1. INTRODUCTION
This article argues for a reinterpretation of the Niobe allusion of Iliad 24 (599–620) based on the hitherto overlooked significance of the ‘epic τε’ + aorist indicative in line 602. General practice in commentaries and translations has been to read the line contrary to the ordinary rules of Homeric grammar, interpreting the verb as a single past event rather than a timeless or ‘gnomic’ aorist (the regular interpretation with τε). This approach has led some to puzzle over what the τε is doing here and to devise various implausible and ad hoc explanations for it. My solution is to take this τε at face value and read it in its regular generalizing sense. I establish by examining parallel passages that there is no basis for interpreting this τε in any other way, nor, as some have done, for simply ignoring it.
Reinterpreting the text in this way allows us to solve a problem in the passage noticed since antiquity. The allusion is plainly a ring composition, but its structure is thrown off by lines 614–17. Rejecting the lines as inauthentic brings down an unduly heavy editorial hand, but other suggestions fail to repair the ring structure. As I point out, however, these lines contain a present-tense verb in reference to the eternal existence of Niobe as a weeping rock—just the thing to answer the timeless aorist in line 602. The ring structure can thus be saved without any editorial intervention, and the lines may be regarded as genuine, all by reading epic τε as epic τε.
But this reinterpretation does more than remedy a minor grammatical quirk and save an otherwise out-of-balance ring structure. It reframes our understanding of the passage and Achilles’ motivation for alluding to Niobe when he does and in the way that he does. By reading the introductory verb as a timeless aorist, the grieving of Niobe is presented, from the beginning, as an eternal consequence of a grieving process gone wrong—one that is out of step with established norms that typically allow people to move past their grief when they have lost a loved one. Being denied the ability to eat food—an essential step in the Homeric grieving process—has had devastating consequences for Niobe, as she can never move past her all-consuming grief at the loss of her children, and she is doomed to gnaw at her own sorrows instead. If Priam would like to avoid a similar outcome, Achilles suggests, he must eat, and he must do so right here and now, despite his reservations, behind enemy lines and in the company of Achilles himself.
In this way, an innocent-looking particle turns out to hold the key to a better understanding of the entire passage, solving an age-old structural puzzle along the way. In what follows I provide the relevant passage (§1.1), then introduce its problems more fully and my proposed solutions (§1.2) and provide a road map for the rest of the paper (§1.3).
1.1. Passage containing the Niobe allusion
I give the passage in full in (1) below, to be referenced throughout the article. All translations are mine. The Greek text is West’s, except that iotas adscript have been converted to iotas subscript.Footnote 1 I have offset certain portions of the text in (1) to visually reflect the structure of the passage as analysed below. Verbs in the aorist and imperfect indicative are emboldened, as is the particle τε. Verbs argued to have timeless (‘gnomic’) reference, translated by the present tense in English, are underlined. The verb of principal concern is the aorist indicative ἐμνήσατο, tentatively rendered as past tense ‘thought’, following the consensus interpretation of translations and commentaries up to now,Footnote 2 though I argue for a revised interpretation as presential ‘thinks’. The verb ἐμνήσατο is accordingly underlined in (1), as well as emboldened.
1.2. Problematization and proposed solutions
Of the numerous peculiarities in Achilles’ recounting of the Niobe myth to Priam, lines 614–17 have stood out since antiquity as potentially spurious.Footnote 3 Among other difficulties, these lines interrupt the ring structure of the passage, which RichardsonFootnote 4 lays out as in (2), here slightly modified for presentation.
Richardson explains the apparent intrusion of lines 614–17 as providing ‘a parallel with the conclusion of Akhilleus’ speech (lines 619–20), for just as Niobe continues to mourn her children, so will Priam lament his son on his return to Troy’.Footnote 6 While not implausible, this suggestion does nothing to remedy the ring structure, which would thus have three ‘A’ elements, with the second intruding between C2 and B2.
