize Penelope's psychological state in a form that conduces to sympathy. By manifesting Penelope's fear, agitation, trepidation, and uncertainty in a mythological and cultic motif that the audience would already have associated with very similar emotions, (67) the atmosphere would trigger in the audience a visceral response approximating Penelope's own emotional state. Plato has the eponymous rhapsode of Ion 535C boast that a good Homeric performance could have just such an effect. Ending the dream-dialogue with a direct quote from Odysseus's own katabasis brings the underworldly atmosphere and its emotional effect to a culmination just as the [TEXT NOT REPRODUCIBLE IN ASCII], without any sensible justification, denies Penelope a resolution to the uncertainty that has ruined the past ten years of her life.
Then, suddenly, the disjunctor upends the syntax of that direct quote, dispelling the discomforting katabantic atmosphere for the audience with a humorous inversion of its previously disconcerting culminating verse. The basic notion that laugher expresses pleasurable relief at the sudden abatement of discomfort or disconcertion is an explicit or implicit component of almost all humor theories (68) and has been clinically proven by psychological studies. (69) Here, the sudden relief assuages the audience's discomfort over both Penelope's mental state and Telemachus's safety on an irrational level, reinforcing their rational knowledge that Athena will protect Telemachus and restore Penelope's husband to her. (70) Timing Penelope's relief immediately after the joke both prevents the tense katabantic atmosphere from being disrupted prematurely and synchronizes the relief that the joke produces in the audience with Penelope's own relief, thus maintaining the sympathy the audience feels with her. It also juxtaposes the relief both Penelope and the audience feel with the most directly relevant passage, namely the six-verse conclusion of the Telemachia in which the suitors lay their ambush for Telemachus.
These phenomenological effects would be hollow, however, if Penelope's relief were not justified within the story. The generally accepted explanation--that Penelope is relieved by the dreams reassurance about Telemachus (71)--seems incontrovertible, but it is also incomplete, for it does not explain why Penelope's relief is only mentioned after the [TEXT NOT REPRODUCIBLE IN ASCII] leaves, where it is uncomfortably juxtaposed with the apparition's refusal to tell her about Odysseus.
Let us look briefly at how Penelope's relief is constructed (Od. 4.839-41):
[TEXT NOT REPRODUCIBLE IN ASCII]. And she sprung up from sleep, The daughter of Ikarios; for her dear heart was cheered, Since (72) a clear dream rushed to her at night's milking-time.
The dream's visual clarity, its [TEXT NOT REPRODUCIBLE IN ASCII], secondarily implies the dream's trustworthiness in accordance with the distinction Penelope herself later draws between visual dreams, which are 'true,' and aural dreams, which are untrustworthy (Od. 19.565-7). Its predominant effect, however, especially when combined with the neuter [TEXT NOT REPRODUCIBLE IN ASCII] rather than with the anthropomorphizing [TEXT NOT REPRODUCIBLE IN ASCII], (73) is to associate Penelope's relief first and foremost with her visual experience of the dream as a whole. (74) But because Homer uses [TEXT NOT REPRODUCIBLE IN ASCII] elsewhere only to express the visual clarity with which gods appear to mortals, (75) here the word, focalized through Penelope, reiterates her belief that the [TEXT NOT REPRODUCIBLE IN ASCII] is divine and thus suggests that she now experiences the dream as though it were a divine epiphany. Divine epiphany-scenes usually describe their percipients' emotional reactions only after the gods leave, even when those reactions correspond with language spoken earlier in the scene, (76) and Penelope, who initially mistakes the [TEXT NOT REPRODUCIBLE IN ASCII] for her actual sister, does not even recognize the divine character of her dream until well into its course.
The epiphanic tenor of [TEXT NOT REPRODUCIBLE IN ASCII] reflects the fact that the word still retains in Homeric poetry its etymological association with gleaming white radiance. (77) That association helps frame Penelope's relief in terms of the passage's katabantic imagery by creating a photic contrast between [TEXT NOT REPRODUCIBLE IN ASCII] near the beginning of Odyssey 4.841 and [TEXT NOT REPRODUCIBLE IN ASCII] at the verse's end. Mediated by the verb "rushed," the contrast identifies the source of Penelope's relief with the sudden disruption of the night's darkness by what she now perceives as her dream's divine brightness. (78) If the phrase [TEXT NOT REPRODUCIBLE IN ASCII] earlier in the passage means "dark apparition" rather than "insubstantial apparition," this sudden shift would be even more pronounced. "Night" picks up the scene's underworldly tenor. We may think, for instance, of Odysseus's description of the region where he begins his katabasis as perpetually covered by a sunless and "baneful night" ([TEXT NOT REPRODUCIBLE IN ASCII], 11.19), but the association extends well beyond that passage. (79) The specification [TEXT NOT REPRODUCIBLE IN ASCII] enhances this implication by indicating that the dream arrives at evening twilight, (80) when people traditionally begin their katabaseis. (81) Because people traditionally return from the underworld at morning light, the mention of the dream's brightness only at the apparition's departure marks Penelope's own return from her emotional katabasis. Her concomitant relief, therefore, seems to dramatize the formulaic phrase [TEXT NOT REPRODUCIBLE IN ASCII] (happy from death: Od. 9.62, 9.565,10.133), which in Homer still retains traces of its original meaning "having returned from death," and which continues to be associated with seeing light down through the classical period. (82) Thus, Penelope's dream first manifests (rather than causes) her emotional katabasis and then brings her back from it when she eventually perceives the brightness that represents the dream-figure's divinity, the dream's visual clarity, and consequently the veracity of its news about Telemachus. Its humor ensures that the audience experiences a parallel emotional journey.
