The Center for Hellenic Studies

Homer’s Versicolored Fabric: The Evocative Power of Ancient Greek Epic Word-Making

  Bonifazi, Anna. 2012. Homer's Versicolored Fabric: The Evocative Power of Ancient Greek Epic Word-making. Hellenic Studies Series 50. Washington, DC: Center for Hellenic Studies. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Bonifazi.Homers_Versicolored_Fabric.2012.

Chapter 1. The Cognitive Presence of the Absent Hero (Odyssey 1–4)

The common thread of the first three chapters of this monograph is Odysseus and anaphoric references to him, in particular, (ἐ)κεῖνος and αὐτός. Rather than being merely technical and necessary linguistic devices that recall his person, such references offer a remarkable contribution to the poetic account of the Ithacan hero in his literary substance. By “literary substance,” I mean the complex of motifs underlying Odysseus’ figure, which range from his narrative role to the interlacing dynamics of his disguise and of the journey of his soul. In this respect, the anaphoric references that I will discuss are instances of what I would call literary grammar.

The term “cognitive” in the title of this chapter recalls a general idea of cognition—that is, the mental process of acquiring knowledge and understanding. I would like to draw attention to Odysseus’ presence in books 1–4 of the Odyssey, despite his physical absence from the plot. This presence, I argue, reflects what happens at the level of cognition. While the first books of the poem are being performed or read, the recipients somehow perceive Odysseus in their consciousness. My use of “consciousness” builds upon the work of the linguist Wallace Chafe, which delves into the strong links between linguistic phenomena, perceptions of information, and states of consciousness. [1] In the course of my analysis, I will ask questions about Odysseus’ presence in the consciousness of three main agents: the primary speaking ‘I’, the external audience, and the internal audience. [2] Such an investigation advances a relatively {13|14} new perspective in the study of Homeric poetry—namely, the cognitive one; in particular, it delves into the narrative and performative technique of placing references to characters throughout the story, which is essential to the understanding and the processing of information; it is where grammar, performance, and cognition are most synergetic. [3] An examination of these references enhances our understanding of the range of uses of some words beyond the current scholarly consensus.

Though the name “Odysseus” is uttered by the primary speaking ‘I’ of the Odyssey an average of twenty-five times per book, the most interesting references to the Ithacan hero are the anaphoric ones—that is, linguistic expressions that are meant to have Odysseus as the referent. For the moment, I am taking into account a quite broad definition of “anaphora,” which is respectful of its etymology (ἀνά + φέρω = “bear back”), as well as open to its cognitive extensions: “Anaphora is a reference to objects that have already previously figured in discourse or are generally known.” [4] An example in English:

“This is not fun at all,” said Sarah. The girl was quite upset.


“The girl” is an anaphoric expression implying a referent whose verbal trace is the previously mentioned name “Sarah.” We do not know anything about her. If this text were within a literary text, it would be difficult to determine the literary values and the cognitive implications underlying the phrase “the girl.” We might, perhaps, think that the writer assumes a bit of distance after the direct quote of Sarah’s words, as if “the girl” were an external qualification of the statement, but we probably could not go much further. Conversely, in the Odyssey, there is one anaphoric expression that has been the subject of {14|15} numerous comments by numerous scholars throughout the centuries and whose literary significance has been unquestionable. It appears in the very first line of the poem:

Ἄνδρα μοι ἔννεπε, Μοῦσα, πολύτροπον, ὃς …

Odyssey 1.1

About the man, tell me, Muse, [the one] of many ways, who …


The incipit of the Odyssey is an anaphoric expression implying a referent. [5] The referent is Odysseus. This referent has no previous verbal trace at all. [6]

The withholding of Odysseus’ name at the very beginning of the poem (until line 21) has been much commented upon. I will review some of the interpretations of this linguistic fact because they show how much the name, naming, name-withholding, and anaphors exemplify several literary themes and motifs of the Odyssey. Therefore, they indirectly show that anaphoric expressions may have a poetic significance within Homeric linguistic choices.

Goldhill and Clay connect the incipit with the overall theme of Odysseus’ namelessness throughout the poem; their statements may be seen as representative of the communis opinio. Goldhill sees in andra … polutropon an “uneasy tension between paradigmatic model and unique individual typical of the representation of heroes.” In particular, in andra “there is to be recognized a paradigmatic and normative representation of what it is to be a man in society, an announcement that the narrative to come will explore the terms in which an adult male’s place is to be determined.” [7] Polutropon, conversely, identifies the special qualities of a special andra; it marks the distinctive manyness of the Ithacan hero. [8] Clay highlights Odysseus’ anonymity: “… [T]he hero as he is first introduced is anonymous. He is in no way defined by his patronymic.” She continues: “the {15|16} absence of Odysseus’ name characterizes not only the proem; for throughout a good part of his adventures, the hero is nameless.” [9]

Odysseus’ namelessness is the main topic of the first chapter of Bernard Fenik’s Studies in the Odyssey, which is even titled “The nameless stranger.” The argument starts with the question of why Odysseus in book 7 does not reveal his name to Arete as the queen wonders about his identity (7.237–239). Even though Fenik does not specifically comment on andra … polutropon at 1.1, his investigation of “the business of name-giving in the Odyssey” [10] is a useful contribution to our comprehension of the recurrent patterns of concealing Odysseus’ identity and of delaying the recognition scenes: “It would, in a word, contradict an unchanging bent of the Odyssey if the hero did answer Arete and name himself directly.” The ultimate purpose of the poet, in Fenik’s view, is to produce “an elaborate range of emotions and ironies,” the effects of which “he [the poet] concentrates on from the beginning of his poem to the end.” [11] A clear instance of these emotions and ironies relating to Odysseus as the nameless stranger is—in Fenik’s analysis—the common pattern of Eumaeus’, Eurykleia’s, and Penelope’s regret and sorrow about Odysseus’ absence, as expressed in front of Odysseus himself (Odyssey 14.40–41; 19.369, 127–129 respectively). More importantly, Fenik sees a polarization in the scholarship between those who emphasize Odysseus’ own reasons for maintaining his namelessness (“the persons of his [the poet’s] imagination must … bear within themselves the motives for their actions” [12] ) and those who think of the poet’s reasons for Odysseus’ withholding of his name (“the ‘intention’ was that of the poet” [13] ). In the first case, the focus is upon Odysseus’ state of mind—that is, his temporary loss of self-awareness—whereas in the second case, the poetic concern is with maintaining typical Odyssean situations. [14]

While the debate about Odysseus’ namelessness in Fenik’s work involves either the main character or the primary speaking ‘I’, in Norman Austin’s “Name Magic in the Odyssey,” the analysis centers rather upon the primary speaking ‘I’ and the internal audience. Austin writes of a “deliberate circumambulation around the name” and of an “unequivocal evasion.” [15] His analysis of the pattern {16|17} of withholding Odysseus’ name includes Penelope’s periphrases at 4.724–725 and 814–815 (“ … my brave husband, lionhearted, excellent in all forms of aretē among the Danaans”), the series of anaphoric expressions referring to Odysseus in the exchange between Telemachus and Athena in book one (the most prominent of which is at line 115, when Telemachus is seated among the suitors ὀσσόμενος πατέρ’ ἐσθλὸν ἐνὶ φρεσίν “seeing in his mind’s eye his brave father”), Calypso’s words at 5.138–144, and Eumaeus’ concern with the name of his master in book fourteen. [16] N. Austin favors the rhetorical explanation of periphrasis as a form of amplificatio “used for dramatic emphasis,” [17] and, most of all, he interprets Odysseus’ withheld name in the proem as a mark of fusion between the primary speaking ‘I’ and the internal audience (that is, those who practice the “circumambulation” around his name): “The Odyssey introduces Odysseus obliquely because that is the way in which sympathetic characters consistently introduce or talk about Odysseus: the andra of 5.1 is virtually a formulaic periphrasis for Odysseus through the poem … The poem aligns itself with those who treat the name of Odysseus with reverence.” Thus, Odysseus’ namelessness can be seen as a purposeful strategy having to do with the “internal” perception of the main hero, rather than with modern readers’ concerns about the vagueness and mysteriousness of indirect references; indeed, Odysseus’ namelessness might well amplify a certain reference to character rather than conceal it. My own analysis supports this latter reading, which is cognitive and literary at the same time. In the framework of the secondary literature about andra, Rüter makes a central point: “Wenn das Proömium den Namen des Odysseus nicht nennt, so ist dies keine ‘Unbestimmtheit des Ausdrucks’ und ‘Undeutlichkeit der Meinung’, der Name ist auch nicht einfach ‘vergessen’, und auch die Feststellung, daß der Held Odysseus heiße, sei beim Hörer vorausgesetzt und stehe erst beiläufig α 21, trifft die Sache nur ungenau. Jeder entnimmt den ersten vier Versen ohne Mühe, daß der Held nur Odysseus sein kann.” [18] Odysseus is simply indicated as the most obvious referent.

To sum up, an understanding of andra at 1.1 might reflect the different points of view of the modern reader, of the external audience, of the internal audience, of the main character himself, or of the primary speaking ‘I’. All seem to have good reasons to use and to appreciate andra instead of “Odysseus.” Modern readers take this as a striking omission, and their interpretation is driven by their literary comprehension of Odysseus’ identity as a leading issue of the poem. The external audience finds meaning in an oblique reference to {17|18} the main character of the story, which might well have had a specially marked quality (andra refers to shared knowledge about the social status of heroes and is followed by the adjective polutropon, traditionally characterizing Hermes [19] ); they might also have appreciated the novelty of the lack of a traditional feature, such as the patronym. Finally, in light of N. Austin’s argument, the point of view of the internal audience—more specifically, the part of it that is still loyal to Odysseus—might coincide with the practice of the primary speaking ‘I’, which is to have Odysseus and Odysseus’ name so much on their mind that they refer to him almost exclusively via periphrases and pronouns, because they know (and share) a great deal of information about him; they also venerate the master and are going to experience his progressive recognition. The main character himself, Odysseus ipse, might have fully approved such an incipit to the extent that it reflects his process of losing, disguising, changing, and finding more identities. The primary speaking ‘I’ might have considered all of the foregoing in order to engage the audience(s) as much as possible, not only by uttering those specific words, but also by hinting metanarratively at the power of words, which is a prominent theme of his own version of Odysseus’ nostos. [20]

All that has been said so far should not be kept apart from the grammatical classification of andra as an anaphoric expression with Odysseus as the referent. On the contrary, all that has been said is tightly connected to the reading of andra as an anaphoric expression. Anaphors can be more powerful and more effective than we think. A further step in my argument is that the grammatical classification of andra as an anaphoric expression with Odysseus as the referent still misses a fascinating detail: there is no verbal antecedent indicating the referent. I am arguing that this lack—which has specific cognitive reasons, as I will show in the next section—plays a relevant role in terms of the literary value of the phrase. Odysseus is the most present referent in the mind of the primary speaking ‘I’ and of the external audience, who share the event of the performance; he is most present in the minds of the internal characters as well, whose actions are basically focused on Odysseus’ physical absence. Insofar as we expand our knowledge of some grammatical phenomena, we are better able to comprehend the literary functions and meanings of what Bauman (1977) calls “verbal art.” Kühner and Gerth, Smyth, and all the scholars who many decades ago wrote about, for example, the use of ancient Greek third-person pronouns—which are anaphors par excellence—could not foresee that many years later studies in linguistics, in cognitive psychology, and in artificial intelligence would have increased our understanding of anaphora. The main goal in {18|19} what follows—which coincides with the main goal of the monograph itself—is to draw from these recent works some cues that might allow us to complete (if not to revise) the account of some Homeric linguistic features; this, in turn, will have consequences in our translations and in our teaching.

Alternative readings of third-person pronouns: accessibility of the referent, narrative functions, and pragmatic impact

Any secure investigation into the interpretation of anaphors and their discourse relevance has to exceed the scope of syntax. That is why in recent years the processing of anaphoric expressions has enjoyed increased interest in linguistics, in cognitive psychology, and in computer science. As one might imagine, the theoretical frameworks and the scholarly models are numerous and different. For the purpose of my study of archaic Greek poetry, I am selecting the findings (and the related terminology) of a restricted set of works, which concern three interrelated aspects of referential choices: the cognitive accessibility of the referent, the narrative functions of anaphors, and the attitudes of the speakers towards the entity recalled. [21]

The first aspect represents a significant innovation with respect to traditional accounts of anaphors, especially those with which classicists and philologists are acquainted. In fact, it implies a radical change in the understanding of what an anaphor is and how it is processed. It also constitutes the theoretical basis for the two remaining aspects I associate with this—that is, the discourse functions and the speaker’s standpoint.

When we say that a certain anaphoric expression has a referent, we assume that the referent is an extralinguistic entity, which can be either fictional or non-fictional. [22] For example, in the above cited text “‘This is not fun at all,’ said Sarah. The girl was quite upset,” “Sarah” is the name of such an entity. According to classical accounts of anaphors, the referent has to be mentioned in some way in the text. We understand that “the girl” is an anaphor for “Sarah,” insofar as “Sarah” is in the text. This model of explanation is called the “referent in the text” model. Reference is seen as “the relationship between a pronoun and its antecedent, this antecedent being a recent full noun phrase.” [23] Yet, as many {19|20} scholars have already queried, what happens when we process sentences that include a third-person pronoun referring to something that is not in the text? Let us consider, for example, the following example:

[Context: a neighbor’s father has been in the hospital for a week already] Anne to her neighbor, seeing her looking haggard:

How is he? [24]


In this case, “he” is a third-person pronoun the understanding of which cannot be explained according to the “referent in the text” view. In fact, the inference that “he” refers to the neighbor’s father is due to the extralinguistic context: Anne and the neighbor share a previous knowledge about the physical condition of the father and, presumably, also about the neighbor’s concern. Thus, “he” is an anaphor referring not to an entity that has been previously figured in discourse, but to an entity about which the discourse participants share some previous knowledge. I quote an even subtler example showing a pronoun whose referent is an entity present somewhere else than in the text:

Kill an active, plump chicken. Prepare it for the oven, cut it into four pieces and roast it with thyme for 1 hour. [25]


In the linguistics literature, this is a famous example. The referent of each “it” is actually a dead and no longer active chicken, instead of what is verbally mentioned as “an active, plump chicken.” Yet, “it” is presumably processed in a successful way by any reader of the recipe. What does this mean? When we process a pronoun, we retrieve the “mental representation” of the entity corresponding to the referent. For instance, the mental representation of the chicken selected for the quoted recipe allows the reader to recreate an appropriate fictional context and to monitor the chicken in its various phases of preparation. This model of explanation is called the “referent in the mind” model, or cognitive model of reference. [26] The advantages of such a model are remarkable. First, it is possible to explain why in everyday language we regularly use pronouns without previously mentioned referents (for example, A to B outside a classroom: “How was it?”), and why authors of literary or artistic texts purposefully play with the same feature (for example, think of Robert Redford’s movie “A River Runs Through It”). Second, this cognitive model allows for a much more flexible notion of antecedent; what triggers the inference about the referent can be a linguistic item or an extralinguistic item or even an implied extralinguistic {20|21} item (as in the following exchange, where the implied extralinguistic item is “parents”: A: “Did you know that little Peter recently became an orphan?” B: “Really? He must miss them terribly”). [27] Third, as Emmott and Cornish argue, such a model makes anaphora and deixis cognitively similar: in both cases, a linguistic feature may refer to the extralinguistic context without the antecedent necessarily being a verbal one [28] (as in the following question: [Context: A young goat suddenly enters the open front door. A to B, observing the scene in fascination:] “What do you think it’s looking for, exactly?” [29] ). Fourth and last, the distinction between backward (anaphoric) and forward (cataphoric) mention of the intended referent is neutralized at the cognitive level. Even pronouns referring “back” to already familiar subjects are cognitively projected forwards, to the extent that new information involving them is going to be added (think of the referent in the mind of “journey” in the following: “Mary went from Paris to Istanbul by train and coach. It took a whole week”). [30] Overall, anaphora becomes a phenomenon related to discourse memory, which encompasses both linguistic and extralinguistic features of communication. [31]

If we adopt the “referent in the mind” model, we can better appreciate the narrative strategies used to recall a certain character who has already been introduced. Instead of explaining the linguistic choices on a “referent in the text” base (for example, when we say that distal pronouns are used for referents that are mentioned far above), we can think of pronouns and of other anaphoric expressions as ways to retrieve the mental representations of the corresponding referents. This retrieval is part of an overarching cognitive activity in which every reader is involved. Whenever we read a narrative, we recreate a fictional context that includes characters and locations that we constantly monitor: we track the overt as well as the covert presence of groups and of individual characters; we monitor them in relation to the different contexts to which they are bound. [32] In this respect, the mental representations of the participants in the {21|22} narrated events are essential: they crucially help us to process the pronouns related to them and allow us to explain their behavior. [33] The text we read may draw our attention to one or more characters in a certain contextual frame, and this may result in the use of different linguistic features that recall those characters or that activate them in our consciousness. [34] In other words, different characters acting in different contextual frameworks are recalled in different ways according to the different focus of attention that is drawn to them. In the reading process, the focus of attention is constantly updated, so that everyone can “fine-tune their perception of what the predication currently being constructed … or interpreted … is about.” [35] From Cornish’s study of the linguistic rendering—in English and in French—of anaphors according to the different activation degrees of the referents, I report the following scenarios:

First possible situation: “the referent is assumed to be at the forefront of the addressee’s (and the speaker’s) consciousness”; in other words, the referent is highly accessible information. [36] There are no competing referents. An unaccented pronoun or null anaphor [37] will be used.

Second possible situation: “the intended referent is not in focus, but is both familiar to the addressee at that point in the discourse and ‘activated’”; the accessibility of the referent is medium. An accented third-person pronoun [38] or a demonstrative pronoun will be used.

Third possible situation: “the referent is assumed to be inactive, that is, of low accessibility”; it may be “located in long-term memory,” as an {22|23} item of “encyclopedic, stereotypical, or personal knowledge.” A distal demonstrative pronoun or full definite description or a full name will be used. [39]


I have now introduced some background notions that inform my approach to the analysis of the occurrences of third-person pronouns in Homer. First, pronouns are not simply considered as syntactical devices used to recall a previously mentioned entity; rather, they activate the mental representation of an entity. The verbal reference to this entity may or may not appear in the text. This model for the interpretation of anaphors is the “referent in the mind” model. The cognitive process underlying the use of pronouns is forward-oriented: recall adds new information about the referent in question. Different pronominal choices may correspond to different activation degrees of a certain character at a certain point of the narration: null anaphors and unaccented pronouns may signal that the referent is already active or easily accessible; accented or demonstrative pronouns, conversely, may signal that the referent is going to be re-activated; finally, distal demonstrative third-person pronouns and definite nouns may signal that the referent has a very low accessibility.