BrüggerFootnote 7 follows Richardson’s view fairly closely, though he assigns a numeral, rather than a letter, to each portion of the primary ring structure (his 1, 2, 3 = A1, B1, C1 in [2] above), answered by those numerals prime (his 3′, 2′, 1′ = C2, B2, A2 in [2] above). He makes Richardson’s explanation more explicit by treating line 613 as the beginning of a secondary ring structure, assigned the letter value A (Niobe ate), followed by 614–17 as B (Niobe’s continued weeping), and these are answered in turn by 618–19 as A′ (exhortation to eat) and 619–20 as B′ (Priam’s continued weeping). Thus, in Brügger’s representation, three of these passages are simultaneously lettered (A, A′, B′), referring to their role in the secondary ring structure, and numbered (3′, 2′, 1′), referring to their role in the primary ring structure (corresponding to C2, B2, A2 in [2] above). But this still does not resolve the problem of 614–17, to which he assigns the letter B only, without an accompanying numeral-prime value. Despite the added layer of complexity, lines 614–17 are still left floating, with no explicit connection to what comes before (A1, B1, C1), in which respect they differ from the other lettered portions of the passage (A, A′, B′ = 3′, 2′, 1′ = C2, B2, A2).
While I agree with these commentaries that lines 614–17 are not spurious and can be sensibly understood in the context of the ring structure of the passage, my proposal differs in that it views lines 614–17 as answering specifically to line 602 (C1), taking these together with the preceding line (613) under the umbrella of C2 (see [4] below). This has the benefit of anchoring lines 614–17 to the first part of the passage, rather than having to suppose that they are introduced only in order to look ahead to the ‘continued weeping’ of lines 619–20. This view is justified by particular verbal parallels between the two portions of the text, as I will now describe.
My proposal for salvaging the ring structure of (1) comes about by making sense of another—this time grammatical—oddity which occurs earlier in the passage, in line 602, repeated in (3).
Scholars have been at pains to explain the apparently aberrant use of the particle τε here,Footnote 8 which typically has a generalizing function, often used in general relative clauses, as in line 616 in (1) above (see also [11] in Section 3), or co-occurring with the present or timeless aorist found in aphorisms or epic similes (hence the particle is often called ‘epic τε’), as in (5) in Section 2 and (15), (18)–(19) in Section 3. But the context of (1) seems to call for a preterital interpretation of the aorist rather than a timeless or ‘gnomic’ interpretation, and scholars have unanimously interpreted it as such (Denniston,Footnote 9 Ruijgh,Footnote 10 ChantraineFootnote 11 ).
As I will show, however, past-referring aorists in such contexts do not occur with τε elsewhere in Homer. Moreover, given that gnomic aorists are virtually always augmented in epic (PlattFootnote 12 ), ἐμνήσατο is suited to timeless/gnomic interpretation (rendered by the English present tense), while the subsequent, augmentless preterites in the passage are past referring (rendered by the English past tense). I therefore read line 602 as a statement that still applies in the present: ‘For even Niobe thinks of food’, referring to her eternal sorrowing as a weeping rock.
The suspected lines 614–17 may then be understood in terms of the ring structure of the passage, taken together with line 613 in answer to line 602. While line 602 has an aorist with present reference (ἐμνήσατο σίτου ‘she thinks of food’), lines 613–17 resolve the timeless aorist into a past-referring aorist, σίτου μνήσατο ‘she thought of food’ (line 613), and a verb in the present tense, νῦν δέ … κήδεα πέσσει ‘even now she broods on [or, more precisely, stews on or chews on] her sorrows’. The timeless aorist, looking both backward and forward, is duly answered by a past and a present tense, relating Niobe’s episodic past to her eternal present. The food metaphor is carried through in the choice of πέσσει ‘she broods’ in line 617, as πέσσω means literally ‘cook, digest, process food’ and so metaphorically ‘stew on, chew on, process emotions’.Footnote 13 It is a deliberate food related pun recalling σίτου in lines 602 and 613.
I give my revision of Richardson’s schema in (4).