The joke in this passage is what humor theorists, speaking technically, call a pun, a joke whose incongruity arises from ambiguities in its connector's signifiers, (83) often with a familiar phrase providing the dominant sense on which the non-dominant sense plays. (84) In essence, a pun asks its audience to pretend that two words or phrases that sound the same actually are the same. Because a pun's local logic is discarded as ridiculously untenable once the pun is over, it follows that every pun implicitly reconstructs and reconfirms linguistic conventions in such a way that the speaker demonstrates his dominance over language by re-appropriating its operations as his own. (85) Through the process of breaking down and rebuilding the linguistic machine, the Odyssey poet displays his recognition of, and mastery over, the conventions of epic language with a maneuver that parallels other ways in which the two Homeric epics assert their superiority by reinventing traditional epic language and structures, such as reconfiguring traditional type-scenes in novel ways. (86)
I say superiority rather than excellence, for if modern puns imply a contest of wits in which the punster proves his verbal superiority over his rivals, (87) how much more pronounced must this implication have been for the agonistic Greeks, who engaged in such verbal sparring during symposia, and for their rhapsodes, who performed competitively? Internal evidence supports this theory. Bruce Louden (1995, 43-4) concludes that Homeric characters use wordplay as a means of exerting power over others:
Puns indicate manipulation by the speaker. As in flyting speeches and curses, when Homeric characters engage in wordplay, they similarly index their power, their ability to exert control over hostile opponents. Athene, Odysseus and Penelope, among others, use wordplays to their benefit. We might contrast the example of the suitors, who attempt to use the same devices but whose essential lack of power is instead revealed with great irony.
Laughter among Homeric characters is almost always condescending, often laughed in triumph over an enemy who is thought defeated, (88) and it would be strange indeed if laughter within Homers poems were completely different from his audience's laughter. Perhaps the poet singing Penelope's dream laughs specifically at Iliad 4, whose own wordplay with the phrase [TEXT NOT REPRODUCIBLE IN ASCII] seems quite feeble in comparison, or even at himself (or another rhapsode) when performing the Odyssean Nekyia, which apparently fails to take any advantage of the potential for wit inherent in addressing the phrase to a ghost. Regardless, people who provoke laughter, even laughter at themselves, forge social bonds with their audiences which promote solidarity and facilitate persuasion. (89) With Penelope's dream, the Odyssey poet takes advantage of humor's power to draw his audience into laughing alongside him as they solve his clever linguistic game and he then surpasses his rivals. (90)
Ahl, F. 1985. Metaformations: Soundplay and Wordplay in Ovid and Other Classical Poets. Ithaca.
Amory, A. 1966. "The Gates of Horn and Ivory." YCS 20: 1-58.
Anghelina, C. 2010. "The Homeric Gates of Horn and Ivory." MH 67: 65-72.
Arend, W. 1933. Die typischen Scenen bei Homer. Berlin.
Attardo, S. 1994. Linguistic Theories of Humor. Berlin.
Barnouw, J. 2004. Odysseus, Hero of Practical Intelligence: Deliberation and Signs in Homer's Odyssey. Lanham, MD.
Bremmer, J. 1983. The Early Greek Conception of the Soul. Princeton.
Brennan, T. C. 1987. "An Ethnic Joke in Homer?" HSCP 91: 1-3.
Brillante, C. 1990. "Scene oniriche nei poemi omerici." MD 24: 31-46.
--. 1991. Studi sulla rappresentazione del sogno nella Grecia antica. Palermo.
Brown, C. 1989. "Ares, Aphrodite, and the Laughter of the Gods." Phoenix 43: 283-93.
Burrows, R. 1965. "Deception as a Comic Device in the OdysseyCW 59: 33-6.
Chantraine, P. 1968. Dictionnaire etymologique de la langue grecque: Histoire des mots. Paris.
Clarke, H. 1969. "The Humor of Homer." CJ 64: 246-52.
Clarke, M. 1999. Flesh and Spirit in the Songs of Homer: A Study of Words and Myths. Oxford.
Colakis, M. 1986. "The Laughter of the Suitors in the Odyssey." CW 79: 137-41. de Jauregui, M. H. 2011. "Priam's Catabasis: Traces of the Epic Journey to Hades in Iliad 24." TAPA 131: 37-68.
de Jong, I. 2001. A Narratological Commentary on the Odyssey. Cambridge.
Denniston, J. D. 1950. The Greek Particles. Second edition. London.
Desmidt, D. B. 2006. "Horn and Ivory, Bow and Scar." CQ 56: 284-9.
Dodds, E. R. 1951. The Greeks and the Irrational. Berkeley.
Duckworth, G. E. 1930. Foreshadowing and Suspense in the Epics of Homer, Apollonius, and Vergil. New York.