Cornish presents a general framework within which further discourse-related and pragmatic considerations can be added. As we shift our focus from the cognitive processing of undifferentiated sequences of sentences to the comprehension of sequential units of discourse, we come to consider the overall organization of discourse and the “global status of characters.” [40] The speaker’s use of referential devices rests on both the “local context of speaking” and the structural properties of the discourse. [41] Within narrative discourse, for example, entities are referred to not only as “the immediate informational context permits” but also under specific “narrative conditions”—that is, the centrality of some events, the relation between embedded episodes and the main storyline, the role of some characters. [42] In other words, the distance to the last mention of the referent, the need to disambiguate between multiple referents, and the availability of the referent in a local informational context are insufficient to account for the actual use of anaphoric expressions in cohesive stretches of discourse—that is, in texts. [43] Clancy (1980) identifies episode boundaries as structural elements that considerably influence referential choices. By taking {23|24} into account the wide range of possible referential choices from null anaphor to full noun phrases, Fox (1987) argues that the narrative use of full noun phrases when a pronoun might have been expected signals “the hierarchical structure of the text,” for instance, it marks the beginning of a new narrative unit. [44]

Amongst scholars working on participant tracking in Latin literature from a pragmatic perspective, Bolkestein and Van de Grift (1994) suggest analogous interpretations. Less expected anaphoric choices are connected to discourse segment boundaries or, more generally, to developments of discourse that are accomplished in what the authors call “a ‘non-default’ way”: for instance, when a “topic-to-be” is introduced or when a switch at the level of perspective or of type of discourse occurs. [45] By examining the pragmatics of referential choices in samples from Virgil, Ovid, and Tacitus, Kroon (2009) provides a framework that permits the identification of the discourse switches that motivate unexpected occurrences of ille. This framework is itself a typological division of different “modes of discourse”: narrative, report, description, information, argument and direct discourse. [46] All the works analyzed by Kroon show that the interpretation of referential devices in cohesive units of discourse should include not only the accessibility degree of the referent but also the narrative functions fulfilled by the strategic location of the device at particular points of the discourse.

A further—and final—aspect, exquisitely pragmatic in nature, should be underscored in the analysis of the discourse relevance of referential expressions: the attitudes of the speakers towards the entity recalled. In other words, the utterance of an anaphoric expression might be determined by its specific communicative purposes, which include the interaction between the speaker and the addressee(s) or between the narrator and the narratees. The various attitudes to which I will call particular attention are social ranking, emotional appraisal, and visual standpoint. In Bonifazi 2009b, I grouped these features under the label “pragmatic conspicuity.” As for Latin anaphors, Kroon (2009) once again offers insights related to this topic: remoteness and proximity of anaphoric demonstratives may relate to the psychological level of distance or nearness; anaphors may convey the perspective from which a certain referent {24|25} is “seen” by the author or the speaker. [47] Before exploring these claims through an analysis of some Homeric passages, I will set forth two basic notions that inform my own view of some usages of κεῖνος and αὐτός. The first notion is that of “emotional deixis,” while the second is that of “empathic anaphor.” Lakoff 1974 highlights some English uses of the demonstratives “this” and “that” that are “linked to the speaker’s emotional involvement in the subject-matter of his utterance.” [48] They create vividness sometimes because they convey the speaker’s own closeness to the subject-matter, sometimes because they imply closeness between speaker and addressee—sympathy, solidarity, or camaraderie. Lakoff calls these instances of “emotional deixis.” Interestingly, the emotive uses of “that” are found to be surprisingly close to those of “this.” Since the current chapter will delve into the Homeric uses of κεῖνος, I quote here one of Lakoff’s examples of emotional-deictic “that”:

That Henry Kissinger sure knows his way around Hollywood!


While marking spatial distance, “that” actually “establishes solidarity and implies shared emotions.” Through the mention of a common subject, a link between speaker and addressee emerges, forged by culturally shared knowledge or opinion. [49]

The notion of “empathic anaphor,” as formulated by Daneš and by Conte, will also be useful in my present investigation. Daneš states that in modern literary prose co-referentiality “is not based on the vocabulary and/or on some conceptual system,” nor is it “derivable/predictable/recoverable from preceding co-text.” Rather, co-referential expressions “bring forth new characteristics of the respective discourse subjects and contribute to text development.” [50] Pragmatic factors such as evaluation, emotion, attitudes, or point of view may play a role in the stylistic choices of anaphors. Conte supports the same point; empathy by the speaker is a relevant component of pronominal choice. Conte includes the following excerpt from James’ The Bostonians in her analysis:

To talk to those people about the South—if they could have guessed how little he cared to do it! He had a passionate tenderness for his own country, and a sense of intimate connexion with it which would have made it as impossible for him to take a roomful of Northern fanatics into his confidence as to read aloud his mother’s or his mistress’s letters. To be quiet about the Southern land, not to touch her with vulgar hands, to {25|26} leave her alone with her wounds and her memories, not prating in the market-place either for her troubles or her hopes, but waiting as a man should wait, for the slow process, the sensible beneficence, of time—this was the desire of Ransom’s heart.

H. James The Bostonians (1886). Harmondsworth: Penguin Books, 1966, p. 44; cf. Conte 1999:78 (my italics)


The novel’s protagonist, Basil Ransom, has been asked to talk about his life in the South of the United States at a meeting of feminists in Boston. The author’s adoption of Ransom’s point of view becomes evident in the anaphors: the choice of “her” for “the Southern land” (following an initial “it”) conveys empathy and emotional nearness by the subject toward the referent.

Case-studies from Homer

In order to illustrate the cognitive processing, the interpretation, and the interrelation of anaphoric devices, I have chosen three Homeric passages as study cases. My aim is to show that all of the pragmatic features that I have just introduced can be found in the Homeric poems. Furthermore, the complexity of the interrelation between these features is closely linked to the literary side of the reading, in terms of participant tracking, participant “hierarchy,” discourse structure, narrator’s empathy, and characters’ empathy. My first passage shows how a referent is made cognitively accessible through its mental representation:

Ἦμος δ’ ἠριγένεια φάνη ῥοδοδάκτυλος Ἠώς,
ὤρνυτ’ ἄρ’ ἐξ εὐνῆφιν Ὀδυσσῆος φίλος υἱός,
εἵματα ἑσσάμενος, Ø περὶ δὲ ξίφος ὀξὺ θέτ’ ὤμῳ,
Ø ποσσὶ δ’ ὑπὸ λιπαροῖσιν ἐδήσατο καλὰ πέδιλα,
Ø βῆ δ’ ἴμεν ἐκ θαλάμοιο θεῷ ἐναλίγκιος ἄντην.
Ø αἶψα δὲ κηρύκεσσι λιγυφθόγγοισι κέλευσε
κηρύσσειν ἀγορήνδε κάρη κομόωντας Ἀχαιούς.
οἱ μὲν ἐκήρυσσον, τοὶ δ’ ἠγείροντο μάλ’ ὦκα.
αὐτὰρ ἐπεί Ø ῥ’ ἤγερθεν ὁμηγερέες τ’ ἐγένοντο,
Ø βῆ ῥ’ ἴμεν εἰς ἀγορήν, Ø παλάμῃ δ’ ἔχε χάλκεον ἔγχος

Odyssey 2.1–10; the primary speaking ‘I’ to the audience  [51] {26|27}

Odysseus’ dear son, when early-born Dawn, rosy-toed, appeared,
woke up from his bed;
after putting his clothes on, he carried his sharp sword over his shoulder,
under his sleek feet he bound fair sandals,
and went out from the room, comparable to a god, in the face.
He directly gave instructions to the loud-voiced heralds
that they might announce the assembly to the long-haired Achaeans.
The heralds made the announcement, and the men gathered in a quite short time.
Then, once they were assembled all together,
he went to the place of meeting, with a bronze spear in the hand


Since the end of book one (cf. 425–444), Telemachus is a character in focus; that is, he is at the center of attention of the primary speaking ‘I’ and of the recipients as well. At line 2 of book two, he is recalled as Ὀδυσσῆος φίλος υἱός (“the dear son of Odysseus”). The following description of his actions implies that Telemachus is a highly accessible referent; indeed, he is the only possible referent. Accordingly, at lines 3–7 a series of 4 null anaphors occur (the symbol Ø conventionally stands for null anaphor, or zero). [52] At lines 6–7, he is said to call some heralds to convene the assembly of the Achaeans. οἱ μὲν and τοὶ δ’ (8) activate in the performer’s and in the audience’s mind the easily accessible entities corresponding to the heralds and Achaeans respectively, who have just been mentioned. At lines 9–10, the heralds, the Achaeans, and Telemachus are recalled via null anaphors. This means that at that point the mental representation of all these referents in the discourse memory of the performance participants (and of us readers, as well) is fully retrieved; all these characters are “on the stage” and activated.

My second and the third case-studies will show possible narrative functions of anaphors and possible attitudes of the speakers towards the entity recalled.

Let us first consider the proem of the Odyssey, lines 1–10. Our goal is to check the referential choices against their discourse relevance and the “global status of characters.” The source of the utterances is the primary speaking ‘I’. The subject of the analysis is the set of characters that are introduced. [53] When we examine the words of our examples, it is important to leave aside syntactical distinctions, such as the grammatical case in which a certain character is introduced or recalled; accusative, genitive, and dative cases are as relevant as {27|28} the nominative. Analogously, the type of clause including the reference is not relevant. The results show that the primary speaking ‘I’ not only provides a map to the people involved, but he also cues the recipients to a hierarchy of each character’s discourse relevance. Here is the text:

Ἄνδρα μοι ἔννεπε, Μοῦσα, πολύτροπον, ὃς μάλα πολλὰ
πλάγχθη, ἐπεὶ Τροίης ἱερὸν πτολίεθρον Øἔπερσε·
πολλῶν δ’ ἀνθρώπων Ø ἴδεν ἄστεα καὶ νόον Ø ἔγνω,
πολλὰ δ’ ὅ γ’ ἐν πόντῳ πάθεν ἄλγεα ὃν κατὰ θυμόν,
ἀρνύμενος ἥν τε ψυχὴν καὶ νόστον ἑταίρων.
ἀλλ’ οὐδ’ ὧς ἑτάρους Ø ἐρρύσατο, ἱέμενός περ·
αὐτῶν γὰρ σφετέρῃσιν ἀτασθαλίῃσιν Ø ὄλοντο,
νήπιοι, κατὰ βοῦς Ὑπερίονος Ἠελίοιο
Ø ἤσθιον· αὐτὰρ ὁ τοῖσιν ἀφείλετο νόστιμον ἦμαρ.
τῶν ἁμόθεν γε, θεά, θύγατερ Διός, εἰπὲ καὶ ἡμῖν.

Odyssey 1.1–10; the primary speaking ‘I’ to the audience

About the man, tell me, Muse, [the one] of many ways, who many places
had been driven, after he sacked the sacred city of Troy.
He saw the cities of many human beings, and learned about their noos.
He suffered many pains in his spirit on the sea,
struggling for his own life and for the nostos of his companions;
but he did not save the companions, even though he strove to;
for they ruined themselves because of their own recklessness,
fools, who ate the cattle of the Sun god;
and then, he [the Sun god] took away from them the day of nostos.
About these things, from some point of this tale, tell us as well, goddess, daughter of Zeus.


First- and second-person references stand for individuals who are external to the map of the characters involved in the story. They are:

• line 1 “I” (μοι)
• line 1 “you” (Μοῦσα)
• line 10 “you” (θεά, θύγατερ Διός)
• line 10 “we” (ἡμῖν)


At this point in my analysis, I will not engage the question of who is physically and symbolically behind these references; rather, I simply underscore {28|29} a well-known fact about them: they encapsulate the proem by starting and ending it according to a chiastic structure (first – second – second – first person). Once we set aside the personae involved in the performance, readers process any references to characters (and possibly to objects as well) as part of the contextual frames that are recreated. [54] People such as “many human beings” (πολλῶν … ἀνθρώπων, 3) exist within the story-world but they are not known to act in any particular context at any particular point; technically, they are said to be “unbound.” [55] Instead, other people mentioned are involved in the actions and events described, which constitute the contextual frames recreated by the reader: traveling, suffering, being destroyed by recklessness, and taking away the nostimos day. The participants in these events are the only ones recalled by means of pronouns in the text; they are Odysseus, the companions, and the Sun God. The only exception is the pronoun τῶν (10), which resumes all the mentioned topics. [56] In the following table, I summarize the verbal traces of the people involved in the sequence of the narrative flow. [57] The verbal traces in question are anaphoric expressions, including also relative pronouns and possessive adjectives. This inclusion is due to some specific morpho-syntactic characteristics of the Homeric language. The meanings and the functions of ὁ/τό, ἑ, ἕ (σφε- for the plural) and ὅς—coming from IE roots *so-/to-, *IE *yo-, and IE *swe– respectively—in Homeric epic and in archaic lyric are not easy to distinguish: the weak demonstrative pronouns ὁ/τό work also as definite articles, as simple third-person pronouns, and as correlative pronouns; ἕ forms oscillate between purely anaphoric and reflexive functions; ὅς forms are both relative and possessive. [58] In the analysis that I am proposing, they all contribute to mark the local informational context concerning the characters as well as their discourse relevance over the entire poem. {29|30}












































































































Anaphoric Expression Referents Type of Anaphor
(1) Ἄνδρα (… πολύτροπον) Odysseus noun
(1) ὅς Odysseus Relative pronoun
(2) Ø (ἔπερσε) Odysseus Null anaphor
(3) Ø (ἴδεν) Odysseus Null anaphor
(3) Ø (ἔγνω) Odysseus Null anaphor
(4) ὁ Odysseus Weak demonstrative pronoun
(4) ὃν (…θυμόν) Odysseus Possessive adjective
(5) ἥν (…ψυχήν) Odysseus Possessive adjective
(6) (οὐδ’) … Ø (ἐρρύσατο,) Odysseus Null anaphor
(7) αὐτῶν The companions Personal pronoun
(7) σφετέρῃσιν The companions Possessive adjective
(7) Ø (ὄλοντο) The companions Null anaphor
(9) Ø (ἤσθιον) The companions Null anaphor
(9) ὁ The Sun god Weak demonstrative pronoun
(9) τοῖσιν The companions Weak demonstrative pronoun
(10) τῶν The topics just mentioned Weak demonstrative pronoun

I have already suggested that ἄνδρα (1) may be considered an anaphoric expression without any verbal antecedent. From a cognitive standpoint, listeners and performers do not need to name Odysseus, as he represents a most accessible entity (I remind the reader of the case “How is he?”). The relative pronoun ὅς triggers an entire chain of cognitive activities that I have explained in a different work: [59] its utterance coincides with the beginning of the reenactment of the past, and its function is to trigger the mental representation of properties, genealogies, and famous deeds of mythical characters. As such, it draws the attention of the recipient to the person about whom a (long) story flashback is going to start. Then, a little series of null anaphors follows (at lines 2 and 3, verbs ἔπερσε, ἴδεν, ἔγνω have no grammatical subject), which, according to Cornish’s scenario, reflects an unchanged status in focus and no competing {30|31} referents. At line 4, the weak demonstrative pronoun ὁ [60] does not serve to reset the focus upon Odysseus, because he is already in focus and, so far, there are no competing referents. The cognitive and narrative reason for choosing ὁ registers the combined functions of the particle δέ, which opens a new discourse act, [61] and of γε, which modifies ὁ itself. γε gives prosodic and semantic prominence to ὁ (as it typically does, along with personal pronouns); [62] it emphasizes something relationally new about somebody referentially old. [63] The relationally new component is something that only recipients sharing a large amount of contextual knowledge with the speaker can grasp; we modern readers can simply observe that the new discourse act of line 4 zooms in on a closer description of what Odysseus struggled for and took care of in his existence (it also concludes the series of alliterative π’s of the first 4 lines of the poem). Thus, δ’ ὅ γ’ not only introduces a single different discourse act, but also marks a new discourse unit. The idea of a closer narration of Odysseus’ life is confirmed by the two possessive adjectives ὅν (θυμόν, 4) and ἥν (ψυχήν, 5). Later in the text, before Odysseus’ name is uttered for the first time (line 21), two more possessive adjectives referring to Odysseus are added: οἷσι (φίλοισι, 19) and ἥν (γαῖαν, {31|32} 21). What does that mean in our cognitive-pragmatic analysis of third-person pronouns? The use of reflexive marks (“his own”) is a sign of the speaker’s emotional nearness to the character, as if the character’s standpoint (if not his/her mental state) were accessible to the speaker. [64] In fact, all of the four reflexive marks just mentioned (lines 4, 5, 19, and 21) relate to what is deeply near and dear to Odysseus throughout his nostos: “his own” thumos, “his own” psykhē, “his own” philoi, “his own” gaia. Interestingly, the only other subjects for whom a reflexive mark is used in the proem are also the only other referents of null anaphors—namely, the companions: “because of their own recklessness” corresponds to the Greek possessive adjective σφετέρῃσιν + intensifier αὐτῶν (7); [65] “they perished” and “they ate [the Sun cattle]” in Greek are verbs without grammatical subject (ὄλοντο, 7, and ἤσθιον, 9.) Unlike the instance of Odysseus in the proem, the verbal trace of the referents is present (ἑταίρων, 5 and ἑτάρους, 6). Like Odysseus, the companions are also fully activated entities and not of emotional indifference to the speaker (see the exclamation-comment νήπιοι, 8). The next third-person pronouns are ὁ τοῖσιν (9), the referents being the Sun God and the companions, respectively. As is typical of Homeric information structure, two side-by-side pronouns heading a sentence retrieve the mental representation of easily inferable referents, about whom something relevant is going to be said. The Sun God has been introduced by means of the epithet + name phrase Ὑπερίονος Ἠελίοιο at line 8. The two entities appear to be linked: “What about him and them next? He took away the day of [their] nostos,” which is a crucial piece of information about the denied nostos of some heroes traveling back from Troy. [66] Finally, τῶν (10), which refers “backward” to all that has been mentioned in the previous lines, cognitively projects “forward” the reference to the thematic material from which the singer is going to draw. [67] {32|33}

On the whole, the use of anaphoric expressions and of pronouns in this segment of text is meaningfully related to the poetic substance of the text. The verbal traces of the referents contribute in a quite significant way to the identification of a cognitive, emotional, and narrative hierarchy of the poem’s characters. Listeners and readers are driven by the primary speaking ‘I’ to retrieve the mental representation of Odysseus and of his companions as the central characters of the poem, along with the Sun God, who is the reason for the denied nostos for the companions. The cognitive emphasis on these two entities and on the Sun God parallels the thematic emphasis on nostos, which culminates in the contrast between those who ultimately attain a nostos and those who do not.

This monograph does not consider analytically the cognitive processing of Homeric names. However, the name + epithet phrase Ὑπερίονος Ἠελίοιο (8) prompts a point about the relationship between the cognitive relevance of pronouns and that of names, names + epithets, and nouns + epithets. Bakker links the different choices amongst them to the status of the character in the listener’s consciousness, which is very relevant to the current topic. The different possibilities he identifies are ultimately close to Cornish’s outline: the more a certain character is inactive or far from the “stage,” the more names and nouns + epithets are required in their introduction. [68] Bakker also identifies a discourse function underlying the Homeric use of names: when marked by δέ or αὐτάρ, they signify a “new step in the progression of the narrative, a step signaling that the god or hero in question will be the frame, or theme, for the moment or moments to come.” [69] Likewise, I would argue that names in Homer can cognitively perform the anaphoric function of pronouns. While the hypothesis of names as anaphors seems to be new to the linguistics literature, [70] the strongly evocative power (and even magic) of names, especially “speaking names,” is not at all new to Homerists. [71] Any utterance of names, nouns + epithets, or names + epithets, independently of its more or less close relation with the contextual content, works as what is called, in computer science, a hyperlink: it opens a cognitive window on the mythical background and ency- {33|34} clopedic knowledge that do, in fact, coincide with a full mental representation of the referent by the primary speaking ‘I’ and the recipients: his (her) relatives, main deeds, physical and moral characteristics, fame, worship, and so on. [72] With the utterance of a single name, multiple chains of association are established in the mind of the receiver. Such associations can also be created by the utterance of names in secular and colloquial contexts; I invite the reader to think of an academic meeting in which a new colleague shakes hands with a senior faculty member who says: “[first and last name], nice to meet you.” The particular reality of academic meetings makes the utterance of one’s first and last name signify much more than the transmission of new information, and both the new and the senior colleagues are aware of that, thanks to their pragmatic competence. [73] To return to Homer, the first mention of Odysseus’ name in the Odyssey, a few lines after the proem, illustrates the anaphoric function of names within the “referent in the mind” model. Poseidon (Ποσειδάωνος, 20) is mentioned as the only god who does not favor Odysseus’ nostos. The primary speaking ‘I’ explains: ὁ δ’ ἀσπερχὲς μενέαινεν / ἀντιθέῳ Ὀδυσῆϊ πάρος ἣν γαῖαν ἱκέσθαι (“he remained relentlessly angry / with godlike Odysseus, until his return to his own country,” 20–21 [tr. Lattimore 1967]). The oppositional force of Poseidon and the antagonism between the two (cf. ἀντιθέῳ) is indirectly linked with the passive and the active hate that mark Odysseus, symbolically and etymologically. [74] Thus, uttering “Odysseus” strongly evokes the mental representation of the referent along with a series of specific connotations.