For my interpretation of πέσσω as ‘process (food/emotions)’ see the lexical study provided in the appendix at the end of this article.
1.3. Road map
The rest of this article is structured as follows: After exploring the narrative implications of my reinterpretation of the text (§2), I show how the grammatical facts of Homer support reading the aorist in line 602 as timeless (§3) in contrast to the other preterites of the passage, which are past referring (§4). I then summarize and conclude (§5).
2. WEEPING ROCKS (BUT FOOD IS BETTER)
The ‘real-life’ Niobe, as legend has it, can be seen to this day in the form of the Weeping Rock on Mount Sipylus. And she is still ‘weeping’: the natural formation of porous limestone appears to ‘weep’ after a rain as water seeps through it.
Fig. 1: Weeping Rock (Ağlayan Kaya), Mount Sipylus (Spil Dağı), Manisa Province, Turkey.Footnote 14
The notion of a weeping rock even has precedent within the Iliad itself, in the epic simile in (5).
The same comparison is made in reference to Agamemnon’s tears at Il. 9.14–15.
Sipylus is explicitly mentioned in line 615 of (1) above, part of the suspected lines 614–17. Aside from these lines’ disruption of the ring structure of the passage, several objections were put forth against them in antiquity,Footnote 15 of which two are worth considering here, presented in (6).
Against (6a) I argue that a rock’s inability to eat is precisely the point of the passage; against (6b) that the passage is not so much a consolation as an admonition. Both points hinge on the fact that we are not actually told that Niobe eats, only that she ‘thought of food’.
Homer’s is the only extant version of the Niobe myth in which her thinking about food is mentioned, and Achilles clearly means for Priam to draw a parallel not only between Niobe’s grief and his own but also between his own refusal of food and Niobe’s failure to eat in her time of mourning. Richardson rightly notes that Niobe’s petrification is not a punishment from Zeus but a necessary consequence of her unresolved grief.Footnote 16 The reason for this outcome, I suggest, is that she cannot properly process her sorrows.
As I will argue in what follows, the prescribed Homeric procedure for moving on from mourning includes: (i) eating after a grief-induced fast and (ii) conducting the proper funeral rites of the deceased.Footnote 17 But Niobe cannot carry out her children’s funeral rites, and she cannot eat. She can therefore never move past the ‘brooding’ stage of grief, which Priam is himself currently in, according to his own declaration in line 639 (see discussion below and cf. [25] in the appendix).
It seems, therefore, that part of Achilles’ point in relating this narrative to Priam is to explain why, in enemy territory, in the home of the most dangerous of all Greeks, the best thing for Priam to do is to sit down to a meal, rather than immediately depart as he had sensibly requested to do (24.552–7). Achilles himself has just eaten (24.475–6), so his invitation is not motivated by his own hunger but rather by his own advancement in his grieving for Patroclus.
An important step in the processing of grief in Homer, particularly after a prolonged fast, is the consumption of food. Achilles’ relentless fasting in Book 19 (see especially lines 228–31, 303–21 and 340–55) has only lately concluded (24.475–6) after an exhortation by his mother to eat, sleep and lie with a woman, in (7).
Achilles had made his own exhortation to feast during Patroclus’ funeral, in (8).
While the rest of the Achaeans are eating their meal, the funeral pyre serves to ‘eat up’ the body of Patroclus (πῦρ ἐσθίει) along with the Trojan victims and other sacrificial offerings. Achilles explicitly denies this right to Hector’s body, thereby also denying consolation to Priam, who is mentioned by name, in (9).Footnote 19
In the Odyssey, norms of eating and lamentation are reinforced by the words of Pisistratus in the house of Menelaus. In response to the hosts and guests weeping over their supper for Odysseus, presumed dead (Od. 4.184–5), Pisistratus says that there is nothing wrong with lamenting the dead, as long as it is not simultaneous with supper, so their crying should be delayed until the next morning (4.193–5). Menelaus agrees and urges everyone to ‘think again of supper’ (δόρπου δ’ ἐξαῦτις μνησώμεθα) and put aside weeping until the proper time (4.212–15).Footnote 20
In light of these parallels, we may return to our scene in Iliad 24. After Achilles has yielded to Priam his right to consolation by returning Hector’s body, Priam must observe due propriety in carrying out his mourning process, first by concluding his fast and taking food, and then by sleeping, before he can return home to complete Hector’s funeral rites. Only Achilles is said to have ‘taken his fill of lamentation’ (γόοιο τετάρπετο δῖος Ἀχιλλεύς, 24.513) when the two of them weep together in Achilles’ residence.