Fernandez-Galiniano, M. 1992. "Commentary on Odyssey XXI-XXII." In Russo et al. 1992,131-312.
Finkelberg, A. 1993. "Anaximanders Conception of the 'Apeiron.'" Phronesis 38: 229-56.
Frame, D. 1978. The Myth of Return in Early Greek Epic. New Haven.
Friedrich, R. 1981. "On the Compositional Use of Similes in the Odyssey." AJP 102: 120-37.
Golden, L. 1990. "To [TEXT NOT REPRODUCIBLE IN ASCII] in The Iliad." HSCP 93: 47-57.
Gruner, C. 1997. The Game of Humor: A Comprehensive Theory of Why We Laugh. New Brunswick.
Gunn, D. 1971. "Thematic Composition and Homeric Authorship." HSCP 71: 1-31.
Haller, B. 2009. "The Gates of Horn and Ivory in Odyssey 19: Penelopes Call for Deeds, Not Words." CP 104: 397-417.
Halliwell, S. 2008. Greek Laughter: A Study of Cultural Psychology from Homer to Early Christianity. Cambridge.
Harris, W. 2009. Dreams and Experience in Classical Antiquity. Cambridge, MA.
Hart, W. 1943. "High Comedy in the Odyssey." University of California Publications in Classical Philology 12: 263-78.
Hawley, R. 1968. "The Antiphonal Muse: Comic Sub-Theme in the Iliad" CW62: 81-2.
Hewitt, J. 1928. "Homeric Laughter." CJ 23: 436-47.
--. 1929a. "Humor in Homer and Vergil." CW 22: 169-72.
--. 1929b. "Humor in Homer and Vergil." CW 22: 177-81.
Hey, O. 1908. Der Traumglaube der Antike: Ein historischer Versuch. Munich.
Highbarger, E. L. 1940. The Gates of Dreams: An Archaeological Examination of Vergil, Aeneid VI.893-899. Baltimore.
Holowchak, M. 2002. Ancient Science and Dreams: Oneirology in Greco-Roman Antiquity. Lanham, MD.
Hundt, J. 1935. Der Traumglaube dei Homer. Greifswald.
Hunt, W. I. 1890. "Homeric Wit and Humor." TAPA 21: 48-58.
Kessels, A. H. M. 1969. "Ancient Systems of Dream Classifications." Mnemosyne 4: 389-424.
--. 1978. Studies on the Dream in Greek Literature. Utrecht.
Kirk, G. 1990. The Iliad: A Commentary, II: Books 5-8. Cambridge.
Koestler, A. 1964. The Act of Creation. London.
Lefcourt, H. 2001. Humor: The Psychology of Living Buoyantly. Waterloo.
Levaniouk, O. 2008. "Penelope and the Pandareids." Phoenix 62: 5-38.
Levine, D. 1982. "Odyssey 18: Iros as Paradigm for the Suitors." CJ 77: 200-4.
--. 1983a. "Homeric Laughter and the Unsmiling Suitors." CJ 78.2: 97-104.
--. 1983b. "Penelopes Laugh: Odyssey 18.163." AJP 104: 172-8.
Levy, E. 1982. "Le reve homerique." Ktema 7: 23-41.
Lieshout, R. G. A. van. 1980. Greeks on Dreams. Utrecht.
Louden, B. 1995. "Categories of Homeric Wordplay." TAPA 125: 27-46.
--. 2006. The Iliad: Structure, Myth, and Meaning. Baltimore.
Louis, P. 1948. "Sur le sens du mot [TEXT NOT REPRODUCIBLE IN ASCII] chez Homere." RPh 3: 63-72.
Lowenstam, S. 1993. The Scepter and the Spear: Studies on Forms of Repetition in the Homeric Poems. Lanham, MD.
MacHovec, F. 1988. Humor: Theory, History, Application. Springfield, IL.
Mackie, C. J. 1999. "Scamander and the Rivers of Hades in Homer." AJP 120: 485-501.
McKinlay, A. P. 1957. "On the Way Scholars Interpret [TEXT NOT REPRODUCIBLE IN ASCII]." AC 26: 12-39.
Meltzer, G. 1990. "The Role of Comic Perspectives in Shaping Homers Tragic Vision." CW 83: 265-80.
Messer, W. S. 1918. The Dream in Homer and Greek Tragedy. New York.
Morreall, J. 1983. Taking Laughter Seriously. Albany.
Morris, J. 1983. "'Dream Scenes' in Homer, a Study in Variation." TAPA 113: 39-68.
Nagler, M. 1974. Spontaneity and Tradition: A Study of the Oral Art of Homer. Berkeley.
Newton, R. 1998. "Cloak and Shield in Odyssey 14." CJ 93: 143-56.
Oberhelman, S. 1993. "Dreams in Graeco-Roman Medicine." In W. Haase and H. Temporini, eds., Aufstieg und Niedergang der romischen Welt. Berlin. Part II, vol. 37.1, 121-56.
O'Quin, K., and J. Aronoff. 1981. "Humor as a Technique of Social Influence." Social Psychology Quarterly 44: 349-57.
Parvulescu, A. 1985. "Homeric ([TEXT NOT REPRODUCIBLE IN ASCII]) [TEXT NOT REPRODUCIBLE IN ASCII]." Glotta 63: 152-9.