The subject of my third analysis is the use of different third-person pronouns that refer to the same character. My purpose is to shed some light on anaphors as potential signals of the speaker’s social and/or emotional viewpoint. Helen’s speech to Aphrodite in Iliad 3 is my example. Alexander has just survived Menelaus’ onslaught during their duel. Aphrodite, who rescued him, has placed him in his bedroom and has made him look like a most charming man (379–382). Then, the goddess reaches Helen and invites her to join Alexander. Helen recognizes Aphrodite and says:

δαιμονίη, τί με ταῦτα λιλαίεαι ἠπεροπεύειν;
ἦ πῄ με προτέρω πολίων εὖ ναιομενάων
ἄξεις, ἢ Φρυγίης ἢ Μῃονίης ἐρατεινῆς,
εἴ τίς τοι καὶ κεῖθι φίλος μερόπων ἀνθρώπων· {34|35}
οὕνεκα δὴ νῦν δῖον Ἀλέξανδρον Μενέλαος
νικήσας ἐθέλει στυγερὴν ἐμὲ οἴκαδ’ ἄγεσθαι,
τοὔνεκα δὴ νῦν δεῦρο δολοφρονέουσα παρέστης;
ἧσο παρ’ αὐτὸν ἰοῦσα, θεῶν δ’ ἀπόεικε κελεύθου,
μηδ’ ἔτι σοῖσι πόδεσσιν ὑποστρέψειας Ὄλυμπον,
ἀλλ’ αἰεὶ περὶ κεῖνον ὀΐζυε καί ἑ φύλασσε,
εἰς ὅ κέ σ’ ἢ ἄλοχον Ø ποιήσεται ἢ ὅ γε δούλην.
κεῖσε δ’ ἐγὼν οὐκ εἶμι· νεμεσσητὸν δέ κεν εἴη·
κείνου πορσανέουσα λέχος· Τρῳαὶ δέ μ’ ὀπίσσω
πᾶσαι μωμήσονται· ἔχω δ’ ἄχε’ ἄκριτα θυμῷ.

Iliad 3.399–412; Helen to Aphrodite

O Goddess, why are you eager to beguile me with these words?
Will you lead me further somewhere among the well-peopled cities,
either in Phrygia or in lovely Maeonia,
if someone is dear to you even there, among human beings?
Is it because now Menelaus, after defeating noble Alexander
wants to carry me home, even if I am hateful?
Or is it because now you stand beside me, here, and you are meditating deceits?
Sit beside him after reaching him; leave the path of gods;
don’t turn any more to Olympus with your feet,
but lament all the time over that one, and take care of him,
until he makes you his wife, or he will make you his concubine.
Over there I don’t want to go. It would arouse nemesis,
if I were to share the bed of that one. The Trojan women will all
blame me afterwards. And I already have confused sorrows in my heart.


Helen’s speech mixes wrath, bitterness, self-pity, and passion; she moves from a “startling and insulting development of the sexual theme” to “short and staccato, perhaps almost sobbing assertions of refusal, shame and self-pity.” [75] According to Ebbott, Helen’s feelings of shame, nemesis, and self-blame are to be specifically connected to her acts of mourning (cf. Iliad 24.761–775) and inform her character throughout the poem. [76] What is relevant here is how Alexander is recalled by Helen. After putting the reference to herself in the middle of the line, that is, in a position perfectly corresponding to the mention of Alexander’s name in the preceding line, as Kirk notices (cf. δῖον Ἀλέξανδρον, 403 and στυγερὴν ἐμέ, {35|36} 404), [77] Helen recalls her beloved five times in six lines by means of four different pronouns: αὐτὸν, 406; κεῖνον, 408; ἑ, 408; ὅ, 409; and κείνου, 411 (εἰς ὅ, 409, is a subordinating relative construction including neuter ὅ). Independently of the euphonic and metrical convenience of putting αὐτός (more often) and κεῖνος (less frequently) after prepositions such as παρά, περί, and ἀμφί, the choice of αὐτόν at 406 has a significant communicative value. It is neither a mere mechanical device employed to refer to Alexander, nor is it a necessary one: Helen does not seem to be interested in disambiguating between Menelaus and Alexander in “Go and sit beside him”; moreover, she does not need to add any nuance concerning his sameness or his selfness (as αὐτός may convey in Homer, even in non-nominative cases). Though I will discuss the Homeric uses and meanings of αὐτός more extensively in chapter 3, I outline here a pragmatic and cognitive reading of αὐτόν in this passage. The pronoun conveys the centrality of the referent (Alexander) at the thematic, at the visual, and at the emotional level. On the thematic level, Helen from this point on treats Alexander as the central subject for whom different actions are imagined (going to him, suffering over him, taking care of him). The visual aspect places Alexander at the center of attention; what follows is acted out around him, in both Helen’s imagination and Aphrodite’s behavior. The emotional aspect, finally, reflects Helen’s sardonic depiction of Aphrodite’s supposed concern and deep passion for the man. κεῖνον at 408 should not, then, be categorized as a variant of a “plain” third-person pronoun with Alexander as the referent; [78] rather, it is the most appropriate anaphor within the context of the described action (suffering for/lamenting about Alexander by Aphrodite). As we will see in later sections of this chapter, κεῖνος is frequently used to imply mourning and lament over a dead (or a supposedly dead) person. In the quoted passage, the imagined lament by Aphrodite overlaps with Helen’s own lament (cf. 412 “I already have confused sorrows in my heart”). [79] In addition, κεῖνον at 408 lexically cross-references the preceding κεῖνος uttered by Aphrodite—cf. 391 κεῖνος ὅ γ’ ἐν θαλάμῳ καὶ δινωτοῖσι λέχεσσι—where it deictically points at the appearance of Alexander (“Look at him! There he is, in the room with the bed of circled patterns”) and to the next κεῖνος (411), through which Helen seems to be determined to keep herself away from Alexander, both physically and emotionally (“I am not going to serve the bed of that man”). Still, despite the lexical repetition, each κεῖνος corresponds to different communicative intentions and different meanings, not only when it is uttered by different speakers (Helen or Aphrodite), but also {36|37} when it is uttered by the same speaker (Helen reporting Aphrodite’s lament and Helen expressing her own distance). The pronominal choice κεῖνος has, therefore, many facets—more, perhaps, than one might expect. I anticipate here its most striking pragmatic characteristic: κεῖνος in the Homeric poems is regularly uttered by characters; the primary speaking ‘I’ almost never uses κεῖνος. ἑ (408) is the accusative form of a pronoun that stems from IE *se-/swe- and, from Aristarchus on, has been considered a “regular” anaphoric pronoun (unlike accented ἕ, which has been considered a reflexive pronoun). [80] From the same IE root, we have dative οἷ, genitive οὗ (or ἕο, εἷο, ἕθεν), and the possessive adjective ὅς. As an unaccented pronoun not accompanied by any particle, ἑ marks a highly accessible referent and corresponds to the in-focus status of the referent; there is no competing referent and there is no switch in the imagined scene. [81] Conversely, after the null anaphor that precedes the verb ποιήσεται, ὁ at 409 has a different value. The referent is easily accessible as well, but requires a slightly higher level of activation in the consciousness of the listener, as is evident in the presence of the particle γε, whose pragmatic force is to mark a switch in the imagined scene. ὁ enhances the focus on further actions to come of an already known referent; not differently from relative pronouns, its basic cognitive function is to project forwards the relevance of the referent. As I have argued elsewhere about Pindaric language, the weak demonstrative pronoun ὁ in archaic poetry can have a recognitional function; the same holds for ὁ definite articles and ὅς relative pronoun. “Recognitional” indicates that a specific knowledge about the referent—already shared by the speaker and the listener due to some common past experience—is reactivated when something new is going to be told concerning him (or her). [82] The particle γε might contribute to this process by stressing the paradoxical novelty introduced by the ongoing discourse act—that is, “he even might make you your slave.” With respect to Cornish’s cognitive model of interpretation of anaphors, null anaphors reflect the in-focus status of the referent, with no signal of a switch of discourse level. The weak demonstrative pronoun ὁ/τό, however, while also reflecting the in-focus status of the {37|38} referent, may signal a switch at the level of discourse, as a new discourse act is going to involve its referent. Different particles accompanying the anaphor can mark this switch and can specify what kind of new act is going to be performed. Finally, the richness of the third-person pronouns chosen by Helen—and by the primary speaking ‘I’ as he reports her words—to refer to Alexander marks a variety of intentions. All of them are connected to the emotional attitude of either Helen or Aphrodite towards the referent. The complexity of these references matches the passionate and conflicted tone of the speech.

Grammatical accounts of (e)keinos

Ancient Greek demonstrative pronouns can be understood as a way to recall characters (much less frequently objects) whose “presence” is reactivated because something relevant about them is going to be said. Scholarly attention to ὅδε and οὗτος pronouns has usually focused upon the borderline between the merely anaphoric (or cataphoric) and the deictic functions. [83] Conversely, attention to (ἐ)κεῖνος and to αὐτός is, generally speaking, limited to their anaphoric function, as they mainly work as third-person pronouns. Their so-called “emphatic” use is linked to remoteness for the former and to either sameness of selfness for the latter. This section of my study retrieves some basic assumptions about the meaning and the use of (ἐ)κεῖνος, as a preliminary to an analysis of Odysseus pronominalized as κεῖνος in the first books of the Odyssey. Odysseus αὐτός, as well as αὐτός in general, will be considered in chapter 2 and chapter 3 respectively.

Havers (1906) identifies three basic meanings of (ἐ)κεῖνος—namely “that” (deictically referring to what is far in space or in time), “the other” (i.e. what is on the other side, such as enemies or dead people), and “he/she/it” (regular third-person pronoun). [84] Its fundamental characterization as a demonstrative of the third person agrees with Brugmann’s tripartition between Ich-Deixis (cf. ὅδε), Du-Deixis (cf. οὗτος), and Jener-Deixis (cf. (ἐ)κεῖνος). [85] The referents of (ἐ)κεῖνος typically do not participate in the utterance situation (they are neither the speaking “I” nor the co-present “you”). [86] Remoteness related to the past suggests the idea of “well-known” (cf. Latin ille); in the positive sense, {38|39} it expresses venerability. [87] Remoteness concerning the co-text of (ἐ)κεῖνος [88] indicates that the antecedent occurs at some distance from the pronoun itself. Literature confirms, but also contradicts, this. LSJ explains: “when οὗτος and ἐκεῖνος refer to two things before mentioned, ἐκεῖνος prop. belongs to the more remote, in time, place or thought … ; but ἐκεῖνος sts. [sometimes] = the latter.” [89] Analogously, White and Morgan write, with respect to Xenophon’s prose: “[ἐκεῖνος] used prop. of a person or thing remote in thought or actual distance from the speaker or subject in hand, but sometimes refers to one near at hand or lately mentioned, and even to the subject itself … ; it is often used as a strong form of the pers. pron.” [90] The reference to what is presumably not involved in the utterance situation is refuted by some recorded uses of ἐκεῖνος “with simple demonstrative force” (LSJ quotes Odyssey 18.239–240 Ἶρος ἐκεῖνος … / ἧσται, which is translated “Iros sits there” and Thucydides 1.51.2 νῆες ἐκεῖναι ἐπιπλέουσιν, which is translated “There are ships sailing up”). [91] To sum up, literature confirms uses of (ἐ)κεῖνος as a deictic as well as a marker of anaphoric distance, but this distance is also said to be blurred. The basic reference seems to be to individuals excluded from the communicative situation (third persons), though from time to time (ἐ)κεῖνος marks somebody or something that does appear to the sight of the speaker. Finally, some interactional components of the choice of (ἐ)κεῖνος with respect to other third-person pronouns are hinted at, such as the idea of being well-known. With the latter in mind, I point out a couple of further important components related to (ἐ)κεῖνος that have already been considered by scholars. The first, again from LSJ, is about the use of (ἐ)κεῖνος “for things, of which one cannot remember or must not mention the name.” [92] The second is suggested by Bakker in his comment on Herodotean indirect speeches: the standpoint of the speaker whose words are reported by the historian seems to be retained through the use of αὐτός and ἐκεῖνος, which distinguish between the speaking “I” (referred to as αὐτός) and a co-present “you” (referred to as ἐκεῖνος). [93] ἐκεῖνος may work as the mark of the {39|40} addressee in indirect speeches. The further interactional components I would identify concern the social convenience of choosing (ἐ)κεῖνος instead of other pronouns, on the one hand, and the pragmatic association of (ἐ)κεῖνος-subjects with ‘non-I’ subjects, on the other. As far as the meanings of (ἐ)κεῖνος in Homer are concerned, I cite here a representative statement: “ἐκεῖνος, ou κεῖνος , ne sert jamais à désigner une personne ou une chose aperçues au loin. Il s’applique le plus souvent à une personne absente, à une chose placée hors de la vue.” [94]

The following analysis questions the understanding of (ἐ)κεῖνος as a label indicating somebody (less frequently something) far in space or in time or, as Monro says, “in the order of thought.” [95]

The significance of keinos referring to Odysseus in the first four books of the Odyssey

The first four books of the Odyssey depict the negative situation in Ithaca due to Odysseus’ absence, even as they also prepare the eventual narration of Odysseus’ nostos by reference to tales of other Greek heroes’ nostoi, which parallel Odysseus’ nostos, either in similarity or contrast. [96] Thalmann, who titles a chapter of his monograph on the Odyssey “The Hero Absent,” emphasizes both of these aspects and concludes that they ultimately point to “the triumph of Odysseus’ homecoming.” [97] The absence of the main hero leaves space for telling the stories of other people and other heroes, which ultimately serves to foreground the specific adventures and peculiar story of Odysseus. This latter point bridges the first section of this chapter to the current one. I have argued that Odysseus is a highly accessible referent for the external audience from the very beginning of the poem. In what follows, I will show that he is most present in the mind of the internal audience as well. He is a kind of common cognitive landmark for the people that act in Ithaca during his physical absence. The {40|41} linguistic trace of that is the demonstrative pronoun (ἐ)κεῖνος, [98] which, curiously enough, is uttered almost exclusively by characters; with only a very few exceptions, κεῖνος is never used by the primary speaking ‘I’, not only in the first four books but also in the rest of the poem. According to Cornish’s model, a distal demonstrative pronoun in English and in French may correspond to a low accessibility of the referent, the knowledge about him/her being located in long-term memory. My analysis of the Homeric uses of κεῖνος will present a picture that is more complex. The need to recall somebody who was not a fully activated referent in the immediately preceding discourse is certainly there. However, further communicative intentions mark the utterance of κεῖνος. In brief, they concern a temporary visual recall of the referent and some social/emotional implications relating to that recall. [99] Furthermore, the repeated use of κεῖνος instead of Odysseus’ name has the effect of creating a cross-referencing mark of identity, which may be consciously exploited by the speakers (as we will see in chapters 2 and 3). As Odysseus’ disguise and recognition proceeds, κεῖνος marks different aspects of the hero and different communicative intentions by the individuals surrounding him. These cognitive and social/emotional implications of the use of such a demonstrative may well explain why the primary speaking ‘I’ avoids its utterance.

In book one, Telemachus asks Athena-Mentes if he (she, actually) is a xenos of his father:

καί μοι τοῦτ’ ἀγόρευσον ἐτήτυμον, ὄφρ’ ἐῢ εἰδῶ,
ἠὲ νέον μεθέπεις, ἦ καὶ πατρώϊός ἐσσι
ξεῖνος, ἐπεὶ πολλοὶ ἴσαν ἀνέρες ἡμέτερον δῶ
ἄλλοι, ἐπεὶ καὶ κεῖνος ἐπίστροφος ἦν ἀνθρώπων.

Odyssey 1.174–177; Telemachus to Athena-Mentes

Also, tell me truthfully this, so that I may be aware:
are you visiting for the first time, or are you a xenos of my father?
Because many men used to come to our house,
other men, as also that one was used to turn in any direction among people.


In her Narratological Commentary on the Odyssey, De Jong writes: {41|42}

Telemachus refers to his father as κεῖνος (cf. 177), the pronoun used to refer to an absent person. As such it is the most appropriate pronoun for those at home to refer to Odysseus; indeed, fifty-nine out of a total of eighty-nine instances of (singular) κεῖνος/ἐκεῖνος in the Odyssey concern him. [100]


While I agree that κεῖνος is “the most appropriate pronoun” used to refer to Odysseus, I claim that the reason is not simply because he is absent. keinos is also the most economical way to evoke a number of aspects involving the relationship between the ones who utter keinos and Odysseus as keinos. The total number of occurrences of this pronoun in singular masculine forms in the Odyssey is 103. Of those, 59 have Odysseus as the referent, which represents a remarkable average (57%), especially if we compare this outcome with the highest number of instances of keinos referring to a character other than Odysseus (10, for Telemachus), on the one hand, and with the highest number of occurrences of keinos referring to a male character in the Iliad (7 instances out of 39 [18%] have Achilles as the referent), on the other hand. The passage quoted above concludes Telemachus’ speech addressed to Athena-Mentes (158–177). During this speech, he refers to his father, “a man whose white bones are somewhere, rotted by the rain for sure, and lie on the land” 161–162), by using another κεῖνος (163; this passage will be commented upon below) and by using ὁ (166 and 168). “Of my father” at 175 translates the adjective πατρώϊός. Thus, “that one” at 177 connects to other anaphoric expressions occurring within the same speech, all of them replacing “Odysseus.” [101]

In other passages, the reference to Odysseus as κεῖνος seems to allude to the distance between the absent hero and his relatives and friends; [102] as such, it implies a certain degree of venerability (“that one” as “that famous man,” “the one whom we all love and respect”). The spatial notion of distance is somehow interacting with the social distance toward a person who ultimately deserves praise and honor. [103] Moreover, the utterance of κεῖνος in the Odyssey works as a cross-referencing keyword. Nagy points out an essential feature of oral poetics, namely “meaning by way of reference.” [104] In oral poetics, reference “is not restricted to the immediate context but extends to analogous contexts {42|43} heard in previous performances.” [105] Taking an example given by Nagy himself, the “meaning” of the rolling πῆμα (“pain”) in relationship to Achilles at Odyssey 8.81–82 (“then it was that the beginning of pain starting rolling / upon both Trojans and Danaans”) cannot be disassociated from the “reference” to the Iliadic theme of πῆμα at Iliad XVII 688 (“a god is letting roll a pain upon the Danaans”). [106] The utterance of these thematic words within the performance becomes a cross-referencing activity, which basically signals the “continuous narration extending into the current narrative.” [107] I submit that the numerous uses of κεῖνος uttered with the deliberate intention of referring to Odysseus reflect the same kind of cross-referencing activity. κεῖνος becomes a reference to a thematic “package” concerning an overall perception of the Ithacan hero that is shared among the characters on the stage in Odyssey 1–4. What are the components of this package? Together with spatial distance, fame, and venerability, I point out two communicative intentions that frequently characterize the performative context of κεῖνος in several instances: visualizing Odysseus’ desired appearance and lamenting his supposed death.