In this way, Priam’s consumption of food serves to alleviate his mourning rather than perpetuate it. It is not Niobe’s weeping, then, that is meant to be compared to Priam’s, pace Richardson, who takes lines 614–17 of (1) in connection with lines 619–20.Footnote 21 And, pace Schmitz,Footnote 22 Niobe is not presented ‘as a model for the grieving Priam to temporarily abandon his pain and turn to the meal’. Rather, Priam’s ability to consume food (successfully) is meant to be contrasted with Niobe’s lack of it, and by extension his proper mourning process is to be contrasted with her improper one. It is not that Niobe ‘ate and was petrified’ (per [6b] above) but that she did not eat and was petrified.
My interpretation is thus in direct opposition to the typical view of the passage, expressed most recently by BierlFootnote 23 that ‘Niobe … remembered to eat (613)—unlike Priam as well as to some extent Achilles’. Niobe ‘thought of food’, it is true, but this does not mean she did any eating. Indeed, her suffering is coextensive with her frustrated desire for food—both unending—such that she can only ‘chew on’ her sorrows to this day (νῦν δέ … κήδεα πέσσει). Priam in fact uses this same phrase himself in line 639 (κήδεα μυρία πέσσω, ‘I have been brooding on my countless sorrows’) in reference to his fasting and grieving up to now, which, he says, is alleviated in so far as he has now, at long last, taken food and drink (24.635–42, partially quoted in [25] in the appendix).
In his grief, Priam must take care not to suffer Niobe’s fate. He may accomplish this, according to Achilles, by observing the correct mourning procedure in consuming food, as Niobe cannot (and never could). Unlike Niobe, Priam may now cease from his grief-induced madness in order to conduct a proper burial for his son. Achilles meanwhile completes his own grieving process, as outlined by his mother (7) above, in laying with Briseis (24.675–6). Only at this point, when sleep has overtaken everyone, does Hermes come in the night to lead Priam back to Troy.
It is therefore an essential, even ritual part of Priam’s mourning process to take food. While his son will of course be ‘much lamented’ (πολυδάκρυτος), Priam’s weeping does not continue endlessly ‘even now’ (νῦν δέ) as Niobe’s does. Priam’s meal is not meant to be a temporary distraction from suffering; it is a means by which he may begin to move on to the next stages of his grieving process (including more weeping, but not an endless amount). Achilles thus urges Priam to emulate Niobe in thinking of food in order to avoid her fate. Whereas she can only think of food, he can actually eat; whereas she can only gnaw on her sorrows, he can actually process them and move on. Achilles’ allusion to Niobe is thus not merely a precedent for Priam’s emulation but also a warning, along the following lines.
Emulation: Even Niobe thought of food after the death of her children, and we should do likewise.
Warning: But she, unlike you, has no choice but to consume her own grief eternally, which you may avoid by undertaking the proper grieving procedure (eating).
Mythological allusions are not uncommonly made as warnings in Homer, such as the Lycurgus narrative at Il. 6.128–41, partially quoted in (12) in Section 3.
Crucially, the eternal nature of Niobe’s endless gnawing of her sorrows is anticipated, ring compositionally, by line 602 only if we read it (in accordance with the ordinary rules of Homeric grammar) as a timeless or ‘gnomic’ aorist. I turn now to this point.