Pelliccia, H. 1995. Mind, Body, and Speech in Homer and Pindar. Gottingen.
Podlecki, A. 1971. "Some Odyssean Similes." G&R 18: 81-90.
Pollmann, K. 1993. "Etymologie, Allegorese und epische Struktur: Zu den Toren der Traume bei Homer und Vergil." Philologus 137: 232-51.
Pucci, P. 1986. "Les 'figures' de la Metis dans l'Odyssee." Metis 1: 7-28.
--. 1987. Odysseus Polytropos: Intertextual Readings in the Odyssey and the Iliad. Ithaca.
--. 1998. The Song of the Sirens: Essays on Homer. Lanham, MD.
Rapp, A. 1948. "The Dawn of Humor: The Transition from Ridicule to True Humor Seen in Homer." CJ 43: 275-80.
Raskin, V. 1985. Semantic Mechanisms of Humor. Dordrecht.
Ready, J. 2011. Character, Narrator, and Simile in the Iliad. Cambridge.
Redfield, J. 1975. Nature and Culture in the Iliad. Chicago.
Richardson, N. 1993. The Iliad: A Commentary, IV: Books 21-24. Cambridge.
Ritchie, G. 2004. The Linguistic Analysis of Jokes. London.
Rozokoki, A. 2001. "Penelopes Dream in Book 19 of the Odyssey." CQ 51: 1-6.
Russo, J. 1968. "Homer against His Tradition." Arion 7: 275-95.
--. 1982. "Interview and Aftermath: Dream, Fantasy, and Intuition in Odyssey 19 and 20." AJP 103: 4-18.
--. 1992. "Commentary on Odyssey XVII-XX." In Russo et al. 1992, 3-130.
Russo, J., M. Fernandez-Galiniano, and A. Heubeck, eds. 1992. A Commentary on Homer's Odyssey, III: Books XVII-XXIV. Oxford.
Scott, W. 1974. The Oral Nature of the Homeric Simile. Leiden.
--. 2009. The Artistry of the Homeric Simile. Lebanon, NH.
Scully, S. 1990. Homer and the Sacred City. Ithaca.
Seidensticker, B. 1982. Palintonos Harmonia: Studien zu komischen Elementen in der griechischen Tragodie. Gottingen.
Sikes, E. E. 1940. "The Humour of Homer." CR 54:121-7.
Slatkin, L. 1996. "Composition by Theme and the Metis of the Odyssey." In S. Schein, ed., Reading the Odyssey: Selected Interpretive Essays. Princeton. 223-38.
Stanford, W. B. 1939. Ambiguity in Greek Literature. Oxford.
Thalmann, W. G. 1988. "Thersites: Comedy, Scapegoats, and Heroic Ideology in the Iliad: TAPA 118: 1-28.
Turkeltaub, D. Forthcoming. "Penelope's 'Stout Hand' and Odyssean Humor." JHS.
--. Forthcoming. "Penelope's Lion, [TEXT NOT REPRODUCIBLE IN ASCII] Husband, and [TEXT NOT REPRODUCIBLE IN ASCII]- Destroying Pain." CJ.
Versnel, H. S. 1987. "What Did Ancient Man See When He Saw a God? Some Reflections on Greco-Roman Epiphany." In D. van der Plas, ed., Effigies Dei: Essays on the History of Religions. Leiden and New York. 42-55.
Walde, C. 2001. Die Traumdarstellungen in der griechisch-romischen Dichtung. Leipzig.
West, S. 1988. "Commentary on Odyssey I-IV." In A. Heubeck, S. West, and B. Hainsworth, eds., A Commentary on Homer's Odyssey, I: Books I-VIII. Oxford. 51-248.
Whitman, C. 1958. Homer and the Heroic Tradition. Cambridge, MA.
Williams, F. 1986. "Odysseus's Homecoming as a Parody of Homeric Formal Welcomes." CW 79: 395-7.
Zervou, A. 1990. Ironie et parodie: Le comique chez Homere. Athens.
(1.) Oberhelman 1993, 132-3.
(2.) See in particular Stanford 1939, 98, 101-14 and Louden 1995 with his useful bibliography. Ahl 1985 remains the most comprehensive work on Greco-Roman wordplay. Pucci (1986) and Slatkin (1996, 235-7) consider the Odyssey's wordplay an extension of its principle characters' [TEXT NOT REPRODUCIBLE IN ASCII].
(3.) Louden 1995, 32. For a general discussion of comic irony and parody in Homer, see Zervou 1990.
