(Ebook) Epic Grief: Personal Laments in Homer's: Iliad by Christos Tsagalis ISBN 9783110896251, 3110896257
(Ebook) Epic Grief: Personal Laments in Homer's: Iliad by Christos Tsagalis ISBN 9783110896251, 3110896257
https://ebooknice.com/product/epic-grief-49061734
https://ebooknice.com/product/untitled-58598012
https://ebooknice.com/product/performing-grief-bridal-laments-in-
rural-china-1668864
https://ebooknice.com/product/the-iliad-webster-s-thesaurus-
edition-1935048
(Ebook) Inscribing Sorrow: Fourth-Century Attic Funerary Epigrams by
Christos C. Tsagalis ISBN 9783110201321, 3110201321
https://ebooknice.com/product/inscribing-sorrow-fourth-century-attic-
funerary-epigrams-1948134
https://ebooknice.com/product/homer-the-iliad-1714374
https://ebooknice.com/product/the-iliad-webster-s-chinese-simplified-
thesaurus-edition-2019362
https://ebooknice.com/product/the-iliad-a-new-translation-by-peter-
green-50599128
https://ebooknice.com/product/the-iliad-webster-s-spanish-thesaurus-
edition-1972760
Epic Grief Personal Laments in Homer s Iliad Christos
Tsagalis Digital Instant Download
Author(s): Christos Tsagalis
ISBN(s): 9783110896251, 3110896257
Edition: ebook
File Details: PDF, 3.41 MB
Year: 2012
Language: english
Christos Tsagalis
Epic Grief
wDE
G
Untersuchungen zur
antiken Literatur und Geschichte
Herausgegeben von
Gustav-Adolf Lehmann, Heinz-Günther Nesselrath
und Otto Zwierlein
Band 70
by
Chris to s Tsagalis
ISBN 3-11-017944-X
A CIP catalogue record for this book is available from the Library
of Congress
© Copyright 2004 by Walter de Gruyter GmbH & Co. KG, D-10785 Berlin
All rights reserved, including those of translation into foreign languages. N o part of this book may
be reproduced or transmitted in any form or by any means, electronic or mechanical, including
photocopy, recording, or any information storage and retrieval system, without permission in
writing from the publisher.
Printed in Germany
Cover design: Christopher Schneider, Berlin
Acknowledgements
Καιρός τοϋ σπείρειν, καιρός τοϋ θερίζειν wrote George Seferis many
years ago,* but it is only now, as I write these lines, that I feel the depth
of this metaphor like the good ploughman who has sown his field and
eagerly awaits harvest time. I only hope that both seeds and field were
good.
This book began its long journey to the Homeric seas as a PhD
dissertation submitted to the Classics Department of Cornell University at
the end of 1997. Its present state is much different, and I feel the need to
dwell for a while on the various changes done for this thorough revision.
One completely new chapter ("Distance, Closeness and Mors
Immatura: Common Motifs in the Iliadic Personal Laments"), has been
added. Chapters 1 and 4 ("The Morphology of the Iliadic Personal
Laments" and "Ars Allusiva: Intratextual Readings in the Iliadic Personal
Laments") have been thoroughly revised and the appendix on Iliadic short
obituaries has been considerably enriched. A chapter on the poetics of
Thetis' personal lament (to appear in QUCC) and the second appendix of
my thesis ("Localization and Metrical Shape of Lament Terms in the
Iliad'') have been omitted.
Many people have helped me improve my work in various ways. For
a first acquaintance with deixis and its poetics, I would like to express my
gratitude to Lucia Athanassaki and Natassa Peponi for inviting me to
attend a Conference in Delphi on Deixis in Greek and Latin Literature. I
also offer special thanks to two of my colleagues in Crete, Stavros
Frangoulidis and Yannis Tzifopoulos, as well as to Prof. George
Christodoulou in Athens, Michael Lipka and Andreas Markantonatos (to
whom I owe the first part of the book's title) in Patras for their
encouragement and insistence on the importance of turning this
dissertation into a book. For their generosity in offering valuable criticism
on this study, I would like to thank Fred Ahl, Kevin Clinton, Richard
C.C.T.