3. EPIC TE
Homeric commentaries and grammars are generally at a loss to explain the use of τε in line 602 of (1). Naively, one might read it as an instance of generalizing τε (‘epic τε’), giving the aorist a timeless (presential) sense: ‘even Niobe thinks of food’. But scholars have unanimously reasoned that the τε here cannot have a generalizing sense because Niobe’s story is confined to the remote mythic past, not the eternal present. A couple of ad hoc explanations have been given. Denniston speculates, ‘Here, perhaps, a historic precedent is taken as equivalent to a general proposition’.Footnote 24 Though noting the oddity, Brügger endorses this view: ‘τε bei einer konkreten, quasi-historischen Schilderung ist auffällig und wird durch den mythologisch-generalisierenden Kontext erklärt’ (‘τε in a concrete, quasi-historical description is striking and is explained by the mythological-generalizing context’).Footnote 25 But is τε really explained by the ‘mythological-generalizing context’? If it were, we should expect to find parallel passages in Homer in which τε shows a similar function. But, as will be seen below, this is not the case.
It is true that the phrase καὶ γάρ (as we have in line 602) is fairly frequent as a means of introducing an exemplary precedent for what has been asserted,Footnote 26 especially a mythological one, as in (10).Footnote 27
Occasionally καί on its own is used to introduce such narratives, as in (11).Footnote 28
But none of these examples seems ‘equivalent to a general proposition’, as Denniston puts it (my italics), nor is it clear what the meaning of such an equivalence would be.
Chantraine treats line 602 in (1) under his section on the particle chain γάρ τε, with a subgroup of examples qualified as having ‘une valeur éventuelle et contingente’ (‘a potential and conditional value’),Footnote 29 comparing it to examples (16) and (17) discussed below, though neither of those examples contain an aorist indicative. He interprets the γάρ τε in line 602 with the following translation: ‘Niobé, elle-même, un jour a songé à manger’ (‘Even Niobe thought someday of eating’). It is difficult to understand exactly how Chantraine was conceptualizing this, however.
These explanations raise empirical questions to be investigated. First, following on Denniston’s explanation, we may ask whether there are any parallel passages in Homer in which τε is used where the mythic past is invoked as a historical precedent for the current situation. In particular, since such occasions are often introduced by καὶ γάρ, occurrences of the particle chain καὶ γάρ τε are predicted (if Denniston’s assessment is correct) to provide good parallels for the meaning of ‘a historic precedent … taken as equivalent to a general proposition’. In connection with Chantraine’s interpretation, we may ask whether the particle group γάρ τε is ever used in reference to a single (episodic) past event elsewhere in Homer, and so whether we have good parallels for a single-event reading in 602.
My investigation shows that neither Denniston’s nor Chantraine’s proposals are well founded, since there are no good parallels for their interpretations elsewhere in Homer. Whether alone or in conjunction with other particles, τε is not used in contexts where a concrete historical episode is invoked as a precedent (paradeigma) to be applied to a current situation.Footnote 30 All sentences containing καὶ γάρ τε are generic or gnomic (timeless) in meaning, never referring to a single, episodic event in the past. Those containing γάρ τε may be ‘contingent’ (with the conditional complementizer εἴ περ ‘even if’), but more importantly all of them involve a notion of repeated action, whether generic, iterative-habitual or gnomic (see respectively [16], [17] and [18]–[19] below). Hence the occurrence of (καὶ) γάρ τε in line 602 of (1) should be interpreted likewise, with a multiple-event reading rather than a single-event reading, since the latter turns out not to be supported by any Homeric parallel.
Let us first address the absence of τε in mythic or historical allusions (paradeigmata) in Homer and then treat its occurrence in the two relevant particle chains, καὶ γάρ τε and γάρ τε, in turn. The first question posed above may be answered categorically no: paradeigmata referring to single past episodes do not employ (non-connective) τε. Consider, for example, (12) and (13), where we find the formula οὐδὲ γὰρ οὐδέ. This formula is elsewhere used in existential perfect contexts, of the type ‘no one has ever done this before’, taking the lack of precedent as grounds for current unlikelihood.Footnote 31
In other cases we find the simile formula ὡς ὅτε ‘as when’ introducing a narrative about a single past occurrence with the (augmentless) aorist indicative, as in (14). Cf. similarly Nestor’s allusions to his youth, cited in n. 33 below. Unlike the generic similes, however, τε is not used (contrast [5] above).