Cross-referencing keinos and the visualization of Odysseus

οὐ γάρ πώ τινά φημι ἐοικότα ὧδε ἰδέσθαι
οὔτ’ ἄνδρ’ οὔτε γυναῖκα, σέβας μ’ ἔχει εἰσορόωσαν,
ὡς ὅδ’ Ὀδυσσῆος μεγαλήτορος υἷι ἔοικε,
Τηλεμάχῳ, τὸν ἔλειπε νέον γεγαῶτ’ ἐνὶ οἴκῳ
κεῖνος ἀνήρ, ὅτ’ ἐμεῖο κυνώπιδος εἵνεκ’ Ἀχαιοὶ
ἤλθεθ’ ὑπὸ Τροίην, πόλεμον θρασὺν ὁρμαίνοντες.
τὴν δ’ ἀπαμειβόμενος προσέφη ξανθὸς Μενέλαος·
οὕτω νῦν καὶ ἐγὼ νοέω, γύναι, ὡς σὺ ἐΐσκεις·
κείνου γὰρ τοιοίδε πόδες τοιαίδε τε χεῖρες
ὀφθαλμῶν τε βολαὶ κεφαλή τ’ ἐφύπερθέ τε χαῖται.
καὶ νῦν ἦ τοι ἐγὼ μεμνημένος ἀμφ’ Ὀδυσῆϊ
μυθεόμην, ὅσα κεῖνος ὀϊζύσας ἐμόγησεν
ἀμφ’ ἐμοί, αὐτὰρ ὁ πυκνὸν ὑπ’ ὀφρύσι δάκρυον εἶβε,
χλαῖναν πορφυρέην ἄντ’ ὀφθαλμοῖιν ἀνασχών.
τὸν δ’ αὖ Νεστορίδης Πεισίστρατος ἀντίον ηὔδα·
Ἀτρεΐδη Μενέλαε διοτρεφές, ὄρχαμε λαῶν,
κείνου μέν τοι ὅδ’ υἱὸς ἐτήτυμον, ὡς ἀγορεύεις·

Odyssey 4.141–157; Helen to Menelaus; Menelaus to Helen; Pisistratus to Menelaus {43|44}

“For I say I never saw anyone so similar,
neither man nor woman, and veneration takes possession of me as I look at him,
as this one here resembles the son of great-hearted Odysseus,
Telemachus, whom in the house, still young,
that man left, when the Achaeans, because of me, impudent,
went beneath Troy, urging reckless war.”
Menelaus of reddish-yellow hair answered by saying to her [Helen]:
“Now I am realizing this as well, wife, as you make the comparison;
these feet and these hands really are such as the ones of that one [i.e. Odysseus’ ones],
and the quick glances of the eyes, and the head, and the hair atop.
And right now, I was recalling memories of Odysseus
and I was telling about him: how much that one suffered and wailed
for me. At that point, he [Telemachus] shed a dense tear from his eyelids;
he had held the purple cloak before his eyes.”
At that point, the son of Nestor Pisistratus said before him:
“Son of Atreus, Menelaus, cherished by Zeus, leader of the army,
this is truthfully the son of that one, as you are arguing.”


The four instances of keinos included in this passage allow us to expand our understanding of the range of uses and meanings of such a demonstrative, as it refers to Odysseus. The first, occurring in the adjectival construction κεῖνος ἀνήρ (145), retrieves background knowledge about Odysseus’ and the Achaeans’ role in the initial decision about the war against Troy (145–146). It also triggers a little series of further uses of keinos uttered by Menelaus and by Nestor’s son Pisistratus. At line 152, an act of lament is going on, and the use of keinos in this context will be analyzed in the next sub-section. The two instances of keinos that begin lines 149 and 157 introduce, instead, the idea of potential visibility. As a matter of fact, in lines 149–151 Menelaus resumes the comparison of the appearance of the young man in front of him to Odysseus, which Helen had already expressed at lines 141–144. However, unlike what Helen says (“this man has a likeness to the son of great-hearted Odysseus”), Menelaus’ comparison goes into greater detail. He posits the resemblance of Telemachus’ feet, hands, eyes, and hair to the same features of Odysseus by establishing a close relationship between what he actually sees and what he mentally visualizes. The keywords for this cognitive visual operation are κείνου, the particle γάρ, and the marks of demonstratio ad oculos τοιοίδε + τοιαίδε (149). Placing κείνου at the very beginning of both the sentence and the line might have corresponded to a paralinguistic effect of vocal emphasis by the performer (such as a higher pitch {44|45} level), which ultimately serves the information structure. [108] The new and the most relevant information conveyed by that utterance is the sudden visualization of Odysseus’ feet, hands, eyes, and hair (their epiphanic appearance, so to speak) at the exact moment in which Telemachus’ feet, hands, eyes, and hair are scrutinized (the verb νοέω at 148 includes such a recognition process [109] ). So, Menelaus starts the explanation of what he perceives by saying κείνου γάρ, where γάρ has the double function of explicative connective between the previous and the following discourse act (“I am realizing that resemblance too; indeed”) and of a particle marking the visual evidence of a certain fact (“Of that man—yes, here we are—”). [110] At 157, Pisistratus uses keinos to convey the same visual connection, in briefer terms (“this one is truly the son of that one”). [111] In book one, Athena-Mentes uses keinos with the same intention in a very similar context involving the recognition of Telemachus’ resemblance to Odysseus: αἰνῶς μὲν κεφαλήν τε καὶ ὄμματα καλὰ ἔοικας / κείνῳ, “As for the head and the fine eyes, you terribly resemble that man,” 208–209). Later, at 212, he (the goddess, actually) remembers the first time he met Odysseus and concludes “since that time I have not seen Odysseus nor has that man (κεῖνος) seen me.” The visual implications of the latter keinos are quite clear. [112] Likewise, Nestor in book three speaks about his own σέβας, “awe,” once he perceives in Telemachus’ appearance the strong similarity with the father (εἰ ἐτεόν γε / κείνου ἔκγονός ἐσσι· σέβας μ’ ἔχει εἰσορόωντα “if it’s true that you are his son; a sacred respect seizes me as I look at you,” 122–123); later in his speech, he refers to Athena’s closeness to Odysseus as visually perceptible (οὐ γάρ πω ἴδον ὧδε θεοὺς ἀναφανδὰ φιλεῦντας, / ὡς κείνῳ ἀναφανδὰ παρίστατο Παλλὰς Ἀθήνη, “Indeed, I never saw the gods loving one in such a manifest way, as Pallas Athena was standing by that one,” 221–222). In sum, the visual relationship established by the speaker in the moment in which s(he) utters keinos—possibly accompanied by a verbum videndi—hints at the cognitive process of sudden visualization of a {45|46} subject “on the spot” even though the subject is not physically reachable, as if the speaker had momentarily experienced the materialization of his/her presence—the experience, that is, of an epiphany. We can now better contextualize the very first Odyssean occurrence of keinos that has Odysseus as the referent:

εἰ κεῖνόν γ’ Ἰθάκηνδε ἰδοίατο νοστήσαντα

Odyssey 1.163; Telemachus to Athena-Mentes

If they were to see that one come safely back to Ithaca

 

Telemachus is complaining about the reprehensible behavior of the suitors. The verbal antecedent of the anaphoric form is not far (ἀνέρος, 161, two lines before), nor is there any risk of ambiguity. Yet, κεῖνος—followed by γε [113] —is a most appropriate way to refer to Odysseus: it conveys distance, venerability, and potential visibility at the same time (cf. the verbum videndi ἰδοίατο). Telemachus is longing for his father’s nostos; for an instant, he imagines what would happen if he really should appear. The physical absence of the main hero is being replaced by his potential visibility, his potentially available presence. [114]

Cross-referencing keinos and lamenting Odysseus’ supposed death

ξεῖν’, ἐπεὶ ἂρ δὴ ταῦτά μ’ ἀνείρεαι ἠδὲ μεταλλᾷς,
μέλλεν μέν ποτε οἶκος ὅδ’ ἀφνειὸς καὶ ἀμύμων
ἔμμεναι, ὄφρ’ ἔτι κεῖνος ἀνὴρ ἐπιδήμιος ἦεν·
νῦν δ’ ἑτέρως ἐβόλοντο θεοὶ κακὰ μητιόωντες,
οἳ κεῖνον μὲν ἄϊστον ἐποίησαν περὶ πάντων
ἀνθρώπων, ἐπεὶ οὔ κε θανόντι περ ὧδ’ ἀκαχοίμην,
εἰ μετὰ οἷσ’ ἑτάροισι δάμη Τρώων ἐνὶ δήμῳ,
ἠὲ φίλων ἐν χερσίν, ἐπεὶ πόλεμον τολύπευσε.
τῶ κέν οἱ τύμβον μὲν ἐποίησαν Παναχαιοί,
ἠδέ κε καὶ ᾧ παιδὶ μέγα κλέος ἤρατ’ ὀπίσσω.
νῦν δέ μιν ἀκλειῶς Ἅρπυιαι ἀνηρέψαντο·
οἴχετ’ ἄϊστος ἄπυστος, ἐμοὶ δ’ ὀδύνας τε γόους τε
κάλλιπεν· οὐδέ τι κεῖνον ὀδυρόμενος στεναχίζω
οἶον, ἐπεί νύ μοι ἄλλα θεοὶ κακὰ κήδε’ ἔτευξαν.

Odyssey 1.231–244; Telemachus to Athena-Mentes {46|47}

Guest, since you ask me and inquire about these matters:
there was a time in which this house was rich and blameless,
while that man was among his people.
Now, instead, gods wanted something different; they devised evils.
They made that one invisible before all humans. I would not grieve for him dead in this way,
if he had been defeated with his companions in the land of Troy,
and if he accomplished war and was in the arms of his friends.
For him all the Achaeans would have made a tomb,
and he would have gained great glory for his son from that moment on.
But now the Snatchers have carried him off, with no glory for him.
He is not to be seen, not to be asked about, and he left sorrows and laments
to me. And I am grieving and groaning not only about that one,
as in fact the gods have caused other wicked cares for me.


This segment includes a variety of anaphoric expressions that indicate Odysseus, without ever making direct mention of the name itself. κεῖνος ἀνὴρ at line 233 seems to confirm the cross-referencing function I introduced above, since it echoes three uses of κεῖνος previously uttered by Athena-Mentes (lines 199, 209, and 212). A further κεῖνον at line 235 reinforces the hypothesis of a cognitively economical way to refer to a specific person, regardless of (modern) grammatical rules: the verbal trace of the referent does not occur far above (κεῖνος ἀνήρ is just two lines before), nor can the use of κεῖνος be justified by any risk of referential ambiguity (here the subjects are only the gods and Odysseus). [115] About κεῖνος at 235, here I limit myself to saying that the entire relative clause is acutely constructed in order to convey more meanings, and the choice of κεῖνος is part of the strategy. I will offer a separate comment on this clause at the end of the next chapter. In coincidence with a series of counterfactual actions {47|48} pondered by Telemachus at lines 236–240—“… for I would not be so full of akhos even if he were dead. … The Achaeans would have made a grave for him, and he would have won great kleos also for his son afterwards,” which are underscored by the redundant presence of κε at lines 236, 239, and 240 [116] —Odysseus is the highly accessible referent of null anaphors (see 236–238), of weak demonstrative οἱ at 239, and of μιν at 241. Then, at lines 242–244, there is what I identify as a temporary metapoetic switch to a different register in the performance, i.e. from “ordinary” epic narrative to ritual lament. The explicit performative verb στεναχίζω (243), [117] occurring right in the middle of this micro-lament, is the clearest marker of what is going on at this point of the performance, as well as the unquestionably related terms ὀδύνας, γόους (242) and ὀδυρόμενος (243). A specific metrical configuration encapsulates this moment: all three lines are holodactylic and are preceded and followed by “regular” hexameters; furthermore, 243 and 244 start both with a runover (κάλλιπεν and οἶον respectively), [118] which suggests a special interlacing between those lines, if not a prosodic correspondence to the mimetic sobs that the performer might have produced at that moment. [119] Most of all, the architecture of sounds makes line 242 an especially prominent one: beyond the masterful assonant pair ἄϊστος ἄπυστος [120] there is a rhythmical sequence based on the sounds oi + st + st, which is repeated twice (it is immediately recognizable if one reads the line aloud; in a transliterated scriptio continua it looks like oikhetaistosapustosemoidodunastegoouste). [121] This deliberate insertion of a lament-act includes the demonstrative pronoun keinos {48|49} to refer to Odysseus; the lamenting “I” (Telemachus) establishes a close relationship with the subject of lament and mourns for the supposed death of the hero of the poem. [122]

Telemachus is certainly not the only one who expresses these fears and these feelings in the first four books. Menelaus utters very similar words at 4.108–110 (ἐμοὶ δ’ ἄχος αἰὲν ἄλαστον / κείνου, ὅπως δὴ δηρὸν ἀποίχεται, οὐδέ τι ἴδμεν, / ζώει ὅ γ’ ἦ τέθνηκεν “I have a grief that is constantly unforgettable, for that one, (about) how he is away for so long, and we don’t know anything, whether he is alive or has died”) and 181–182 (ἀλλὰ τὰ μέν που μέλλεν ἀγάσσασθαι θεὸς αὐτός / ὃς κεῖνον δύστηνον ἀνόστιμον οἶον ἔθηκεν “Well, the thing must be that a god in person became indignant and made that one alone wretched, with no return”). Penelope’s reply to the ghost of her sister Iphthime at iv 832–834 includes an echo of mourning keinos referring to Odysseus: εἰ δ’ ἄγε μοι καὶ κεῖνον ὀϊζυρὸν κατάλεξον, / ἤ που ἔτι ζώει καὶ ὁρᾷ φάος ἠελίοιο, / ἦ ἤδη τέθνηκε καὶ εἰν ’Αΐδαο δόμοισι “please tell me also about that woeful man, whether he is still alive and sees the light of the sun, or if he is already dead and in the house of Hades”). [123] That Penelope here thinks of her husband and not of her son, who is her immediate cause of worry, is confirmed by two further uses of keinos, which serve to disambiguate the referents: at 819, τοῦ is Telemachus and ἐκείνου is Odysseus (Penelope is the speaker); likewise, at 836 κεῖνον is Odysseus (the ghost speaks). [124] To sum up, with reference to Table 1: twenty-nine out of fifty-nine overall occurrences of keinos with Odysseus as {49|50} the referent are condensed in the first four books of the poem. Keinos serves the thematization of Odysseus not only as a character who is distant from the utterance situation, but also as one whose fame and earned respect is shared among friends and relatives. Furthermore, two relevant intentions are marked by the utterance of keinos: namely, visualizing his potential physical presence and lamenting his supposed death. [125] The grievous absence and the desired presence of Odysseus are complementary aspects of the overall perception of the hero by the characters of the first four books of the Odyssey. Moreover, the thematization of Odysseus as keinos makes keinos a cross-referencing word, according to the principle of “meaning by way of reference.” [126] κεῖνος works as a signpost for Odysseus’ status at a higher hierarchical level within the frame of the entire story told in the Odyssey; its discourse relevance is not only local, concerning particular contexts of speech, but also global, implicating the overall role and significance of the hero.

Significantly, from book five to eight there is no use of keinos that has Odysseus as its referent, with the notable exception of 9.456–457. Polyphemus speaks to the ram to which Odysseus clings and uses κεῖνος to refer to that Oὖτις that is responsible for his own downfall: “If only you could think like us and only be given / a voice, to tell me where he [κεῖνος] is skulking away from my anger …” (tr. Lattimore 1967). keinos here meaningfully conveys the physical distance imagined by the Cyclops and the actual, simultaneous visual nearness or presence of the Ithacan hero under the ram. It is easy to deduce why Odysseus, apart from this exception, is not referred to by keinos in books five to eight: book five introduces Odysseus on the stage for the first time; he even becomes the performer of his own epos from eight to twelve. Thus, attitudes toward, and emotions provoked by, Odysseus, as well as the hero’s own self-perception, drastically change within the narrative. Book thirteen represents an intermediate moment in which the main hero, who is no longer the performer of his adventures, is “seen” only by the primary speaking ‘I’ and by the external audience. Odysseus is considered as the undoubted true Odysseus arriving home, without being observed by anyone else on the stage. The absence of any keinos related to him is motivated by the fact that there is no visual feedback about him by any mortal. Athena is with him, of course, but apart from her he is completely alone. {50|51}

Visual and social/emotional implications of the utterance of (e)keinos elsewhere

The visual and social implications of (e)keinos when it refers to Odysseus in the Odyssey can also be detected in other passages, beyond the figure of Odysseus, beyond the Odyssey, and beyond Homer. Let us begin with the visual evidence, which foregrounds the tight relationship between the (e)keinos-subject and the speaking “I” and which includes (e)keinos accompanied by verba videndi. In the Odyssey itself, in book ten, the companions, released from Circe’s spell, turn from pigs into men again, and recognize Odysseus:

ἄνδρες δ’ ἂψ ἐγένοντο νεώτεροι ἢ πάρος ἦσαν
καὶ πολὺ καλλίονες καὶ μείζονες εἰσοράασθαι.
ἔγνωσαν δέ με κεῖνοι …

Odyssey 10.395–397; Odysseus to the Phaeacians

They became men all of a sudden, younger than before
and much handsomer and taller to see.
Those ones recognized me …


κεῖνοι resumes the visual appearance of the companions after the metamorphosis. Its position adjacent to ἐμέ highlights the subjects involved in the recognition process: “those ones” and “me.” A few lines later, the same visual link in a similar cognitive and emotional process is shown: “As when the calves … returning to the farm-yard … run about their mothers, so those men, once they saw me with their eyes [ὣς ἐμὲ κεῖνοι, ἐπεὶ ἴδον ὀφθαλμοῖσι] pressed around me, weeping” (Odyssey 10.410–415). [127] Also independently of the presence of an {51|52} “I” mark, keinos implies visual activities not only in Homer, but also elsewhere, as in the incipit of Sophocles’ Electra:

Ὦ τοῦ στρατηγήσαντος ἐν Τροίᾳ ποτὲ
’Αγαμέμνονος παῖ, νῦν ἐκεῖν ’ ἔξεστί σοι
παρόντι λεύσσειν, ὧν πρόθυμος ἦσθ’ ἀεί.
τὸ γὰρ παλαιὸν Ἄργος οὑπόθεις τόδε

Sophocles Electra 1–4; the pedagogue to Orestes

O son of the one who once led the army in Troy
son of Agamemnon, now that you are here it is possible
that you watch those things there, which you always desired.
Ancient Argos is here, indeed, which you long for,


The pedagogue and Orestes have reached Mycenae and are looking at the landscape appearing in front of them. ἐκεῖνα does not convey any remoteness; [128] it conversely indicates places that are in sight of the participants to the dialogue hic et nunc (cf. τόδε used to indicate the first entry of the list, namely the plain of Argos). ἐκεῖνα could have been chosen because of the visual activity the pedagogue is suggesting. [129]

More evidence highlights the pure deictic force of the demonstrative; in these cases, Bailly makes (e)keinos equal to an adverb of place. For example, in {52|53} Iliad 19 Zeus is moved to pity by Achilles’ lamenting over Patroclus’ corpse (315–337) and asks Athena why she abandoned Peleus’ son:

τέκνον ἐμόν, δὴ πάμπαν ἀποίχεαι ἀνδρὸς ἑῆος.
ἦ νύ τοι οὐκέτι πάγχυ μετὰ φρεσὶ μέμβλετ’ ’Αχίλλευς;
κεῖνος ὅ γε προπάροιθε νεῶν ὀρθοκραιράων
ἧσται ὀδυρόμενος ἕταρον φίλον· οἱ δὲ δὴ ἄλλοι
οἴχονται μετὰ δεῖπνον, ὁ δ’ ἄκμηνος καὶ ἄπαστος.