(4.) Louden 1995, 32 on Irus's name; Williams 1986 on Eumaeus's welcome of Odysseus in Od. 14; Newton 1998 on Odysseus's Thoas-story. Pucci (1987,161) suggests that the Odyssey "smiles at the high-minded lions of the Iliad." See Levy 1982, 34 and 37 on Athena's appearance to Odysseus at Od. 20.30-55 as a play upon normal dream typology; cf. Eurycleia waking Penelope at the beginning of Od. 23. Louden (2006, 53-9, 74-9, 80-111, 142-3) writes that many Iliadic scenes parody narrative patterns, by which he means that they employ what Aristotle calls "low mimetic" instead of "high mimetic" (54-5). Louden (rightly, I think) denies that these parodies are "necessarily comic" (80), but Aristotle identifies "low mimetic" as an essential distinction between comedy and tragedy. Some pre-Parry critics believed that formulaic epithets in infelicitous contexts (Penelope's [TEXT NOT REPRODUCIBLE IN ASCII] in Od. 21.6, for instance) could be ironic or humorous. See Zervou 1990, 6, 17-97; Lowenstam 1993, 16-7; and Turkeltaub, forthcoming in JHS. On other forms of Homeric humor, see notes 18 and 19 below.
(5.) Gunn (1971, 15-7) distinguishes visitations and dreams, though most scholars do not. Visitation scenes are those in which "a supernatural figure visits a mortal who has retired for the evening" (Morris 1983, 41). On Homeric visitation-scene typology, see Arend 1933, 56 note 2, 61-3, 99-105; Gunn 1971; Morris 1983; and de Jong 2001, 120-1. For Homeric dream-scene typology, see Dodds 1951, 104-10; Kessels 1978, 133-44; and Levy 1982. For general dream typology, see Dodds 1951, 105-6 and van Lieshout 1980, 12-28. Brillante (1990) analyzes non-dream visitation scenes that use dream typology and later (1991, 30) notes their resemblance to messenger scenes.
(6.) Cf. Od. 21.79. See Kessels 1978, 143 contra Duckworth 1930, 73.
(7.) Kessels (1978, 114 note 6) and West (1988, 243, ad 4.796) agree.
(8.) West 1988, 244, ad 4.830.
(9.) Hundt 1935, 68-9; van Lieshout 1980, 20; Harris 2009, 43 note 93.
(10.) Pucci 1998, 73.
(11.) E.g, Brillante 1991, 19-20 and van Lieshout 1980, 15, who follows with a valuable anthropological examination of [TEXT NOT REPRODUCIBLE IN ASCII]. Also see Hundt 1935, 17-28 and Kessels 1978, 87.
(12.) E.g., Aristotle, Div. somn. 464a; Plutarch, Moralia 735A; Plato, Ti. 71A and Soph. 266 (cf. Ti. 45D-46A, which compares [TEXT NOT REPRODUCIBLE IN ASCII] in dreams to [TEXT NOT REPRODUCIBLE IN ASCII] in mirrors); Aristotle, Insomn. 462a (cf. Div. somn. 464b, which compares [TEXT NOT REPRODUCIBLE IN ASCII] in dreams to [TEXT NOT REPRODUCIBLE IN ASCII] in water). On philosophical classifications of dreams, see esp. Kessels 1969.
(13.) Pace Holowchak 2002, 21, Patroclus's ghost is called a [TEXT NOT REPRODUCIBLE IN ASCII] (II. 23.65, 100, 106, 221). Because Patroclus apposes [TEXT NOT REPRODUCIBLE IN ASCII] with [TEXT NOT REPRODUCIBLE IN ASCII] in reference to ghosts in the underworld (Il. 23.72), we can deduce that he is also an [TEXT NOT REPRODUCIBLE IN ASCII], but only because he is a ghost, not because he is a dream. Homer does use [TEXT NOT REPRODUCIBLE IN ASCII] and [TEXT NOT REPRODUCIBLE IN ASCII] interchangeably (Clarke 1999, 157-215), but only while discussing ghosts, never dreams. I am therefore also unwilling to take the presumptive leap that Holowchak (2002, 23 note 3) does when he states that Oneiros in Iliad 2 is an [TEXT NOT REPRODUCIBLE IN ASCII]. By Holowchak's argument, Oneiros would also be a [TEXT NOT REPRODUCIBLE IN ASCII], which he dearly is not; cf. Dodds 1951, 122 note 10. According to Pindar (Fr. 131 Snell) the principle that remains posthumously as the [TEXT NOT REPRODUCIBLE IN ASCII] "shows" ([TEXT NOT REPRODUCIBLE IN ASCII]) impending decisions ([TEXT NOT REPRODUCIBLE IN ASCII]) during life; Pindar's diction suggests that what becomes the [TEXT NOT REPRODUCIBLE IN ASCII] after death causes, but is not itself, the dream; on this, see van Lieshout 1980, 37-8.
(14.) Thus Scully 1990,162 note 9: "[TEXT NOT REPRODUCIBLE IN ASCII] does not mean Village but a well-defended 'territory and all the free 'people' who inhabit it."
(15.) Cf. Eustathius, Od. 1.193 (ad 4.802). Homeric [TEXT NOT REPRODUCIBLE IN ASCII] is normally construed specifically as "mist" or "gloom," but the term can refer to vapor of any opacity (Louis 1948) and so does approximate our term "air" (Finkelberg 1993, 232-6).
(16.) Eustathius, Od. 1.193.9-12; Kessels 1978, 156; Brillante 1991, 20-1, 31-2; Hundt 1935, 66-7 (cf. 13-27).