In the negative we find οὐδ’ ὅτε ‘not even when’, as at Il. 3.343–4, in which Paris refers to his abduction of Helen (the verb is ἔπλεον ‘sailed’, an imperfect indicative).
We may also consider the extended historical narratives of Phoenix and Nestor. The tale of Meleager at Il. 9.524–99 is held up by Phoenix as evidence that men of the past used to accept gifts and be appeased when they were angry (lines 525–6), so Achilles should now do likewise. Generalizing τε is not used.Footnote 32 Similarly, the speeches of Nestor, contrasting the days of his youth with the current time, make no use of generalizing τε.Footnote 33 The lack of generalizing τε is particularly striking given that the narratives of both men contain imperfects with a habitual meaning.
By contrast, the clitic chains containing τε uniformly show a generalizing or multiple-event interpretation. We find καὶ γάρ (τίς) τε five times, besides the occurrence in our passage in (1), namely: Il. 1.63, 2.292, 9.502; Od. 19.265, 23.118. All five are gnomic in meaning, referring to timeless truths, always with a verb in the present indicative, as in (15).
Line 602 in (1) is the only instance of the aorist indicative with καὶ γάρ τε. This, coupled with the fact that all other instances of the particle chain involve presents with a gnomic interpretation, strongly supports a gnomic/timeless reading of ἐμνήσατο in (1).
The clitic chain γάρ τε occurs most often with verbs that have a multiple-event reading. In these cases, it ‘introduces an explanation that is based on a fact that occurs frequently or on a general notion’ (‘introduit une explication qui repose sur un fait qui se produit souvent ou sur une notion générale’), often with an ‘idea of repetition’ (‘idée de répétition’).Footnote 34
As mentioned above, however, τε occurs in a couple of passages whose context may, at first glance, be characterized as single-event, which Chaintraine describes as having ‘une valeur éventuelle et contingente’ (‘a potential and conditional value’).Footnote 35 These are given in (16) and (17), the former containing a future indicative and the latter a present subjunctive.Footnote 36
Denniston calls the occurrence in (16) ‘half-general’.Footnote 37 I am unsure what it means to be half-general and am content to say that it is fully general, with the future indicative expressing a predictable outcome, as in If one puts food in a fish tank, the fish will eat it—a general truth that is none the less expressed as a future, contingent on some antecedent event. Compare similarly (19) below, in which the future indicative τελεῖ ‘will fulfill’ is both future and gnomic, with generalizing τε. In (16), the future event is contingent on Agamemnon giving the command, but it is bound to happen if he does so. This example does not therefore differ in kind from the ‘notion générale’Footnote 38 that is typically expressed by sentences with γάρ τε.
Similarly, (17) is not only conditional but also probably iterative in the sense ‘if we keep on getting killed’. If so, Chantraine’s ‘idée de répétition’ (‘idea of repetition’) is seen in this example as well, which therefore does not need to be considered as fundamentally dissimilar to the other instances of γάρ τε.
We do, in fact, find some cases of γάρ τε with the aorist indicative aside from line 602 in (1). As expected, these are consistently gnomic/timeless in meaning, as in (18).
The occurrence in (19) is both contingent and gnomic (so KirkFootnote 39 ).
Cf. (28) in the appendix, where εἴ περ γάρ τε is followed by the aorist subjunctive in a similarly generalizing sense.
In light of this survey, our passage in (1) would be the only instance of γάρ τε + aorist indicative that refers to a concrete single event in the past, if it were so interpreted. But the parallels speak univocally against this, supporting instead an interpretation of line 602 as a general, timeless statement about the eternal reality of Niobe, rather than about her action in the remote past.