Iliad 19.342–346; Zeus to Athena

My daughter, you really abandoned the noble hero, completely.
Achilles didn’t come to your mind at all, did he?
There he is; he sits down before the high-horned ships,
mourning his dearest companion; the others instead
have gone to dinner, but he is fasting from food, he is not enjoying any taste.


keinos in this case marks the visual pointing at Achilles’ location and the focalization on his presence near the ships. [130] The characters may label as keinos somebody (much less often something) that they visually perceive either in the real, physical setting in which they also are (demonstratio ad oculos) or in their imagination (deixis am Phantasma) [131] . In light of the “referent in the mind” model, the difference is not so relevant: in both cases, once we hear/read the utterance of the demonstrative, we (re-)activate the mental representation of the referent (see above, on “Alternative readings of third-person pronouns”). {53|54} This cognitive input linking deixis ad oculos and deixis am Phantasma is crucial as we consider κεῖνος when it occurs within the narration of another’s epiphany:

πᾶν δ’ ἄλσος καὶ βωμὸς Ἀπόλλωνος Παγασαίου
λάμπεν ὑπαὶ δεινοῖο θεοῦ τευχέων τε καὶ αὐτοῦ,
πῦρ δ’ ὣς ὀφθαλμῶν ἀπελάμπετο. τίς κεν ἐκείνου
ἔτλη θνητὸς ἐὼν κατεναντίον ὁρμηθῆναι
πλήν γ’ Ἡρακλῆος καὶ κυδαλίμου Ἰολάου;

Hesiod Shield 69–73; the primary speaking ‘I’ to the audience

The grove and the altar of Apollo of Pagasae entirely
shone, because of the arms of the terrible god, and because of the god himself;
The eyes were flaming as fire. Which mortal being
could have moved against that one,
except for Heracles and glorious Iolaus?


The narrator tells of the approach of Heracles and Iolaus to Cycnus near the cult place of Pagasaean Apollo. [132] The two attend Apollo’s epiphany, which includes traditional motifs such as the exceptional firelight gleaming from the god (and the god’s eyes) and the struggles of mortals to withstand the monstrum.

In the incipit of Sappho’s fragment 31, the speaking “I” recalls as keinos another person in another epiphanic situation: [133]

φαίνεταί μοι κῆνος ἴσος θέοισιν
ἔμμεν’ ὤνηρ, ὄττις …

Sappho fr. 31.1–2; the primary speaking ‘I’ to the “you” addressed

There he is; that one appears to me, equal to gods,
the man who …


The cognitive and emotional relevance of this event as it happens before the eyes of the speaking “I” does not depend on whether the appearance of the male {54|55} in question is imaginative or real. [134] After all, epiphanies are both real and imaginative par excellence: in a single magic moment, humans realize and perceive a transformation of reality; cognitively, that reality was something before and it becomes something else afterwards. [135] In addition to epic and lyric, drama and prose show that keinos can be chosen to indicate the (sudden) appearance of a reality that was not present or not recognized before. This typically happens when (ἐ)κεῖνος occurs beside another demonstrative such as ὅδε or οὗτος. Let us consider, for example, what Oedipus says as he shows himself to the chorus that has just entered in search for him:

ὅδ’ ἐκεῖνος ἐγώ· φωνῇ γὰρ ὁρῶ,
τὸ φατιζόμενον.
[ΧΟ.] ἰὼ ἰώ,
δεινὸς μὲν ὁρᾶν, δεινὸς δὲ κλύειν.

Sophocles Oedipus at Colonus 137–140; Oedipus to the chorus; the chorus

Here I am; I can see with the sound,
as they say.
[Chorus] Oh! Oh!
Fearful to see, fearful to hear!


Oedipus’ words stage his epiphany to the chorus (whose comment “Oh! Fearful to see, fearful to hear!” is confirmatory). [136] I would suggest that ἐκεῖνος marks the sudden appearance of the subject from elsewhere and that ὅδε and ἐγώ mark what is already on the stage. Thus, my paraphrase would not be “This person here is that one you are looking for,” but “Look! Here I am”; [137] in other {55|56} words, what is old to the hearer is ὅδε and what is new is ἐκεῖνος, not vice versa. [138] I argue that the substantial act underlying many utterances of (ἐ)κεῖνος is to establish a visual relationship between the speaking “I” and the (ἐ)κεῖνος-subject. The (ἐ)κεῖνος-subject is initially far way or absent or unseen, and through the utterance (ἐ)κεῖνος that subject becomes—all of a sudden—close or present to the speaking “I” ’s vision. [139] In arguing this claim, I am also conveying my understanding that no demonstrative—whether deictic or anaphoric—ever marks the place of a thing or a person, but rather expresses a certain perceptual relationship (or orientation) that the “I” establishes. [140] Distance and proximity are relative concepts; they may also embrace different components, visual and/or emotional and/or social. In the case of divine or heroic epiphanies, the vision is, of course, charged with ritual and initiatory values; in less religious contexts, it may simply correspond to illuminating connections that improve the speaker’s understanding. A very famous passage from Aristotle’s Poetics illustrates the latter case:

διὰ γὰρ τοῦτο χαίρουσι τὰς εἰκόνας ὁρῶντες, ὅτι συμβαίνει θεωροῦντας μανθάνειν καὶ συλλογίζεσθαι τί ἕκαστον, οἷον ὅτι οὗτος ἐκεῖνος.

Aristotle Poetics 1448b 15–17; Aristotle to the recipients {56|57}

This is the reason why they delight in looking at images: while scrutinizing them it happens that they learn and make inferences about what each item is, for example, “This one … there he is!”


The philosopher is speaking about engaging in mimesis. The interpretation “Oh yes! This character (on the stage) corresponds to that (famous) one I know” could be modified due to the “epiphanic” character of the situation. In the former example from Oedipus at Colonus, the starting information could be ὅδε, with ἐκεῖνος then added; likewise, in the Poetics οὗτος could simply indicate an individual standing on the implied stage, and ἐκεῖνος could reveal the recognition process that brings the speaker to suddenly see on the stage a certain character: “Now I see who this character on the stage is; there he is!” Not incidentally, the referents of both οὗτος and ἐκεῖνος in this passage are missing. Their referents are in the mind and not in the text. In literary studies, these masculine forms have been thought problematic (one scholar even proposed emending the text in favor of neuter forms.) [141] This syntactical “embarrassment” is significant: it is difficult to imagine that Aristotle chose third-person pronouns to refer to individuals who were not previously mentioned in the text. Yet, the connection of these pronouns with the familiar cognitive operations of spectators of theatrical pieces is extremely relevant to the argument of mimesis, and Aristotle’s testimony is quite remarkable. More or less colloquial expressions such as τοῦτο ἐκεῖνο or τόδ’ ἐκεῖνο—mentioned by many grammarians and frequently occurring in prose as well as in comedy (as well as in tragedy)—are, in my view, evidence of the cognitive identification for which I argued above. Whether the target of the expression is the reality indicated via οὗτος or the reality indicated via ἐκεῖνος, by uttering that phrase the speaker is establishing a visual relationship between what (s)he sees and what is before his/her eyes or the mind’s eye. [142]

The visual implications that (I have argued) underlie the choice of (e)keinos may be the basis for the further communicative intentions that I have already identified in the first books of the Odyssey—that is, the lament over the supposed death of someone and the sense of veneration. Both intentions draw from a distance that is social and, in this case, positive. Dead people are recalled as ἐκείνοι in Heraclitus fr. 62: ἀθάνατοι θνητοί, θνητοὶ ἀθάνατοι, ζῶντες τὸν ἐκείνων θάνατον, τὸν δὲ ἐκείνων βίον τεθνεῶτες “immortals are mortals, mortals are immortals, living the death of those ones, dying the life of those {57|58} ones.” [143] Dead heroes are recalled through ἐκεῖνος not only in Homer, [144] but also in Philostratus’ On Heroes. [145] An attitude of veneration towards people already dead is present also in tragedy (and, with parodic effects, in comedy). Let us consider, for example, Admetus’ moving words as he laments his wife’s death:

ἰώ, στυγναὶ
πρόσοδοι, στυγναὶ δ’ ὄψεις χήρων
μελάθρων. ἰώ μοί μοι. αἶ αἶ.
ποῖ βῶ; ποῖ στῶ; τί λέγω; τί δὲ μή;
πῶς ἂν ὀλοίμαν;
ἦ βαρυδαίμονα μήτηρ μ’ ἔτεκεν.
ζηλῶ φθιμένους, κείνων ἔραμαι,
κεῖν ’ ἐπιθυμῶ δώματα ναίειν.

Euripides Alcestis 860–867; Admetus to the chorus

Oh, how gloomy
are the entrances, how gloomy the sight
of this bereaved dwelling. Oh! Alas, alas!
Where should I go? Where should I stay? What should I say? What should I not say?
How I wish I would perish!
Really, my mother bore me luckless.
I admire the dead, I long for those ones,
those dwellings are the ones I love to stay in.


The emotional implication of calling Odysseus keinos for the internal as well as the external audience of the Odyssey is echoed by this passage. The use of keinos conveys not only the tragic distance that separates the living from the dead, but also the impetus to establish contact with the dead, at least on the imaginative level. [146] {58|59}

Praises of heroes and blessings (or makarismos formulas) often include (e)keinos with reference to the object of praise. [147]

ἔνθα δὲ ναιετάουσι παραὶ Διὶ τερπικεραύνῳ
σεμναί τ’ αἰδοῖαί τε· μεγ’ ὄλβιος ὅν τιν’ ἐκεῖναι
προφρονέως φίλωνται ἐπιχθονίων ἀνθρώπων

Homeric Hymn 2 to Demeter 485–487; the primary speaking ‘I’ to the audience

They [Demeter and Persephone] reside there [on Olympus] beside Zeus who enjoys the thunderbolt,
venerated and revered. Considerably blessed is the one whom those ones
favor and hold very dear, among mortals. {59|60}


These lines from the closure of the Homeric Hymn II to Demeter show a masterful interlacing of reciprocal blessings between the two goddesses celebrated (the venerated ones, “those ones”) and the initiated individuals who worship them and who, in turn, are going to be worshipped. As a further significant instance of keinos conveying praiseworthiness I quote Plutarch:

ἐπεὶ δ’ ἀπέβη, τοὺς μὲν ἄλλους στρατηγοὺς οὐδ’ ὁρᾶν ἐδόκουν ἀπαντῶντες οἱ ἄνθρωποι, πρὸς δ’ ἐκεῖνον συντρέχοντες ἐβόων, ἠσπάζοντο, παρέπεμπον, ἐστεφάνουν προσιόντες, οἱ δὲ μὴ δυνάμενοι προσελθεῖν ἄπωθεν ἐθεῶντο, καὶ τοῖς νέοις ἐδείκνυσαν οἱ πρεσβύτεροι.

Plutarch Alcibiades 32.3; Plutarch to the recipients

As he [Alcibiades] disembarked, the people did not seem to look at the other generals whom they were encountering, but they ran towards that one, and kindly greeted him, and escorted him, and put wreaths on him if they could get close to him. And those who could not get closer were seeing the scene from afar, and the older men were pointing at him to the young.

(e)keinos is apt to express social distance when it implies contempt instead of praise, wrath instead of veneration. I offer two examples from lyric and drama of this negative social distance:

ἔν]νεκα κήνας

Alcaeus fr. 283.14; the primary speaking ‘I’ to the audience

because of that one [Helen]


The stigmatization of Helen by Alcaeus in this fragment is representative of what I refer to when I speak of negative social distance. [148] As for drama: in Euripides’ Orestes, as Orestes tells Pylades about the arrival of Helen together with Menelaus, he says: {60|61}

οὐκ ἐκεῖνος ἀλλ’ ἐκείνη κεῖνον ἐνθάδ’ ἤγαγεν.

Εuripides Orestes 742; Orestes to Pylades

It was not that man [Menelaus] but that woman [Helen] who carried that man here.


Some instances of keinos in the Iliad belong to this group, and these are uttered by characters, never by the primary speaking ‘I’. [149] For example, communicating wrath and disdain seems to be the main goal of Apollo: in book twenty (105–106), the god spurs Aeneas on to fight against Achilles by expressing his contempt for the origins of Achilles, who is his human antagonist (καὶ δὲ σέ φασι Διὸς κούρης ’Αφροδίτης / ἐκγεγάμεν, κεῖνος δὲ χερείονος ἐκ θεοῦ ἐστιν “since they say that you were born of Aphrodite, Zeus’ daughter; but that one was born of a minor god”). Odysseus refers to Achilles as keinos in book nine (678), after the embassy is over: κεῖνός γ’ οὐκ ἐθέλει σβέσσαι χόλον “that one, he does not want to quench his rage.” [150] Agamemnon and whoever does not fight are the referents of further instances of keinos that express the speaker’s wrath at 13.109 and 232 respectively.

A last group of instances of (e)keinos signals a peculiar implication: social politeness. [151] Among the meanings of (e)keinos recorded in LSJ, I have already mentioned “for things, of which one cannot remember or must not mention the name.” In addition to Pindaric instances (Pindar has to practice professional politeness in many of his epinician songs; in fact he a master of that [152] ), I would like to submit to the reader that in Sophocles’ Ajax several uses of keinos referring {61|62} to Ajax himself may be interpreted as signs of social politeness. For the characters of the play, their confrontation with the hero’s tragic madness cannot be distinguished from their ultimate respect (and veneration) of him; their use of keinos thus points at the hero without naming him: keinos marks distance but also respect. [153] Furthermore, it is possible that the three instances of (e)keinos in the third stasimon (lines 1194, 1197, and 1198), all of which have the anonymous referent “the inventor of war,” indicate someone whose identity is very well-known to the audience (the chorus at that point is lamenting the disastrous consequences of war and seems to express a quite negative distance towards that individual, which further justifies the use of (e)keinos). Herodotus indirectly offers us an instance of (e)keinos arguably implying multiple communicative intentions. It is included in the second of the reported funerary epigrams for the fallen at the Thermopylae (whose author since ancient times has been thought to be Simonides [154] ):

ὦ ξεῖν’, ἀγγέλλειν Λακεδαιμονίοις ὅτι τῇδε
κείμεθα, τοῖς κείνων ῥήμασι πειθόμενοι

Herodotus 7.228; the dead warriors to passersby

You that are passing by, go announce to the Spartans that in this place
we lie, obedient to the words of those ones.


Traditionally κείνων equals “of them,” as the regular third-person pronoun. Yet, what is at stake here is an arguably stronger reference to the Spartans’ leaders. κείνων is uttered by the dead warriors themselves (note the close relationship between the speaking plural “I”—κείμεθα—and the keinos-subject). κείνων reflects their own point of view, which possibly includes more aspects, such as the famousness or gloriousness ascribed to them, together with a sense of politeness towards those who brought them to death, which is conveyed by pointing at them without naming them. Still, the passersby could probably easily retrieve the actual referents.

I now bring to a close this partial survey of (e)keinos beyond the Odyssey and beyond Homer with some remarks on its uses in prose. Scholars of ancient Greek history or philosophy will perhaps consider the evidence I have presented as {62|63} pertaining exclusively to poetry; uses of ekeinos in prose would not be emotionally charged and would be more “technically” used. A “technical” aspect would be the disambiguation of persons mentioned in the text (typically, ἐκεῖνος refers to the farther of two referents). This is certainly an important role of this demonstrative, but it does not explain why many prose writers employ ἐκεῖνος close to its referent, nor does it account for αὐτῶν and ἐκείνων as forms of possessive non-reflexive pronouns. [155] But an explanation for these uses can be found if we consider the cognitive level of the reception of these pronouns, both as they occur separately and as they occur together. An indirect insight concerning the processing of these anaphoric expressions comes from the work of Bakker. I have already mentioned Bakker’s remark that ekeinos in Herodotus may convey the standpoint of the speakers whose words are being reported. Herodotus’ recipients of the indirect speech are helped in reconstructing the original dialogic situation in which αὐτοί are the speaking subjects and ἐκείνοι are the addressees. Now, let us analyze the following passage by Plato:

[ΣΩ.] Ἐπιθυμητὴς γάρ εἰμι, ὦ φίλε, τῆς σῆς σοφίας καὶ προσέχω τὸν νοῦν αὐτῇ, ὥστε οὐ χαμαὶ πεσεῖται ὅτι ἂν εἴπῃς. ἀλλά μοι λέξον τίς αὕτη ἡ ὑπηρεσία ἐστὶ τοῖς θεοῖς; αἰτεῖν τε φῂς αὐτοὺς καὶ διδόναι ἐκείνοις; [ΕΥΘ.] Ἔγωγε. [ΣΩ.] Ἆρ’ οὖν οὐ τό γε ὀρθῶς αἰτεῖν ἂν εἴη ὧν δεόμεθα παρ’ ἐκείνων, ταῦτα αὐτοὺς αἰτεῖν; [ΕΥΘ.] Ἀλλὰ τί; [ΣΩ.] Καὶ αὖ τὸ διδόναι ὀρθῶς, ὧν ἐκεῖνοι τυγχάνουσιν δεόμενοι παρ’ ἡμῶν, ταῦτα ἐκείνοις αὖ ἀντιδωρεῖσθαι; οὐ γάρ που τεχνικόν γ’ ἂν εἴη δωροφορεῖν διδόντα τῳ ταῦτα ὧν οὐδὲν δεῖται.

Plato Euthyphro 14d–e; Socrates to Euthyphro

[Socr.]: In fact, I am longing for your wisdom, friend, and my attention is drawn to it, so that what you might say does not fall to the ground. So, tell me: what is this service offered to the gods? Are you saying that it means asking them [as entities higher in rank] and giving them [as most venerable beings]? [Euth.]: I do. [Socr.]: But isn’t it the case that asking in a right way is to ask for what we need from those ones? [Euth.]: And so? [Socr.]: And, on the other hand, giving in a right way isn’t equal to giving those ones in turn the presents that those ones need from us? For somehow it wouldn’t be cunning to give someone who gives things the presents that he does not need at all. {63|64}


In this crucial passage towards the end of Plato’ s Euthyphro, Socrates discusses the concept of “holiness.” Socrates is suggesting that it is not what is holy that pleases the gods, but the reverse, it is what pleases the gods that is holy. In the passage in question, Socrates discusses what is correct to ask of the gods and what is correct to give them. I argue that the use of ἐκεῖνος, which is interestingly placed beside αὐτός (the referent being the same, i.e. gods), plays a significant role in the rhetoric of the discussion. The four instances of ἐκεῖνος have a cognitive function, which is to make the referents—the gods—forcefully present to the mental landscape of the recipients. They also have a notable discourse relevance: their referents are pointed out as the most prominent discourse subjects, in both Socrates’ and Euthyphro’s view. Such prominence could have been prosodic too, in either an oral or aural reception of the text. Most of all, the cognitive focus matches the philosophical focus. The accent of Socrates’ view is on the powerfulness and superiority of the gods; we should ask them what we need from those ones, and, analogously (literally “in a parallel way,”, note the two instances of αὖ), we should give those ones what those ones are asking for. The accent—cognitive, philosophical, and maybe even prosodic—is on the powerfulness and superiority of “those ones,” as a socially charged reference: they are the ones who establish what is to be asked and given; they establish what is holy, not the humans. Thus, the use of ἐκεῖνος hardly seems a mere equivalent of “he” in this case; its utterance is attributable to specific cognitive and rhetorical reasons. As for the two instances of αὐτός in the cited passage, my suggestion is that they do not have the same discourse relevance as κεῖνος; they occur immediately after θεοῖς and ἐκείνων respectively, and they keep the focus on gods; still, αὐτός might lexically underscore the reference to higher-rank individuals, as is common with the gods. [156]

The following example, furthermore, suggests that the richness underlying the choice of ἐκεῖνος as third-person pronoun is not exclusive to poetry:

Λυκοῦργον μέντοι τὸν θέντα αὐτοῖς τοὺς νόμους, οἷς πειθόμενοι ηὐδαιμόνησαν, τοῦτον καὶ θαυμάζω καὶ εἰς τὰ ἔσχατα [μάλα] σοφὸν ἡγοῦμαι. ἐκεῖνος γὰρ οὐ μιμησάμενος τὰς ἄλλας πόλεις, ἀλλὰ καὶ {64|65} ἐναντία γνοὺς ταῖς πλείσταις, προέχουσαν εὐδαιμονίᾳ τὴν πατρίδα ἐπέδειξεν.