(17.) Note that this reading does not rely on any connection between wind and Homeric [TEXT NOT REPRODUCIBLE IN ASCII] as potentially distinct from our "air," merely a connection between wind and the "airiness" that is intrinsic to the standard connotation of [TEXT NOT REPRODUCIBLE IN ASCII]. Nonetheless, wind was defined as moving [TEXT NOT REPRODUCIBLE IN ASCII] by Anaximander at the latest (e.g., Anaximander, Frr. 11 and 24 DK; Anaximenes, Fr. 7 DK; and Aristotle, [Mund.] 394b), who used [TEXT NOT REPRODUCIBLE IN ASCII] as roughly equivalent to our "air" (Finkelberg 1993). In Hesiod, Op. 551-3, [TEXT NOT REPRODUCIBLE IN ASCII] is suspended by wind and blows ([TEXT NOT REPRODUCIBLE IN ASCII]) when Boreas drives clouds; cf. Il. 5.865-6 and note 15 above.
(18.) Hunt 1890; Hewitt 1929a and 1929b; Sikes 1940; Rapp 1948; Hawley 1968; Seidensticker 1982, 50-64; Williams 1986; Brennan 1987; Thalmann 1988, 16-26; Zervou 1990; Newton 1998. The most thorough study of Homeric characters laughing is Halliwell 2008, 51-99. Also see Hewitt 1928; Levine 1982, 203-4 and also 1983a and 1983b; Brown 1986, 286-91; Colakis 1986; Russo 1992, 59, 70; Fernandez-Galiano 1992,195; de Jong 2001, 440.
(19.) See Hart 1943, Burrows 1965, Clarke 1969, Golden 1990, Meltzer 1990, Zervou 1990. For a different method, see Turkeltaub, forthcoming in JHS.
(20.) Raskin 1985, 80-92; cf. Koestler 1964.
(21.) On incongruity theories of humor, see Ritchie 2004, 46-68.
(22.) Raskin 1985, 80-92.
(23.) E.g., Plutarch, Mor, Horn. 97 and Heraclitus, Quaest. Horn. 23.9-11, 41.9. On [TEXT NOT REPRODUCIBLE IN ASCII] and wind, see notes 15 and 17 above.
(24.) Hundt 1935, 33-6; Chantraine 1968, 1294-5; Bremmer 1983, 22-3; Clarke 1999, 133-56.
(25.) For passages highlighting this relationship, see Messer 1918, 40-7 and van Lieshout 1980, 78-84.
(26.) See note 11 above; also Hundt 1935, 19-23; Dodds 1951, 110-1; Brillante 1991, 19-20; Clarke 1999, 157-215. Harris (2009, 47-8, 59-60) writes that the association between dream-figures and the dead persists in many modern cultures.
(27.) Homer, Il. 23.72, 23.104; Od. 11.83, 11.213,11.476, 11.602, 20.355, 23.104, 24.14.
(28.) Chantraine (1968) observes that [TEXT NOT REPRODUCIBLE IN ASCII] "ne possede pas d'etymologie."
(29.) Contravening previous thought, McKinlay (1957) argues that [TEXT NOT REPRODUCIBLE IN ASCII] first meant "weak" and subsequently developed the meaning "dark." While he is correct that [TEXT NOT REPRODUCIBLE IN ASCII] does not always mean "dark," I find his treatment of the (partial) evidence he adduces tendentious. Given his admission that there is a "tendency for light phenomena to come into the picture" (18), it would be just as easy to argue that the word developed as traditionally conceived through figurative use early in its history. This development would fit all of the attestations McKinlay lists and also explain cases such as Aeschylus, Fr. 273a.6 Radt, an early instance (missed by McKinlay) in which [TEXT NOT REPRODUCIBLE IN ASCII] can mean "dark" but not "weak."
(30.) Literal death: Hesiod, Fr. 204.142 M.-W; Pindar, Pyth. 12.13. Semi-literal death: Solon, Fr. 4.34 Bergk; Pindar, Isthm. 4.49; Aeschylus, Pers. 223. Entirely metaphorical ("ruin"): Hesiod, Op. 325, 693; Theognis, Fr. 1.192 Bergk; Pindar, Fr. 126.1.
(31.) Hey 1908, 11-2; Messer 1918, 25; Walde 2001, 45-7.
(32.) Messer 1918, 26; Kessels 1978,149-50; Richardson 1993,177; de Jong 2001, 121; Walde 2001, 45-7.
(33.) See note 5 above for dream typology.
(34.) Parvulescu (1985, 152-4) argues that Penelope's dream happens at dusk based on its proximity to dinnertime. His arguments apply just as well to Achilles' dream.
(35.) Achilles seems to distinguish [TEXT NOT REPRODUCIBLE IN ASCII] from [TEXT NOT REPRODUCIBLE IN ASCII] at Il. 23.104, but his language is probably pleonastic. See Clarke 1999, 157-215 and Brillante 1991, 19-20, contra Messer 1918, 18-20 and Hundt 1935, 28-39.
(36.) See note 13 above.
(37.) Both dreamers' primary concern is for a lost son. Hermes "stood over [Priam's] head" ([TEXT NOT REPRODUCIBLE IN ASCII], Il. 24.682); addresses him as "old man" ([TEXT NOT REPRODUCIBLE IN ASCII], Il. 24.683), which, except for when he calls Priam [TEXT NOT REPRODUCIBLE IN ASCII]in 24.362, is how Hermes always addresses Priam; and then reproaches him for still being asleep. Richardson (1993, 177) also notes the resemblance between II. 23.100-1 and Od. 24.5-9.