Moreover, the aorist in gnomes and similes is regularly augmented in Homer.Footnote 40 In (1), ἐμνήσατο is the only augmented aorist besides ἐρρώσαντο in 616, which is plainly timeless, occurring in a general relative clause with τε. There is thus morphological support for reading ἐμνήσατο as timeless as well (see further §4 below).
Perhaps, then, the naïve reading is not so naïve. Perhaps the τε in line 602 is a generalizing τε after all, and ἐμνήσατο should be understood to have a timeless (presential) interpretation. I thus suggest a reinterpretation of the passage as per (20).
The idea is that Niobe thinks of food eternally, timelessly, even now. This notion, as I have said, is (ring-compositionally) expressed again in 614–17: νῦν δὲ … κήδεα πέσσει ‘even now she chews on her sorrows’.
Such a reading requires that the first—and only the first—of the aorists in the Niobe allusion be interpreted as timeless. All other aorists in the narrative portion of (1) must be understood as referring to the remote (mythic) past. The text in fact provides grammatical support for this interpretation, and I turn now to this final point.
4. AUGMENT AVOIDANCE IN SEQUENTIAL NARRATION
Most finite verbs in the mythic narrative of (1) are aorist and all are augmentless except for ἦεν, for which no securely augmentless counterpart exists.Footnote 41 Further, in all cases the lack of augment is metrically assured. The consistency of this lack of augmentation is unusual and conspicuous. Take πέφνεν ‘killed’, for instance, which is typically augmented (31x in Homer) in the sense ‘(has) killed’. It is augmentless only in (1) (line 605) and in five other places in Homer (16%). Augmentless preterites are known to be preferred in narrativesFootnote 42 but seldom appear with such consistency and metrical security as we find in (1). Contrast the variability in augmentation found in the narratives of (10), (13) and (14) above.
I propose that the avoidance of augmented forms in the narrative portion of (1) maximizes contrast between these and the eternal/ongoing events of lines 602 (καὶ γάρ τ’ ἠΰκομος Νιόβη ἐμνήσατο σίτου) and 614–17 (νῦν δέ … κήδεα πέσσει). Most strikingly, augmentless μνήσατο in line 613 stands in contrast to the augmented ἐμνήσατο in line 602. And here form reflects function: ἐμνήσατο is ‘she thinks’, while μνήσατο is ‘she thought’. Recall Platt’s finding that gnomic/timeless aorists are virtually always augmented in Homer.Footnote 43 Whereas ἐμνήσατο introduces the allusion as a whole, which includes both the narrative portion (lines 603–13) and the νῦν δέ ‘even now’ portion (lines 614–17), μνήσατο falls within the narrative itself. The narrative ends by echoing the verb that introduced it, this time without the augment and referring to Niobe’s thinking event as it first occurred in the remote past. In its position at the end of the narrative and immediately preceding line 614 (νῦν δέ), the verb μνήσατο serves to modulate between the events of the mythic past and those that still hold in the present, formally recalling to mind the timeless aorist that begins the passage (ἐμνήσατο) at the very moment of transition between the remote and the ongoing.
By contrast, in (10), (11), (13) and (14) above the verb introducing the narrative is augmentless, referring to single mythic events, such as the deception of Zeus, which happened ‘once’ (ποτε) in the past.Footnote 44 Further, the narratives of (10)–(14) contain a fair mixture of augmented and augmentless forms. This is because, unlike the Niobe allusion, these narratives do not involve eternally ongoing events, only a succession of events that lie firmly in the remote past. As a result, these narratives can afford, without confusion of time reference, considerable admixture of augmented and augmentless verb forms.Footnote 45
But the Niobe allusion requires, as it were, a greater degree of grammatical care and precision than do these ‘single-episode’ type allusions, so as to avoid confusion of time reference and ensure that the remote narrative events are formally contrasted with the timeless ones. Scrupulous consistency of augmentation—no augment for verbs in sequential narration, augment only for gnomic/timeless aorists—serves to achieve this effect. The timeless (presential) value of the augmented aorist in line 602 of (1) is thus made clearer by the resolutely consistent avoidance of augmentation in the expression of the narrative events that follow.