Χenophon Constitution of the Lacedaemonians 1.2; Xenophon to the recipients

Yet, I do admire Lycurgus, who set up the laws for them [the Spartans]; they abide by them and they are prosperous. I consider him extremely wise. That one, in fact, did not imitate the other cities, but he mastered the opposite of what most cities were experiencing, and he demonstrated that his land was excelling in prosperity.


In these (almost initial) lines of Xenophon’s Spartan Constitution, Lycurgus is recalled as ἐκεῖνος not only because he is going to be the absolute focus of the following sentences, but also because the author wants to convey Lycurgus’ fame and his own praise of him.

Conclusion: the force of “that one”

This chapter introduces an alternative approach to Homeric third-person pronouns. As a narrative unfolds, the recipients reconstruct the contextual frame of each point of the narration and monitor the set of characters and objects that are present or involved at any place and time. Any pronoun permits the retrieval of the mental representation of the referent, regardless of whether there is a previous verbal trace indicating it or not. According to the approach that I have proposed, the “referent in the mind” model can replace the traditional “referent in the text” model. The different kinds of pronouns we find in a narrative text correspond to different degrees of activation of the referent in the speaker/narrator’s mind. As for Homeric language, null anaphors imply highly accessible referents; non-accented pronouns (such as ἑ) imply that the referent is highly accessible but the relevant information of the sentence is not about it; weak demonstrative pronouns (such as ὁ) imply that the referent is easily accessible and the relevant information that is forthcoming is about it (or it marks the presence of someone while another person is also mentioned, or it marks a shift in the discourse); demonstrative pronouns (such as (ἐ)κεῖνος) strongly recall of entities that were not previously fully activated. Furthermore, each choice may correspond to the discourse function whereby the referent is mentioned within the narrative structure (the referent can play a local role within longer segments of narrative and it can fit the global discourse function of a single character). Finally, the utterance of “heavier” anaphoric expressions, such as demonstrative pronouns, can carry with it a specific pragmatic force, which may be visual and/or social and /or emotional. This force may {65|66} constitute what I elsewhere call the “conspicuity” of the referent. [157] The results of the analyses of three Homeric passages taken as study cases show that the accessibility of the referent is certainly an important interpretive factor; however, the primary speaking ‘I’ seems to play with the high accessibility of the referent, so as to manipulate its meaning, as in the case of Odysseus in the first lines of the Odyssey. Recalling the hero has to accomplish literary purposes (signaling novelties with respect to previous traditional material) and discourse purposes (marking the global status of the main character from the very beginning). Finally, the pronominalization of Alexander in Iliad 3 shows pragmatic purposes (conveying cognitive and emotional attitudes towards the referent).

I have drawn specific attention to Homeric (ἐ)κεῖνος because the classical accounts concerning its use are contradictory and because Odysseus in the Odyssey is recalled by means of κεῖνος quite frequently. My textual analysis of the occurrences of κεῖνος referring to Odysseus in the first four books of the poem show that the speakers’ uses of κεῖνος express their wish to see him, their own veneration of him, and their lamentation of his supposed death. The chapter ends with a short survey of the use of (ἐ)κεῖνος elsewhere in ancient Greek literature, as a contribution to an overall picture of the complexity underlying the use of such a demonstrative. Some features parallel those already commented upon in my analysis of Odysseus in the Odyssey and others tell more about what is being conveyed: (ἐ)κεῖνος may mark sacred and secular epiphanies, negative social/emotional distance (wrath, contempt), positive social/emotional distance (veneration), and social politeness. Some instances from prose texts show that considering (ἐ)κεῖνος as a “plain” third-person pronoun or as a mark for distant referents hinders a deeper understanding of linguistic communication.

The opening argument of this chapter concerned the challenging idea of “literary grammar”; that is, considering some grammatical phenomena as linguistic signposts for communicative intentions that are relevant to literary meanings and to literary criticism. What are the elements of literary grammar resulting from the analysis so far? First, a conception of grammatical rules that is strictly oriented to written texts does not help in interpreting those instances in which the verbal mention of the referent is missing. Only if our processing of ἄνδρα at Odyssey 1.1 is free from the “problem” of why a verbal mention of the referent is missing will the full literary value of ἄνδρα become clear: the referent of ἄνδρα is in the mind of both the primary speaking ‘I’ and the external audience as a relevant shared knowledge, which parallels the cognitive presence of Odysseus in the mind of the internal characters. Nor does this mean that readers can only comprehend the text in a mode that is oriented exclusively toward writing; cognitive studies tell us, on the contrary, that readers join listeners in {66|67} many aspects of the cognitive processing of textual references. A second element of literary grammar concerns the meanings of demonstratives. As I have argued in a previous work, pointing in archaic Greek poetry does not indicate place, but orientation. κεῖνος does not convey an absolute meaning about the position or place of somebody or something; rather, it conveys the relative meaning of a relationship established by the speaker with the keinos-subject. This relationship is visual and/or social and/or emotional. Behind the utterance of every κεῖνος—in Homer at least—there are standpoints, attitudes, and intentions that motivate that specific linguistic choice instead of another. Moreover, a cognitive approach allows us to enjoy what anaphoric and deictic references have in common, and how much backward projections ultimately have to do with forward projections. A third element is directly related to teaching and translating. In all of the translations I have proposed, κεῖνος is “that one.” My presupposition is that the linguistic competence of the primary speaking ‘I’ was allowing him to choose between ὁ and κεῖνος, ὅδε and κεῖνος, οὗτος and κεῖνος, and αὐτός and κεῖνος in formulating sentences beyond metrical constraints. Oral and written translations should be more respectful of these different anaphoric choices. The following list is a synthesis of the cognitive and pragmatic meanings of κεῖνος in the different contexts and cotexts that have been analyzed so far:

  • that one whom you already know and who now is becoming a relevant piece of information; in particular:

  • that one is now present to my eyes;

  • that one whom all of us know about;

  • that one I/we am/are venerating;

  • that one I/we am/are venerating and lamenting over, since he is dead;

  • that one I/we are detesting or that one I/we want to distance;

  • that one that cannot be mentioned by name (but you all know whom I/we refer to);

  • that one, that is, Odysseus.

Lastly, the interpretive perspective that I have proposed considers the literary content of Homeric poetry not as something that is slightly modified by factors such as word order or particles, but, on the contrary, as something that is constrained by those factors. ὅ γε at Odyssey 1.4, for example, does not introduce Odysseus’ sufferings in the same way as a mere ὁ would do. The next chapter will show further evidence of Homeric mastery in using and playing with words that do not seem to have any particular relevance. {67|}

Footnotes

[ back ] 1. “One way to think of consciousness is as a narrow spotlight that can at any one time be directed at only a small area of the available scene—but a spotlight that wanders constantly, sometimes with purpose and sometimes not. Exactly what kinds of things it illuminates—the contributions made by the senses, by images remembered and imagined, by abstractions, and by affective states—are not easy to specify and disentangle. But at least we may be able to agree that the general notion of consciousness has strong introspective support” (Chafe 1974:111).

[ back ] 2. By “external audience,” I mean all the possible kinds of “there and then” audiences of Homeric performances; it corresponds to the function of “énonciataire,” in Calame’s term (1986:35). I will refer to the whole constituted by the external audience and us as modern readers with the term “recipients.” By “internal audience,” I mean the various characters (sometimes identified, sometimes not, and sometimes just implied) that happen to occasionally be the audience of any speech within the narrated story.

[ back ] 3. Pioneering works exploring cognitive aspects of the Homeric language are Bakker 1997b and Minchin 2001. The former shows the correspondence between the segmentation of Homeric lines and the rhythms of the flow of consciousness (metric units fit information units); the latter interprets some narrative patterns in terms of cognitive scripts. On the importance of studying pronoun references in written as well as spoken texts, see Emmott 1997:8–9. Long ago, the psychologists Sanford and Garrod asserted: “reference resolution in all its forms constitutes the cornerstone of successful comprehension in terms of the reader’s task of building an appropriate mental model of what is being said. If one can understand how reference is resolved, then an understanding of other parts of the general comprehension process will follow automatically” (Sanford and Garrod 1981:89). In ancient Greek studies, reference tracking concerning characters is first explored by Bakker (1997b), on which, see below, n68; in Latin, by Bolkestein and Van de Grift 1994; Luraghi 1998; and Kroon 2009, which includes further bibliographical references.

[ back ] 4. See Bosch 1983:7; Bosch’s full study of the ancient grammarians’ definitions (and concepts) of anaphora is indispensable.

[ back ] 5. The interpretation of ἄνδρα as an anaphor (“the man”) rather than as a generic indefinite noun (“a man”) is already supported by Pierron (1887:5–6): “ἄνδρα équivaut à τὸν ἄνδρα. Ce n’est pas d’un héros quelconque qu’il s’agit.” Furthermore, common readings of this line do not take “the man” as “the man who did so and so” but, rather, as “the man, … of many turns, who did so and so.” In other words, ὅς introduces a non-restrictive relative clause, which lends support for “the man” as appealing to some previous knowledge of the individual in question. On the syntax and the IE origin of Homeric ὅς introducing non-restrictive relative clauses, cf. Bonifazi 2004c:49–54 and 63–68.

[ back ] 6. Kahane’s insights about the use of ἄνδρα throughout the poem are very relevant to the current argument: first, ἄνδρα at 1.1 is taken as a “reference by anaphora to Odysseus” (1994:59); second, later in the poem, “through a generic reference, ἄνδρα hints at Odysseus in particular” (1994:61); see Odyssey 6.180–181; 8.138–139; 18.52–53 and 79–81; 22.31–32. Anaphoric reference, Kahane argues, may depend on the point of view of the speaker, and it “may change with regard to the different states of knowledge of different characters” (1994:64).

[ back ] 7. Goldhill 1991:1–2 and 2, respectively, for the quotations.

[ back ] 8. Pucci 1982: passim.

[ back ] 9. Clay 1983:26 and 27, respectively.

[ back ] 10. Fenik 1974:18.

[ back ] 11. All the latter three quotations are from Fenik 1974:53 (italics in the text).

[ back ] 12. So in Fenik’s translation (1974:13); cf. Mattes 1958:127: “… die Personen seiner “Erdichtung”—wie dies bei realen Personen selbstverständlich der Fall ist—die Beweggründe ihres Handelns in sich selber tragen, und zwar “nach Massgabe” ihres “Charakters” und der Reaktion dieses “Charakters” auf die vorliegende Situation, so, wie dies im Leben auch tatsächlich geschieht.”

[ back ] 13. Focke 1943:132. Fenik also discusses the view of Wilamowitz and others that a lacuna is to be assumed (1974:11).

[ back ] 14. Cf. Fenik 1974:53.

[ back ] 15. N. Austin 1972:6 and 8 respectively.

[ back ] 16. N. Austin 1972:5–10. I will focus on Eumaeus’ concerns in ch. 2.

[ back ] 17. N. Austin 1972:9; he refers to Lausberg 1960:590, where the beginning of the Odyssey is quoted as an example. Austin draws an interesting connection with the incipit of Virgil’s Aeneid (arma virum que cano; Aeneas’ name occurs for the first time only at line 92).

[ back ] 18. Rüter 1969:34–52; 37 for the quotation.

[ back ] 19. See Nagy 1990b:34.

[ back ] 20. Goldhill 1991:5 “… the Odyssey in a self-reflexive way highlights, first, words and their use as a concern.”

[ back ] 21. As for the mental representation of the referents and their accessibility, I refer to Emmott 1995 and 1997; Cornish 1999 and 2005; Chafe 1994. As for referential choices linked to narrative structure, I refer to Marslen-Wilson et al. 1982; Fox 1987; Bolkestein and Van de Grift 1994; Kroon 2009. Finally, as for some interactional aspects of anaphors, I refer to Lakoff 1974; Daneš 1990; and Conte 1999.

[ back ] 22. For more and less radical notions of “reference,” see Emmott 1997:202–203.

[ back ] 23. Emmott 1997:200.

[ back ] 24. Cornish 2002:473; italics in the text.

[ back ] 25. Example from Brown and Yule 1983:202, quoted by Emmott 1997:201; italics in the text.

[ back ] 26. See Emmott 1997:197–204.

[ back ] 27. Example from Cornish (2005:206); italics in the text.

[ back ] 28. Emmott 1997:211; Emmott 1995:87; Cornish 2002:474–475. “Both anaphora and deixis … operate at the level of memory organization, managing the latter by guiding the processing of incoming segments of a text” (Cornish 1999:117). In Bonifazi 2004c, I argue that Apollonius Dyscolus defines third-person pronouns as δεῖξις τοῦ νοῦ, “deixis of the mind” (Apollonius Dyscolus Syntax 2.12 in Uhlig 1910).

[ back ] 29. Example from Cornish (2005:202); italics in the text.

[ back ] 30. Example from Cornish (2005:206n9); italics in the text.

[ back ] 31. “Anaphora is a means of managing the memory representation of the discourse being constructed by the speech participants on the basis of a co-text as well as a relevant context” (Cornish 2005:200). Co-text is understood as the linguistic part of the context and, moreover, denotes the text-internal linguistic context as opposed to the text-external situational context (Glück 1993:344).

[ back ] 32. Emmott 1997:197. In Emmott’s view, “priming” and “focusing” are the two basic mechanisms utilized by the author of any narrative text in order to guide the monitoring activities of the reader. Priming provides the reader with “knowledge of the time, location, and participant set,” while focusing “draws attention to one or more participants within the group” (1997:197). Emmott (1997:217)—recalling Chafe’s metaphor for what is presently in consciousness as “on stage” (Chafe 1974:121)—asserts: “priming is like characters standing on a stage in front of us, whereas focusing is like someone drawing our attention towards one of these characters.”

[ back ] 33. For an extended analysis of the different scenarios in narrative processing, see Emmott 1997:103–194.

[ back ] 34. The “activation,” “non-activation,” and “semi-activation” of consciousness about discourse characters or discourse facts is explored by Chafe 1994 (these are his terms).

[ back ] 35. Cornish 1999:25. The cognitive notion of attentional state was first presented in a seminal article by the computational linguists Grosz and Sidner (1986:179).

[ back ] 36. On the notion of accessibility in reference theory, see in particular Ariel 1990 and 1994; and Fretheim and Gundel 1996.

[ back ] 37. Null anaphor (or zero) means that no words are used to express the anaphoric reference. Some languages, such as ancient and modern Greek, exploit this possibility (technically defined as “pro-drop”—“pronoun dropping” languages), whereas some others do not (English, for example).

[ back ] 38. By “accented pronoun,” Cornish means a pronoun whose utterance is characterized by a high pitch level (in writing, it corresponds to capital letters). Accented pronouns may share a deictic value with demonstrative pronouns (for example, if I say in front of several people “I’d like to have HER in the group,” which may be accompanied by a hand gesture).

[ back ] 39. Cornish 1999:6–7. Different scales of accessibility-marking have been previously theorized in Gundel et al. 1993 and in Ariel 1990.

[ back ] 40. Fox 1987:160.

[ back ] 41. Marslen-Wilson et al. 1982:355. On referential expressions structuring discourse, see also Vonk et al. 1992.

[ back ] 42. Marslen-Wilson et al. 1982:372 and 371 respectively, for the quotations.

[ back ] 43. The first two points (distance to the last mention and disambiguation) are extensively explored in Givón 1983.

[ back ] 44. Fox 1987:168–170; 168 for the quotation. Cornish assigns a role in the transition to new discourse units to demonstrative pronouns: “Demonstrative pronouns and descriptions may also be used to signal transitions to new discourse units, by picking up referents which were the focus of attention in the immediately preceding discourse segment but which are about to be displaced by a new object of focus” (Cornish 1999:27).

[ back ] 45. Bolkestein and Van de Grift 1994:298. Bolkestein 2000 further explores the complexity of anaphora processing in Latin literary texts.

[ back ] 46. The theoretical source of such a division is Smith 2003. A systematic application of this framework to Virgil’s Aeneid is in Adema 2008; for applications to ancient Greek literature, see Allan 2007 and 2009.

[ back ] 47. Kroon 2009:121–123.

[ back ] 48. Lakoff 1974:347.

[ back ] 49. See Lakoff 1974:351–353 (352 for the example).

[ back ] 50. Daneš 1990:129.

[ back ] 51. As I pointed out in my Introduction, every time I quote a Greek text I will include basic information about who the speaker is and whom the speaker is addressing. By including this information, I make plain a cognitive operation—as elementary as it is necessary—that every reader undertakes in the reception of the words; namely, keeping track of the discourse participants. Considering the communicative context of any utterance at any point of the narration may give fresh impulse to the comprehension of the text.

[ back ] 52. See Luraghi 2004.

[ back ] 53. Generally speaking, Homeric third-person pronouns refer to individuals much more frequently than to objects. However, neuter forms appearing in the test-case texts will be considered as well, because the cognitive activities deployed in the processing of the pronouns are the same.

[ back ] 54. A contextual frame is “a mental store of information about the current context, built up from the text itself and from inferences made from the text” (Emmott 1997:121).

[ back ] 55. Emmott 1997:125.

[ back ] 56. Further uses of τῶν opening the line and referring to topics of narration/speech occur at Odyssey 3.101 (= 4.331) τῶν νῦν μοι μνῆσαι, καί μοι νημερτὲς ἐνίσπες “Remember, now, these things for me. And tell them truthfully” (Telemachus to Nestor); Odyssey 4.350 (= 17.141) τῶν οὐδέν τοι ἐγὼ κρύψω ἔπος οὐδ’ ἐπικεύσω “I won’t keep these things secret from you, and I won’t conceal the tale” (Menelaus to Telemachus).

[ back ] 57. The notion of “flow” is explained by Chafe in terms of “the dynamic quality of the movement of information into and out of both focal (active) and peripheral (semiactive) consciousness” (1994:30).

[ back ] 58. For the breaking-down of the distinction between ὁ, as demonstrative pronoun and definite article, and ὅ/ὅς relative pronouns in archaic Greek, see Bonifazi 2004c:61–66. On ἑ as simple anaphoric mark, see Petit 1999:12–17. The accented form ἕ is commonly taken as a reflexive mark; contra Puddu 2005:150–161. About possessive adjectives: we regularly interpret a possessive adjective by asking ourselves “whose?”, thus retrieving the mental representation of the referent to whom an item belongs.

[ back ] 59. Bonifazi 2004c, with relevant bibliography.

[ back ] 60. The uncertain borderline concerning early Greek ὁ (weak) demonstrative pronoun, ὁ third-person pronoun, ὁ definite article, and ὅ relative pronoun is pointed out by different grammars and dictionaries.

[ back ] 61. As I specified in my Introduction, I borrow the notion of discourse act from Hannay and Kroon. They refer to discourse acts as separate communicative steps, each of them involving a “separate registration in discourse memory” (2005:94). More discourse acts may be grouped in a discourse unit, the latter intended as a coherent sequence of more sentences having an overarching communicative function.

[ back ] 62. According to Ruijgh (1971:444), the original scope of γε coincided with single words (first pronouns, then nouns and verbs); only later did the particle’s scope widen to include whole subordinate clauses.