(38.) See note 5 above. Messer (1918, 23 note 68) and Richardson (1993, ad 24.677-86) argue that Hermes' appearance is not a dream. Kessels (1978, 58-9) cautiously agrees. Levy (1982, 27), Brillante (1990, 41-6), and Harris (2009, 38 note 67) argue that it is a dream.
(39.) On the katabantic atmosphere of Priam's journey, see esp. de Jauregui 2011. Also Whitman 1958, 217-20; Nagler 1974,184-98; Redfield 1975, 214-5; Frame 1978,153-5; Mackie 1999, 488-91.
(40.) Brillante 1990, 43-5.
(41.) Dodds 1951, 110.
(42.) Frame 1978, 38-9, 44, 73-5, 168 and de Jauregui 2011, 48-50; cf. Od. 13.79-80, 90-2.
(43.) I examine this figure more thoroughly in Turkeltaub, forthcoming in CJ.
(44.) Cf. Scott 2009, 31-2 and Levaniouk 2008, 20-1.
(45.) Pelliccia 1995, 127 note 32. The other instance (Od. 15.300) reflects this passage.
(46.) Barnouw (2004, 116) writes that [TEXT NOT REPRODUCIBLE IN ASCII] "like mermerizo, conveys ... an inner contention of tendencies or a turbulence of conflicting impulses" and suggests "incipient motion or action."
(47.) Scott 1974, 58-62.
(48.) Ready 2011, 62.
(49.) Friedrich 1981 and Pucci 1987, 158-61.
(50.) As, for instance, Levaniouk (2008, 21 note 78) claims.
(51.) Barnouw (2004, 109-11) concludes that [TEXT NOT REPRODUCIBLE IN ASCII], even when used as "plotting" (in instances of the "how to" type, on which see Pelliccia 1995, 126-8), always indicates in the Odyssey "inner and outer turmoil" (111) and mental states that are "unsettled, disturbed, needing resolution" (109). In the Odyssey "it is marked by this sort of setting [ones entailing 'anxiety' or are 'baneful, evil'] more than by any fixed order of thinking" (109).
(52.) Thus Pelliccia 1995, 127 note 32 on Penelope here; cf. Barnouw 2004, 68, 115-6.
(53.) This is the verb's normal meaning in Homer; see LSJ, s.v. [TEXT NOT REPRODUCIBLE IN ASCII], 1.1 and I.1.b.
(54.) Of the 42 instances of the motif, twenty-three mean to kill (Il. 4.469, 5.176, 7.12, 7.16, 11.249, 11.260, 11.579, 13.360, 13.412, 15.291, 15.435, 16.312, 16.400, 16.425, 16.465, 17.349,17.524, 21.114, 22.335, 24.498; Od. 14.69, 14.236, 24.381), another (Od. 18.238) probably does, and three more (Il. 21.406, 21.425, and Od. 18.242) burlesque killings.
(55.) Cf. Hesiod, Theog. 798; also in Il. 14.359. The verb for all three, [TEXT NOT REPRODUCIBLE IN ASCII], is a "death-marker" (Frame 1978, 21, 73-74 note 68).
(56.) Messer 1918, 41. Hesiod combines the births of Sleep, Death, and Dreams in one verse (Theog. 212).
(57.) West (1988, 243-4, ad 4.809) notes the uniqueness in Homer of Penelope's metaphoric ekstasis, which led Kessels (1978, 108) to reject any hint of ekstasis here. But ecstatic dreams were not unknown in archaic Greece, and the absence of snowstorms or cranes or pigmies from the events of Homer's poems does not stop Homer from using them figuratively.
(58.) Dodds 1951, 135-47. Van Lieshout (1980, 29-33) catalogues ecstatic dreams in Greece.
(59.) Plato, Chrm. 173A; Anth. Pal. 7.42; Vergil, Aen. 6.893-8; Horace, Od. 3.27.39-42; Statius, Silv. 5.3.285-90; Lucian, Ver. hist. 2.32-3. For further allusions, see Ahl 1985, 261-2.
(60.) Knowing the Gates of Dreams' origin could help shed light on Homer's conception of them. Highbarger (1940) traces them back to the Gate of Horns that stood at the entrance to the underworld in Egyptian and Babylonian mythology. Amory (1966, 6-12) effectively dismantles his argument, but Anghelina (2010) revives its most basic points. Ahl (1985, 263-4) discusses some evidence for a Near Eastern tradition juxtaposing ivory and horn (as it pertains to Ovid's Pygmalion story, see pp. 236-70). Russo (1982, 10 note 13 and 1992, 103) argues that Od. 4.809 suggests that the Gates were part of archaic belief; cf. Hundt 1935, 78-81 and Brillante 1990, 43-5. Amory (1966, 14, followed by Rozokoki 2001, 4-6), leans in this direction but remains uncommitted. Kessels (1978, 101-10) and Pollmann (1993, 233) argue that the Gates were invented ad hoc for Od. 19. For a more extensive bibliography, see Desmidt 2006, 284 note 2, which Haller (2009, 398-9) supplements.