Because line 602, corresponding to item C1 in the ring structure scheme of (2) above (§1.2), is not, as Richardson has it, ‘For even Niobe did so’ (i.e. ate)Footnote 46 but rather ‘For even Niobe does so’ (i.e. thinks of food), it cannot be resolved by 613 alone, as this would only include its past reference and not its eternally ongoing signification. I propose, therefore, that line 602 is resolved by both 613 (‘she thought of food’) and 614–17 (‘she continues to chew on her sorrows’) taken together. Thus, in 613–17, the timeless aorist of 602 is decomposed into a past event (613) with eternal consequences (614–17), and the ring structure of the passage becomes unremarkable, as I have represented in (4) above.
5. CONCLUSION
Far from disrupting the ring structure of (1), lines 614–17 in fact make sense of the apparently exceptional use of generalizing τε and the aorist indicative in line 602. This aorist must be understood as having a timeless (presential) value, rather than a preterital one. This slight reinterpretation relieves the passage of a supposed structural difficulty as well as a grammatical one. In effect, we have arrived at a better understanding of the narrative function of Achilles’ allusion to Niobe in terms of proper Homeric grieving procedure. Priam may proceed, having taken his meal, to carry out the funerary rites of his son, thereby concluding the Iliad, while Niobe is left to her eternal sorrow.
APPENDIX: LEXICAL STUDY OF ΠEΣΣΩ
πέσσω (πέπτω) is generally understood to mean basically ‘cook, bake; ripen; digest’ (PIE *pekʷ- ‘make ready for consumption, soften, process’, cf. Skt. pácati ‘cook, digest, ripen’; Lat. coquō ‘cook, roast, prepare food, ripen’; OCS pešti ‘bake; worry, care’; Rus. peč′ ‘bake, scorch’). Though once in the literal meaning ‘ripen, soften’ in the Odyssey (see [21] below), it is exclusively metaphorical in the Iliad (7x), always as a verb of ingestion/digestion in reference to things other than food, especially emotions. It thus corresponds well to various English idioms having to do with food preparation, digestion and cooking extended beyond actual food: ‘grind, mull over’, ‘digest, chew on, chew over’, ‘choke on, be choked up about’, ‘stew on, seethe, be steamed about, brood on’ and by extension ‘nurse’ (of a physical wound, see [26] below).Footnote 47
The word’s cognates in Anatolian have the sense ‘grind, crush’ (Hitt. pakkuške-), which may be represented (in a metaphorical sense) in the collocation χόλον θυμαλγέα πέσσω ‘mull over one’s heart-grieving rage’, in (22) and (23) below. In reference to wounds or sorrows it always refers to those that are not yet—or can never be—healed or remedied, in (24)–(26) below. It refers not to successful digestion but to unsuccessful digestion, hence ‘be choked up about, brood on’ or ‘choke on, brood over’, in (27) below. The only time this verb refers to an accomplished act of (metaphorical) eating or swallowing is when it has the telicizing prefix κατα- ‘down’, in (28) below. Yet even in this case the swallowing down is only temporary, and a lasting grudge endures, as lines 82–3 make clear.
All these senses can be captured nicely (with deliberate vagueness) by the English word ‘process’, applicable to emotional processing as well as food processing—whether by cooking, grinding or other preparation, or else by chewing, digestion or (over-)ripening/softening. I therefore propose that ‘process (food/emotions)’ is the basic sense of the word in Greek and the one we should reconstruct for the PIE root *pekʷ-.
I present below all the passages in Homer that contain a form of πέσσω. In the Iliad it occurs, interestingly, only in direct quotations. I name the speaker and addressee in each such case, enclosed in square brackets beneath the translation. The occurrence in the Odyssey (21) stands apart as being part of a narrative description, as well as being the only instance in Homer where the verb πέσσω means ‘ripen’ (or rather ‘over-ripen’) and refers to literal food.
Open access