[ back ] 63. For a summary of previous literature on γε and a discussion of its communicative functions, see Wakker 1994:308–309. Already LSJ (s.v.) records that γε can be rendered (also) by an “emphasis in pronunciation.” I believe that the prosodic prominence conferred by γε to the single word(s) it modifies has one of the pragmatic functions that prosodic prominence has in modern languages. In writing, prosodic prominence is rendered by using capital letters. Prosodic prominence of a single word (or a group of words) may contrastively introduce a new topic (as in “What did Fred do? Fred ate the BEANS”), but it may also give emphasis to old information (as in “Who called? Pat said SHE called”). In both cases, the speaker conveys prosodic prominence. Of course, the context of communication helps. For example, the accent on “beans” in the former example might imply that while Fred ate the beans, somebody else ate another kind of food, or that Fred usually does not eat beans at all, or that the beans in question were poorly prepared, and so on. All these and other possible implied meanings are what is “relationally new.” On information structure and sentence intonation, see the chapter “Information Structure,” by J. K. Gundel and T. Fretheim in the Handbook of Pragmatics Online, from which the above cited examples come. An instance of γε signaling the prosodic prominence of a contrast of topic is Iliad 19.217–219 κρείσσων εἰς ἐμέθεν καὶ φέρτερος οὐκ ὀλίγον περ / ἔγχει, ἐγὼ δέ κε σεῖο νοήματι γε προβαλοίμην / πολλόν “with the spear you are better and stronger than I am, and not a little; but I, on the other hand, might pass over you in thinking promptly, and by far” (Odysseus to Achilles).

[ back ] 64. As we will see in ch. 3, the use of the reflexive third person—especially when it is not grammatically required—implies an emotional nearness of the speaker towards the third person in question (this is also the case in modern literary works).

[ back ] 65. On ancient Greek αὐτός as intensifier and on the related reading of αὐτῶν in this passage, see ch. 3, n56. S. West (in Heubeck et al. 1988–92, I:72) notes the exceptional position of the genitive before the possessive adjective at line 7, and connects this to two similar formulaic wordings at Iliad 4.409 (κεῖνοι δὲ σφετέρῃσιν ἀτασθαλίῃσιν ὄλοντο “those ones [our fathers], instead, died because of their recklessness,” uttered by Sthenelus with reference to the Seven that fought against Thebes) and at Odyssey 10.437 (τούτου γὰρ καὶ κεῖνοι ἀτασθαλίῃσιν ὄλοντο “because those ones [the companions that were in the Cyclops’ cave] also perished, due to the recklessness of this one [Odysseus],” uttered by Eurylochus to the companions turned from pigs into men again). Both parallels have in common with Odyssey 1.7 a (demonstrative) third-person pronoun heading the line, which has the function of drawing the attention of the recipient to the referents as the new focus.

[ back ] 66. On the IE poetic context of nostos as a way to get life and see light again, see Frame 1978. On the relevance of the nostos theme in this proem, see Bonifazi 2009c.

[ back ] 67. “What comes between the proem and the Muses’ continuous narrative is the minimum of exposition that is needed to mark the point at which she will begin” (Bassett 1923:347).

[ back ] 68. Bakker (1997b:111) identifies four possibilities: any character that is present “on the stage” is active (ὁ δέ-clause); any character that is copresent is near-active (ὁ δέ-clause + either name in the same unit or noun-epithet phrase in next unit; answering-formula + noun-epithet phrase in next unit); any character that is returning on the stage is semiactive (name with δέ + clause in next unit); any character that is appearing is inactive (name with δέ + clause in next unit; τοῦ/τῷ/τὸν δέ-clause + noun-epithet phrase in next unit). With respect to Bakker’s framework, my reading considers ὁ δέ-clauses as requiring more attention and activation in the mental representation of the character behind ὁ, because such clauses represent new discourse acts, which introduce new material to the consciousness of the recipient, and δέ is the mark for that.

[ back ] 69. Bakker 1997b:100.

[ back ] 70. As far as I know, the first linguist to argue this is Geurts (1999).

[ back ] 71. See N. Austin 1972 and Kahane 1994 on the semantically significant localization of vocative and nominative proper names in particular metrical positions.

[ back ] 72. On noun-epithet formulas as indexes for characters, as “epic code for the mythic entirety” of a hero, see J. M. Foley 1999:210.

[ back ] 73. For the notion of pragmatic competence, see the Introduction.

[ back ] 74. See Dimock 1956 and N. Austin 1972, with related bibliography. Well-known puns on “hate” and Odysseus’ name are 19.407 and 275; 1.62; and, specifically related to Poseidon’s hate, 5.340 and 423.

[ back ] 75. Kirk et al. 1985, I:323.

[ back ] 76. Ebbott 1999; see also Worman (2001a:24–25).

[ back ] 77. Kirk et al. 1985–1993, I:324.

[ back ] 78. Especially in classical prose (for example, in Plato and in Xenophon), αὐτός is frequently found close to κεῖνος, as pronouns sharing the same referent (on this phenomenon, see a later section in this chapter).

[ back ] 79. In the Scholia Vetera Helen’s speech is defined as κομματικός; see Erbse 1969–1988, I:430 (ad 403d).

[ back ] 80. Petit (1999:157–159) argues that the distinction between IE *se– and swe– corresponds to the distinction between anaphoric and reflexive pronouns. A close analysis of the Homeric occurrences of ἕ that shows that the accented version also might not convey any reflexivity is Puddu 2005:150–161.

[ back ] 81. In light of the research conducted by Henrichs on the verb φυλάσσειν in relationship to hero-cult practices (1993:166–167), I submit to the reader that φύλασσε in the passage discussed above could include a specific reference to guarding, as in worship contexts, which is in line with the lament-act expressed by περὶ κεῖνον ὀΐζυε.

[ back ] 82. On recognitional demonstratives, see Diessel 1999:105–106 and Himmelmann 1996:230–239. On ὁ pronouns, ὁ definite articles, and ὅς relative pronouns as recognitional in Pindar, see Bonifazi 2004a:405–406. On the “syntax of movement” of the ὁ δέ-clauses, see Bakker 1997b:62–71 and 92–111.

[ back ] 83. “Cataphoric” is here taken as the equivalent of “proleptic”; for Homeric poetry, see Monro 1891:216–218 and Chantraine 1942–1953, II:169. Bakker 1999 delves into the deictic and the dialogic functions of οὗτος. On deictic ὅδε and οὗτος in Pindar, see Bonifazi 2001:39–43 and 2004a:398–401.

[ back ] 84. Havers 1906:3–5.

[ back ] 85. Brugmann 1904. Already Apollonius Dyscolus contrasts the closeness conveyed by ὅδε and οὗτος and the remoteness conveyed by (ἐ)κεῖνος (See Apollonius Dyscolus Pronouns 21.11 and 57.10–12; Syntax. 136.13 in Uhlig 1910). Kühner and Gerth (1955, I:641) share the same view.

[ back ] 86. “In drama it [(ἐ)κεῖνος] refers to what is not present on the stage. … It is, in fact, mainly used to refer to something outside the situation of utterance” (Ruijgh 2006:159).

[ back ] 87. See Havers 1906:4–5; Kühner and Gerth 1955, I:650; Humbert 1972:33; LSJ s.v. 2; Smyth 1980:308. On κεῖνος used in praise contexts by the laudator within Pindaric poetry, see Bonifazi 2004b.

[ back ] 88. For “co-text,” see n31 above.

[ back ] 89. See also Smyth 1980:309.

[ back ] 90. White and Morgan 1892:68.

[ back ] 91. LSJ s.v. 4 (italics in the text). Bailly makes the same two examples and considers ἐκεῖνος as working as an adverb of place.

[ back ] 92. LSJ s.v. 3; see also Bailly, s.v. 4.

[ back ] 93. Bakker (2006a:100–101) translates, and comments upon, the following Herodotean passage: Πέμψαντα δὲ τὸν Κόλχων βασιλέα ἐς τὴν Ἑλλάδα κήρυκα αἰτέειν τε δίκας τῆς ἁρπαγῆς καὶ ἀπαιτέειν τὴν θυγατέρα· τοὺς δὲ ὑποκρίνασθαι ὡς οὐδὲ ἐκεῖνοι Ἰοῦς τῆς Ἀργείης ἔδοσάν σφι δίκας τῆς ἁρπαγῆς· οὐδὲ ὦν αὐτοὶ δώσειν ἐκείνοισι (Herodotus 1.2.3): “‘The king of the Colchians, they say, sent a messenger to Hellas to demand justice for the abduction and to ask back his daughter. The Greeks, it is held, answered him that they had not offered compensation either for the abduction of Io and of Argo. So, neither would they offer it to them ’ [Bakker’s emphasis]. The perspective of the Greeks is continued … as delivered from the point of view of the speakers; … The infinitive sentence blurs the line between speaking and thinking in the representation of an internal point of view: what counts is the motivation or intention for speech represented, not the verbatim repetition of the speech itself.”

[ back ] 94. Magnien 1922:157–158; see also Monro 1891:217; Chantraine 1942–1953, II:169–170; LfgrE s.v.; Wace and Stubbings 1967:138–139; Ebeling 1885:382; Autenrieth 1958:102. The new observations of De Jong (2001) will be discussed later.

[ back ] 95. Monro 1891:217; farness in thought is mentioned also by LSJ and by Smyth (1980:309).

[ back ] 96. Cf. Nestor’s tales about his own, Diomedes’, and Agamemnon’s nostoi in book three (130–198), Menelaus’ extensive tale about his nostos in book four, and the general mention of the nostos of the Greeks back from Troy by Phemius in book one (326–327) and by Nestor in book three (132–133).

[ back ] 97. Thalmann 1992:31–46; 45 for the quotation.

[ back ] 98. In Homeric language, the initial epsilon is quite uncommon (I recorded 184 instances of κεῖνος forms against 18 ἐκεῖνος forms. While κεῖνος is usual in epic and lyric (the aeolic variant being κῆνος), ἐκεῖνος is common in Herodotean and in Attic prose.

[ back ] 99. By “social,” I mean a collective appraisal, given by an entire community through the voice of an individual. “Emotional” captures one’s feelings from a single (inner) perspective. However, these two aspects may blend together.

[ back ] 100. De Jong 2001:25; the comment is about a previous keinos (1.163), which I will discuss below.

[ back ] 101. Cf. West (in Heubeck et al. 1988–92, I:104): “Telemachus continues to refer to his father rather obliquely,” as well as Goldhill 1991:38 (about 14.122) “‘that man’, keinon, is Eumaeus’ reference to Odysseus, which like Penelope’s and Athene’s use of keinon, refuses to name the master.”

[ back ] 102. See the use of κεῖνος referring to Odysseus at 1.199, ii 272 and 274 (Athena speaking), 3.123 and 222 (Nestor speaking), 4.157 (Pisistratus speaking), and 693 (Penelope speaking).

[ back ] 103. See my argument in a later section, about the positive or negative social attitudes expressed by means of (e)keinos.

[ back ] 104. Nagy 1999:xv (his italics).

[ back ] 105. Nagy 1996a:82n53; see also Nagy 1996b:50.

[ back ] 106. Nagy 1999:xviii and 22 (I am reporting Nagy’s translations of the passages).

[ back ] 107. Nagy 1999:xviii.

[ back ] 108. Intonation may greatly influence not only information structure but also, more specifically, anaphoric comprehesion; see, e.g., Fretheim 1996 and Mithun 1996.

[ back ] 109. On the cognitive relevance of νοέω, see ch. 3, n90 and ch. 4, n102.

[ back ] 110. For instances of γάρ near κεῖνος with visual relevance, cf. Odyssey 4.149, xiv 321; Pindar Olympian 6.25, Nemean 10.62, Isthmian 1.17. γάρ alone clearly conveys visual force in Sophocles Electra 4, and in Alcman fr. 3.45 Calame.

[ back ] 111. N. Austin (1972:18) confirms the importance of this moment: “At Sparta the incantation continues. In a miniature anticipation of Odysseus’ identification scene in Scheria, there is a moment of extraordinary finesse when both Telemachus and Odysseus are revealed in a single image. … A tale casually begun becomes an incantation of both father and son. It concludes with the name of Odysseus before Menelaos’s eyes, but reincarnated in the son who conceals his grief as his father will do in Scheria.” The parallel between Telemachus’ and Odysseus’ disguise and recognition is also underscored by Fenik (1974:25–26).

[ back ] 112. In the next section, further passages showing the visual relationship established by the speaking “I” with the keinos-subject will be quoted.

[ back ] 115. When the Ithacan prophet Halitherses foreshadows Odysseus’ imminent homecoming (2.163–176), he calls him both by name (163 and 173) and by κεῖνος (171); likewise, Eurymachus later replies to Halitherses by using both “Odysseus” (182) and κεῖνος (183). It is notable that in the latter case the antecedent of the anaphoric expression occurs just one line before. The grammatical setting at 2.351 is even more significant, as the referent even follows the utterance of keinos: Telemachus tells Eurykleia: “Nurse, come and draw some wine in the two-handled jars, some sweet wine, the finest one you are preserving after the one you keep as you think of that one, hapless [κεῖνον … τόν κάμμορον], if Odysseus, Zeus-born [διογενὴς Ὀδυσεύς], should come from somewhere, after escaping death and the goddesses of doom” (2.349–352). [ back ] I add here that Eurymachus’ reply foregrounds the thematic link between keinos and Odysseus’ supposed death: “[Odysseus] is dead, far away [ὤλετο τῆλ’]; how I wish you had died with that one [κεῖνῳ] as well” (2.183–184); the same link seems to be suggested by κεῖνον … τόν κάμμορον (2.351), uttered by Telemachus.

[ back ] 116. This exemplifies what Wakker says about κε constructing “an imaginary alternative to the actual course of events” (1994:212–213).

[ back ] 117. An “explicit performative verb” at the head of a statement “shows how the ‘statement’ is to be fitted into the context of conversation, interlocution, dialogue, or in general of exposition” (J. L. Austin 1962:85). Examples given include: “I argue …,” “I conclude …,” “I admit …,” “I prophesy ….”

[ back ] 118. I borrow the term “runover” from M. Clark: “A runover occurs whenever a single word or a short phrase within a clause crosses over the boundary of the line” (1997:28).

[ back ] 119. I owe the latter observation to David Elmer (personal communication).

[ back ] 120. Such “impressive asyndeton” (see S. West in Heubeck et al. 1988–92, I:106) reminds one of similar constructions elsewhere: in the Homeric Hymn II to Demeter, Demeter herself is said to be ἀγέλαστος ἄπαστος “without enjoying any laughter or any nourishment,” while grieving over Persephone (line 200); in Odyssey 4.788, Penelope is laying down ἄσιτος ἄπαστος “without food, without nourishment,” as she is very worried about her son’s destiny; in Iliad 9.63–64 Nestor asserts that the man (literally “that one,” ἐκεῖνος) who longs for fighting with his own people is ἀφρήτωρ ἀθέμιστος ἀνέστιος “clan-missing, custom-missing, home-missing.”

[ back ] 121. One of the scholia at 242 says: ἀπὸ ἄλλης ἀρχῆς τὸ “οἴχετ’ ἄϊστος” κατὰ ἀναφώνησιν, “based on a different beginning, ‘οἴχετ’ ἄϊστος’ was performed by means of emphatic modulation of the voice” (Dindorf 1962:44). According to Stroh 2003, “anaphōnēsis,” first defined by Plutarch (Brutus 24.7.4 and Moralia 1071c) and by Aretaeus ( De curatione diuturnorum morborum 2.13.10), did not correspond to the generic Latin term declamatio; rather, it indicated a specific voice exercise that dealt with voice modulation in the performance of poetry; its corresponding Latin term might have been intentio vocis, as used by Seneca (Epistulae 2.15.7), by Pliny (Natural History 28.14[53]), and by Quintilian (Institutio oratoria 1.10.25).

[ back ] 122. Telemachus laments his father’s supposed death and also calls him keinos at 2.351, 3.88 (κείνου δ’ αὖ καὶ ὄλεθρον ἀπευθέα θῆκε Κρονίων “of that one, instead, Cronus’ son has made the death something impossible to know about”), and 3.93 = 4.323 (κείνου λυγρὸν ὄλεθρον ἐνισπεῖν “[in case you would like] to narrate the mournful ruin of that one”).

[ back ] 123. The atmosphere of “sorrowful reminiscence and melancholy” of this passage has been remarked upon by Fenik (1974:22–25).

[ back ] 124. Thus, the unstable variety of solutions presented in English translations of keinos in the Iliad and in the Odyssey (by a regular third-person pronoun, or by proximal demonstrative “this,” or by no pronoun at all, and never by “that one”) do not always help in tracking of the right referent. For example, in book iii, during the exchange between Telemachus, Nestor, and Athena-Mentor, there is an alternation between Odysseus and Agamemnon as the subjects of contrasting nostoi. At 3.218–222, Nestor speaks about Odysseus (via κεῖνῳ, 222); at 234–238, Athena-Mentor talks about Agamemnon (via proper name, ’Αγαμέμνων, 234). At 240–242, Telemachus says “Mentor, let’s speak no longer about these things, however distressed. For that one, nostos is not actual any more [κείνῳ δ’ οὐκέτι νόστος ἐτήτυμος, 241]; I mean, the immortals must have devised death and black doom for him.” Then, at 248 Telemachus again asks Nestor to tell more about Agamemnon’s death (note the patronymic + epithet + name structure ’Ατρεΐδης εὐρυκρείων ’Αγαμέμνων). The referent of κείνῳ is undoubtedly Odysseus, not only because of the corresponding κεῖνῳ at 222 but also because Telemachus’ further questions about Agamemnon’s story are clearly introduced as a different topic: “But now I want to inquire about the rest of the story, and ask about Nestor” (νῦν δ’ἐθέλω ἔπος ἄλλο μεταλλῆσαι καὶ ἐρέσθαι / Νεστορ’, 243–244). Hence, any translation of κεῖνῳ … νόστος as “his homecoming” is misleading, as it makes the reference indeterminate between Agamemnon (just mentioned) and Odysseus.

[ back ] 125. About visualization and visibility: I anticipate here my connection of the visual aspect in keinos with the element κε of the word; I also argue that the aspect of potential vision is what basically characterizes the Homeric use of particle κε in numerous cases (see ch. 2).

[ back ] 126. See above, pp. 42–43.

[ back ] 127. Further instances of adjacent keinos and “I”: Odyssey 11.390 (Odysseus says about Agamemnon’s soul ἔγνω δ’αἶψ’ ἐμὲ κεῖνος, ἐπεὶ ἴδεν ὀφθαλμοῖσιν “That one immediately recognized me, as he saw [me] with his eyes” (very similar words occur at xi 615); Odyssey 12.258 (Odysseus sees some companions being eaten by Skylla and says οἴκτιστον δὴ κεῖνο ἐμοῖσ’ ἴδον ὀφθαλμοῖσι (“that thing was really the most heartbreaking one I saw with my eyes”); Iliad 6.284, where the “I” mark is in the first-person verb (εἰ κεῖνόν γε ἴδοιμι κατελθόντ’ Ἄϊδος εἴσω “if I could see that man descending into Hades”); see also Iliad 1.271; 7.77; 9.312 and 646; 10.126; 15.45; Odyssey 1.212; 6.166; 11.522 and 528 (the last four passages include the “I” mark in a verbum videndi, as in Iliad 6.284); Odyssey 16.103; 17.110 and 112; 23.76; Pindar Olympian 1.100–101; Pythian 1.42–43; Nemean 3.11. More on visual uses of keinos in Pindar in Bonifazi 2004b:287–292. In Pindaric epinician poetry, keinos very often accomplishes the special effect of making far things suddenly close to the performer and to the audience. Heroes and (more rarely) objects of the past become potentially available to the audience’s eyes; likewise, victors and (more rarely) objects of the present are visually pointed out and put in relationship with the ongoing performance. For adjacent keinos and “I” in Euripides, see Electra 1019–1020; Troades 1045; Iphigenia in Tauris 229; Andromache 454; Alcestis 529; Hecuba 830; Hippolytus 321. More instances of keinos involved with visualization occur at Odyssey 1.163 and 209; 3.222; 4.149; 11.418.