(61.) On the meaning of [TEXT NOT REPRODUCIBLE IN ASCII], see Amory 1966, 19-21.
(62.) Il. 5.887 maybe based on the word's association with ghosts in the Odyssey (Kirk 1990, 152).
(63.) Denniston 1950, 392, followed by West 1988, 244, ad 4.831.
(64.) One scholiast proposes that [TEXT NOT REPRODUCIBLE IN ASCII] here means "to report on unknown things" ([TEXT NOT REPRODUCIBLE IN ASCII] H scholion ad Od. 4.837). His implication of sycophancy is unsupported by the other instances of the idiom and still leaves [TEXT NOT REPRODUCIBLE IN ASCII] unexplained. If we construe [TEXT NOT REPRODUCIBLE IN ASCII] as "to report on," then "not known" returns the problem to one of ignorance, or, if the scholiast means "not known to Penelope," then the apparition would have already violated this maxim by talking about Telemachus.
(65.) Raskin 1985, 104-7.
(66.) Though a component of Attardo's model, Raskin (1985, 45-147) explains this particular operation in more detail.
(67.) E.g., Od. 11.43, 630-5.
(68.) See the overviews in Morreall 1983, 4-38 (and his own theory on 39-59) and MacHovec 1988, 27-105. On Homeric laughter specifically, see Hewitt 1928, 437 and Fernandez-Galiano 1992, 195.
(69.) E.g., Lefcourt 2001, 99-126.
(70.) Aristotle (Pol. 1453a30-39) calls the Odyssey's plot structure, which enables all such effects, "comedic." See Turkeltaub (forthcoming in JHS) for a recent discussion.
(71.) Hundt 1935, 65 and West 1988, 243, ad 4.795ff.
(72.) This clause cannot be temporal because Penelope continues to worry well into the dream.
(73.) On the distinction, see Kessels 1978, 174-85 and Levy 1982, 35 note 144.
(74.) The word [TEXT NOT REPRODUCIBLE IN ASCII] expresses the visual clarity of dreams (van Lieshout 1980, 18-9, 24-7, 38-9, 44 and Versnel 1987, 48). On the visuality of dreams in general, see Kessels 1978, 156-7, 194-7.
(75.) Il. 20.131; Od. 3.420, 7.201, 16.161. On [TEXT NOT REPRODUCIBLE IN ASCII] epiphanic use, see Versnel 1987, 48 and Pucci 1987, 110-1 and 1986, 21-3. The adjective [TEXT NOT REPRODUCIBLE IN ASCII] also has this force in other archaic poetry: Hesiod, Frr. 165.5 and 273.2 W.-M. and Stesichorus, Fr. 104.2 Page.
(76.) Cf. Kessels 1978, 87, 158-62. For epiphany dreams in Greco-Roman literature, see in particular Versnel 1987, 48-50 and Harris 2009, 4, 23-62.
(77.) Versnel 1987, 48 and Pucci 1987, 110-1 and 1986, 21-3.
(78.) The standard explanation of Penelopes relief also does not account for the dreams suddenness. Again, a sympathetic effect may be at work here. Beginning with Hobbes in his Treatise on Human Nature (11.13) and Kant in his The Critique of Judgment, several theorists pinpoint suddenness or surprise as a crucial element in producing laughter. See in particular Koestler 1964, 51-63 and Morreall 1983, 38-59.
(79.) See Frame 1978, esp. 20-8, 38-53. In Hesiod, Theog. 756-61, Sleep and Death, on whom the sun never shines, live together in the house of [TEXT NOT REPRODUCIBLE IN ASCII] (757) at the edge of the world near the entrance to Tartarus. The phrase [TEXT NOT REPRODUCIBLE IN ASCII] recurs at Il. 16.567 and 22.102, where it brings death.
(80.) Parvulescu 1985.
(81.) On the timing of katabaseis, see Frame 1978, 21-3.
(82.) Frame 1978, 7-27.
(83.) Attardo 1994, 109.
(84.) Attardo 1994, 138.
(85.) Cf. Louden 1995, 31-3: "distortion" wordplay places the object or person whose name is distorted "more firmly under the speakers control."
(86.) See Russo 1968.
(87.) Gruner 1997, 131-46.
(88.) See esp. Rapp 1948. Halliwell (2008, 51-109) offers the most incisive and thorough treatment; also Hewitt 1928; Levine 1982, 203-4,1983a, and 1983b; Colakis 1986; Brennan 1987; Brown 1989, 286-91; Fernandez-Galiano 1992, 195; Russo 1992, 59, 70; de Jong 2001, 440. On "superiority theories" of humor, see Gruner 1997.
(89.) O'Quin and Aranof 1981 and Lefcourt 2001, 127-40.
(90.) I would like to thank my colleagues at Santa Clara University, the editor and anonymous readers of Helios, and Pietro Pucci for his guidance and support through the years.
|Printer friendly Cite/link Email Feedback|
|Article Type:||Critical essay|
|Date:||Mar 22, 2014|
|Previous Article:||'Blessed is he, who has seen': the power of ritual viewing and ritual framing in Eleusis.|
|Next Article:||Combat trauma and psychological injury in Euripides' Medea.|