[ back ] 128. Here κεῖνος does not seem used as a cataphoric correlative (ἐκεῖνα … ὧν), as the relative clause is a non-restrictive one.

[ back ] 129. See Euripides Troades 487–488, where Hecuba is lamenting over her own daughters, who have been snatched out of her hands from the enemies: κοὔτ’ ἐξ ἐκείνων ἐλπὶς ὡς ὀφθήσομαι / αὐτή τ’ ἐκείνας οὐκέτ’ ὄψομαί ποτε “And there is no hope that I am able to be seen from those ones [in the future], nor that I myself am able to see them any more.” Noticeably, in this passage the visual component overlaps with the desired eye-contact and the (reasonable) supposition of the ruin/death of the girls. Prose seems to employ visual (e)keinos as well. More work should be done on this; for the moment, I record a couple of examples. Xenophon in Hellenica III 4.18 says ἐπερρώσθη δ’ ἄν τις κἀκεῖνο ἰδών “one might/could have recovered strength having seen that” and then explains what “that” refers to—namely, the scene of Agesilaus triumphantly going to offer garlands to Artemis along with his soldiers. See also Thucydides ΙΙΙ 16.4 ἐπειδὴ καὶ ἐκείνους εἶδον “when they [the Athenians] saw those [the Peloponnesians].” Plutarch in Brutus (23.3–4) describes the moving scene of a painting showing Andromache’s farewell to Hector and tells about the woman κομιζομένης παρ’ αὐτοῦ τὸ παιδίον, ἐκείνῳ δὲ προσβλεπούσης “as she was taking from his arms their little son, while her eyes were fixed upon her husband” (tr. Perrin 1918). Interestingly, in her analysis of the different discourse functions of Latin ille, Kroon (2009:121) argues that sometimes ille marks a perspective shift, which may involve “camera” movements. In some cases in particular, “remoteness in terms of perspective has to do with pulling an essentially activated entity … from a relatively remote attentional position to a closer position, which automatically pushes the previous entity in focus out of perspective” (italics on the text). I see a similarity between this process described by Kroon and the visualization of the keinos subject appearing close to the eyes of the speaking “I”.

[ back ] 130. The seven Iliadic occurrences of keinos with Achilles as the referent are not comparable with the conscious cross-referencing keinos used throughout the Odyssey. Here, I invoke the already cited Homeric passages in which the reference to what is physically “there” or “over there” is within reach of the speaker’s eyes: Iliad 3.391, where Aphrodite points at Alexander in his appearance by saying κεῖνος ὅ γ’ ἐν θαλάμῳ καὶ δινωτοῖσι λέχεσσι “Look at him! There he is, in the room and in the carved bed;” Odyssey 18.239–240 νῦν Ἶρος κεῖνος / ἧσται “now Iros sits there” (Telemachus is speaking); Odyssey 22.165–166, where Eumaeus realizes that Melanthius is reaching the chamber and tells Odysseus κεῖνος δὴ αὖτ’ ἀΐδηλος ἀνήρ, … / ἔρχεται ἐς θάλαμον “There he is, that destructive man, … he is going to the chamber;” Iliad 5.604, where Diomedes sees Ares helping Hector and says to the companions καὶ νῦν οἱ πάρα κεῖνος Ἄρης, βροτῷ ἀνδρὶ ἐοικώς “And now there he is: Ares is beside him, looking like a man.” See also Euripides Phoenissae 159–161, where the old servant shows to Antigone where Polyneices is: ἐκεῖνος ἑπτὰ παρθένων τάφου πέλας / Νιόβης Ἀδράστῳ πλησίον παραστατεῖ / ὁρᾷς; “There he is, standing close to Adrastos, by the tomb of Niobe’s seven virgins”). In Aristophanes, the usual way to point at present items “over there” is ἐκεῖνος + deictic particle -ι (see Aristophanes Birds 297 [ΠΙ.] οὑτοσὶ πέρδιξ [ΕΥ.] ἐκεινοσὶ δὲ νὴ Δί’ ἀτταγᾶς “[Pisth.] This is a partridge. [Euelp.] That one, yea, by Zeus, is a francolin”).

[ back ] 131. See Bühler 1965:121–140; Bosch (1983:9–12); as for deixis am Phantasma in ancient Greek texts, see Bakker 2005:155–173; Calame 2004; Bonifazi 2004a:399–400; L. Edmunds 2008.

[ back ] 132. Such a cult place (situated in Pagasae, the Thessalian port whence the Argonauts departed) is mentioned also by the Scholia at Apollonius Rhodius (I 238); for this epithet of Apollo, see Etymologicum Magnum 646,39 and Hesychius s.v. Παγασίτης.

[ back ] 133. The epiphanic character of keinos in this incipit has been pointed out by Nagy (1990a:201n10), who sees this use also in Mimnermus fr. 14.1–3 οὐ μὲν δὴ κείνου γε μένος … / τοῖον ἐμέο προτέρων πεύθομαι, οἵ μιν ἴδον / … πυκινὰς κλονέοντα φάλαγγας “The strength of that one was not [such as yours] … I am getting information about him from my elders, who saw him … driving the ranks of the army in confusion”; see, too, Nagy 2001a:xxvii n20.

[ back ] 134. See Bonifazi 2004b:294–295.

[ back ] 135. Thus, I connect κείνου at Odyssey 4.149 to an epiphanic moment concerning Menelaus’ mental vision of Odysseus through Telemachus’ body (see above, pp. 43–45). I interpret the occurrences of keinos at Iliad 5.604 (where Ares appears) and Odyssey 11.390 and 615 (where Agamemnon and Heracles establish eye-contact with Odysseus) as epiphanic as well. The pretended sudden visualization of either Ino or Agave behind disguised Pentheus is expressed by Dionysus as follows: αὐτὰς ἐκείνας εἰσορᾶν δοκῶ σ’ ὁρῶν “As I see you, it seems to me that I look at those ones, the real ones” (Euripides Bacchae 927).

[ back ] 136. Ruijgh (2006:160) commenting these words by Oedipus speaks about “present revelation.” Even in humorous contexts, ὅδ’ ἐκεῖνος retains its pragmatic force as a marker of an epiphanic moment: after the announcement of the chorus leader, Demos, “master of all Greece,” appears (rejuvenated) on the stage in Aristophanes Knights 1331: Ὅδ’ ἐκεῖνος ὁρᾶν τεττιγοφόρος, τἀρχαίῳ σχήματι λαμπρός “Look! Here he is, wearing a golden cicada, and shining in his old-world figure.”

[ back ] 137. Italian has a very common expression to indicate exactly the phenomenon of perceiving the appearance of something that was previously far away or absent or unseen: “Ecco!” or “Eccolo/eccola!,” the latter including the addition of the clitic pronoun for “him” or “her.” The Latin cognate of “ecco” is ecce, whose root –ce comes from IE *ke-, the same root from which ancient Greek κεῖνος comes. Historical linguistics seems to confirm not only that *ke– is the distinctive feature of κεῖνος (as *ke– + the pronominal root *eno-), but also that the *ke– component was originally supposed to add an “ego-proximity” (on the latter point, see Waanders 1997; and already Schwyzer 1939–1971, I:613 was recording: “(ἐ)κε- hatte ursprünglich nach Lat. ce– u.a. ‘Ich-Deixis’”). On the relationship between the etymology of κεῖνος and the visual component of its meaning, see Bonifazi 2004b. Interestingly, Hittite particles particles kāša und kāšma are commonly translated as “look here!,” “behold.” They are etymologically marked by the pronominal adverb kā, which is cognate of Greek ke-. (see Rieken 2007).

[ back ] 138. In Sophocles’ Antigone, a similar wording occurs: at line 384, the guard—who has just entered leading Antigone—says to the chorus ἥδ’ ἔστ’ ἐκείνη τοὔργον ἡ ’ξειργασμένη “There she is; she’s the one who did the work,” where ἐκείνη captures the visual appearance and the socially negative distance at once. Aristophanes—almost scoffingly—puts in Socrates’ mouth a similar phrase when Strepsiades’ son appears on the stage after the latter’s father has invoked his presence (ὅδ’ ἐκεῖνος ἀνήρ, Clouds 1167).

[ back ] 139. The actual moments of recognition involving Electra and Orestes in Sophocles’ Electra are characterized by the occurrence of κεῖνος. At 1178, Electra answers Orestes’ question about her identity: τόδ’ ἔστ’ ἐκεῖνο, καὶ μάλ’ ἀθλίως ἔχον “Here it is [my appearance], even though it is doing miserably.” Later, at 1222 she exclaims, ἦ γὰρ σὺ κεῖνος; “Are you really that one?” The latter provides an intriguing connection with a previous κεῖνος, uttered by Orestes (1118), τόδ’ ἄγγος ἴσθι σῶμα τοὐκείνου στέγον “know that this here is the urn of his body,“ which encapsulates both the cognitive sense of absence—by Electra who supposes that Orestes is dead—and the perception of his actual presence quite nearby, not just through the urn of his ashes, but in person—by the omniscient audience.

[ back ] 140. On orienting as prevailing over localizing in Pindar, see Bonifazi 2004a:413–414.

[ back ] 141. So Gudeman (1934:34 and 119) in the light of the similar passage in Rhetorica (1371b ἀλλὰ συλλογισμός ἐστιν ὅτι τοῦτο ἐκεῖνo).

[ back ] 142. As for tragedy, cf. Sophocles Electra 1115 (where the verb is δέρκομαι) and 1178; Euripides Ion 554.

[ back ] 143. See also fr. 77.4.

[ back ] 144. Cf. the already mentioned references occurring at Odyssey xi 390 and 615, where Agamemnon and Heracles are recalled as κεῖνος.

[ back ] 145. Protesilaus is pronominalized as ἐκεῖνος in On Heroes 662.25. The use of ἐκεῖνος to indicate dead individuals is also recorded by Kühner and Gerth (1955, I:650).

[ back ] 146. At Odyssey 24.17–19, the ghost of Ajax is labeled as keinos within three lines that look very much like a lament embedded in the epic narration: Αἴαντός θ’, ὃς ἄριστος ἔην εἶδός τε δέμας τε / τῶν ἄλλων Δαναῶν μετ’ ἀμύμονα Πηλεΐωνα. / ὣς οἱ μὲν περὶ κεῖνον ὁμίλεον … “the ghost of Ajax, who was the best for beauty and body, of all the Danaans, together with the blameless son of Peleus. So they assembled around that one … .” Note, as well, that the striking sequence of “st” sounds at line 17 recalls Odyssey 1.242, commented on above. In Iliad 24, the dead Hector is said to be keinos twice, at lines 412–413, by Hermes talking to Priam (ἀλλ’ ἔτι κεῖνος κεῖται ’Αχιλλῆος παρὰ νηῒ / αὔτως ἐν κλισίῃσι “rather, that one still lies by Achilles’ ship, in the shelters, all alone”), and at lines 243–244, by Priam himself addressing the Trojans (ῥηΐτεροι γὰρ μᾶλλον ’Αχαιοῖσιν δὴ ἔσεσθε / κείνου τεθνηῶτος ἐναιρέμεν “you will clearly be easier for the Achaeans to slay since that one is dead.”) In Sophocles’ Ajax (437), Ajax himself recalls his dead father as κεῖνος; in Philoctetes (415), Neoptolemus tells Philoctetes about Ajax’s death by saying ὡς μηκέτ’ ὄντα κεῖνον ἐν φάει νόει “think that that one does not see the light of life any longer.” Pindar uses κεῖνος to praise the ancestors of the victor (Nemean 4.85–86). In Sophocles’ Oedipus Tyrannus (139), Oedipus calls his father Laius ἐκεῖνος; see, also, Euripides Ion 1008 κείνου δὲ κατθανόντος “when that one [Erechtheus, Creusa’s father] died.” Significantly, Euripides makes Medea call her own children κεῖνους in an act of lament that is uttered before she kills them. Here, I report Kovacs’ translation, which exploits italics as a paralinguistic sign of prosodic and semantic prominence: “I do not care if I myself go into exile. It is their experience of misfortune I weep for” (κείνους δὲ κλαίω συμφορᾷ κεχρημένους, Medea 347). A parodic and exhilarating effect of this specific use of κεῖνος is in Aristophanes (Peace 648–656), where Trygaeus says to Hermes about Cleon (the “leather-seller,” cf. 648): “Stop! Stop, Hermes master, don’t speak but let that man [ἄνδρ’ ἐκεῖνον] remain beneath, where he is. For that man [ἐκεῖνος ἀνήρ] is not ours anymore, but he’s yours. Whatever you would call that one [ἐκεῖνον], whether he would be a knave, when he was alive, and also babbling, and also pettifogger, and also agitator, and also disturber with these names all together you will revile him, now that he is yours.”

[ back ] 147. See Alcman fr. 15; Pindar Isthmian 1.17; Olympian 1.101; Pythian 5.57; Euripides Hecuba 627–628, … κεῖνος ολβιώτατος, / ὅτῳ … “… Blessed is that man to whom … .” Αn inscription dating back to the sixth century BCE from Krissa, near Delphi, mentions the dedication of spits (δρα[χμὰ]ς is the supported reading) to Athena and Hera by Phanaristos “so that that one [κνος] too might get kleos aphthiton each time anew” (SIG 15.351). In Plato Republic 368a, Socrates refers to the father of his interlocutors Glaucon and Adeimantus as ἐκεῖνος ὁ ἀνήρ and mentions an elegy composed by Glaucon’s ἐραστής in which Glaucon and Adeimantus were praised for their excellence in battle. Here is the text: (367e) Καὶ ἐγὼ ἀκούσας, ἀεὶ μὲν δὴ τὴν φύσιν τοῦ τε Γλαύκωνος καὶ τοῦ Ἀδειμάντου ἠγάμην, ἀτὰρ οὖν καὶ τότε πάνυ γε ἥσθην καὶ εἶπον· Οὐ κακῶς εἰς ὑμᾶς, ὦ παῖδες ἐκείνου τοῦ ἀνδρός, τὴν ἀρχὴν τῶν ἐλεγείων ἐποίησεν ὁ Γλαύκωνος ἐραστής, εὐδοκιμήσαντας περὶ τὴν Μεγαροῖ μάχην, εἰπών—παῖδες Ἀρίστωνος, κλεινοῦ θεῖον γένος ἀνδρός “As I heard this—well, I always liked the nature of Glaucon and Adeimantus, but also at that time I was utterly glad and I said, ‘Not bad for you; “oh, you sons of that man,” in accord with the beginning of the elegies that Glaucon’s lover composed when you gained good reputation at the battle of Megara: “Sons of Ariston, god-like descendants of a renowned man.” ’ ” Οnce again the demonstrative conveys both famousness and praiseworthiness.

[ back ] 148. Homeric epic echoes the same phrase at Iliad 2.161 (εἵνεκα Ἑλένης) and 177, and at 9.339. Further negative attitudes towards different figures in lyric are expressed by means of keinos in Alcaeus fr. 70.6 and 72.7; Archilochus fr. 130.4, 176.1 and 200; Hipponax fr. 117.10; Stesichorus fr. 46.3. The Greek texts of lyric, iambic, and elegiac poetry in this monograph are quoted according to the following editions appearing in the TLG online: D. L. Page, Poetae Melici Graeci, Oxford 1962 (for Stesichorus, Simonides, Ibycus, Anacreon, and also Alcman, if not indicated otherwise); M. L. West, Iambi et elegi Graeci, Oxford 1972 (vol. 1 for Archilochus and Hipponax; vol. 2 for Mimnermus, Solon and Ananius); D. L. Page and E. Lobel, Poetarum Lesbiorum Fragmenta, Oxford 1955 (for Sappho and Alcaeus); D. Young, Theognis, Leipzig 1971 (for Theognis); H. Maehler, Pindari carmina cum fragmentis, Leipzig 1971 (for Pindar).

[ back ] 149. I have found very few exceptions to this phenomenon in the Homeric poems: at Iliad 2.37 and 482, 4.543 and 21.517, the primary speaking ‘I’ uses the temporal phrase ἤματι κείνῳ to indicate the shared knowledge (between him and the audience) of a well-known day (see De Jong 2004:234–236); these are, by the way, the only Homeric cases in which keinos does not refer to individuals. Αt Iliad 16.648, the narrator is reporting Zeus’ thought about the possibility that Hector might kill Patroclus (Ζεὺς … μερμηρίζων … καὶ κεῖνον … φαίδιμος Ἕκτωρ / χαλκῷ δῃώσῃ “Zeus … meditating on … whether glorious Hector should slay that one [Patroclus],” 647–650). At Odyssey 24.19, keinos refers to Ajax within a line that resembles an embedded collective lament (see above, n146); at Odyssey 13.111, the primary speaking ‘I’ refers to the divine access to the cave of the Nymphs by means of κείνῃ.

[ back ] 150. In comparison to the Odyssey, the distribution of the 36 masculine singular forms of keinos in the Iliad (many fewer instances than in the Odyssey) is much more indistinct, as far as the referents are concerned. 7 occurrences are for Achilles (19.4%), 5 for Zeus, 4 for Alexander/Paris, 3 for Hector, and so on. Achilles is therefore not a privileged subject for the keinos utterances and, more importantly, the underlying intentions of those who call him keinos seem to be quite different. A common element is, of course, the reference to Achilles’ absence from the battlefield (charged with anger at him; see Iliad 5.790 and 14.368).

[ back ] 151. After the publication of the seminal work by Brown and Levinson (1987), politeness has progressively become a subfield of pragmatics. An application of Brown and Levinson’s frameworks to ancient Greek epic is in Lloyd 2004; see also Brown 2006.

[ back ] 152. See, for example, Nemean 5.22, 8.23; Pythian 4.289.

[ back ] 153. See Ajax 28 (Odysseus speaking), 271, 275, 285 (Tecmessa speaking), 755 and 795 (the messenger speaking). Of course, all of that reflects the strong cognitive presence of Ajax in the characters’ minds, which culminates in the veneration of the hero as a corpse. On the centrality of Ajax’s body in this theatrical piece, see Worman 2001b; on Sophoclean reflexes of Ajax’ s hero cult, see Henrichs 1993.

[ back ] 154. See Cicero Tusculan Disputations 1.42.

[ back ] 155. Cf. Smyth 1980:309, where the phenomenon is simply recorded, without any explanation being given; a generic principle of variance is invoked. Janko (1985:10–11) notes the insufficiency of such an account.

[ back ] 156. I will discuss the processing of αὐτός along with its semantic and pragmatic implications in ch. 3. Janko 1985 is entirely devoted to adjacent uses of αὐτὸς ἐκεῖνος that refer to the same entity, from Homer to Libanius. The author claims that αὐτός juxtaposed with ἐκεῖνος is an idiomatic expression, which is “notably rare in poetry of all periods” and which, distributionally, suggests a colloquialism (1985:28). Among the uses analyzed by Janko, there are some elements that have been and will be highlighted in my own investigation of Homeric uses of αὐτός and ἐκεῖνος: the “pseudo-reflexive usage,” the function of “contrastive anaphora,” the reference to “self-identification” in the nominative case, and Plato’s frequent reference to “a remote or authoritative person” (Janko 1985:22, 24, 24–25, and 26 respectively).

[ back ] 157. See Bonifazi 2009b.