Athens
March 2004
Contents
Conclusion 166
Table 1 51
Table 2 112
Bibliography
Indexes
"... the Iliad is a unity in a deeper sense than is sometimes allowed, embodying
a clear and unique vision of the world, of heroism and of life and death.'"
Jasper Griffin
Homer on Life and Death
The Iliad is about the wrath of Achilles and the disasters it inflicted upon
the Achaean army; that is what its proem prepares us for.2 From the very
beginning of the poem, suffering and destruction creep up in the scene
and remain there virtually until the very end. The Iliad begins with pain
and suffering in the Achaean camp (due to the plague sent by Apollo), and
ends with pain and suffering in the city of Troy (during Hector's funeral).
As one of the fundamental themes of this epic, death is not simply
envisaged as a static phenomenon, but is replete with a remarkable
dynamism, since it initiates and later on enhances a sequence of events
that lead the Iliad to its destined end. The killing of Patroclus in Book 16
introduces the peripeteia which will be completed with Hector's death in
Book 22 and his funeral in Book 24. Thus "das große Gedicht vom Tod",
as Marg3 has put it, makes death the prevailing, unbroken theme which
directs the action.
The range of responses to death covers both physiological and psy-
chological reactions such as grief, as well culturally determined conduct
such as mourning.4 The principal poetic by-product of the deaths of warriors
is lament, which is expressed by mourning scenes and description of
1
Griffin (1980) 16.
2
See 1.1-7 (where a reference is not specified, I refer to the Iliad).
3
Marg (1965) 39.
4
See Derderian (2001) 4 referring to the work of Baudy (1980) 129-142.
2 Chapter 1
The most celebrated passage for the purpose of this study is 24. 719-722,
where the Iliad makes the distinction between the θρήνοι and the γόοι.8
5
As in Book 23 for Patroclus and in 24 for Hector.
6
For the loss of minor figures such as the sons of Diocles, Simoeisius etc.
7
There are also non-expressed lament speeches or threnodies (θρήνοι), which would be
normally uttered by professional singers (θρηνψδοί), and unreported γόοι (indicated
or alluded to by specific expressions containing the word γόοι).
* See Alexiou (2002) 102-103, who notes that "...Homeric and archaic usage may have
distinguished θρήνος and γόος according to the ritual manner of their performance,
using θρήνος for the set dirge composed and performed by the professional mourners,
and γόος for the personal weeping of the kinswomen. Further, early instances point to
the θρήνος as more ordered and polished, often associated with divine performers and
a dominant musical element".
9
The text of the Iliad is that of West (1998-2000).
2. Γόοι and Θρήνοι 3
And when they had brought him inside the renowned house, they
laid him
then on a carved bed, and seated beside him the singers
who were to lead the melody in the dirge, and the singers
chanted the song of sorrow, and the women were mourning beside
them.
10
According to Seremetakis (1991) 125, the social parameters of the Homeric θρήνος
(which is solely performed by professional singers, the άοιδοί) have been treated in
Inner Mani as a performative relation. Performative rights and precedence in the order
of performance are often granted to an exceptionally skilled mourner, even if he/she
is only distantly related to the deceased.
11
On the particle τε in epic diction (not only Homeric), see the meticulous and
exhaustive study by Ruijgh (1971) and GH II, 340-345, especially 343-344.
12
In II. 18. 604-606 = Od. 4. 17-19 (τερπόμενοι· μετά δέ σφιν έμέλπετο θείος άοιδός
/ φορμίζων- δοιώ δέ κυβιστητήρε κατ' αυτούς, / μολπής έξάρχοντος, έδίνευον κατά
μέσσου), there is a description of a dance on the Shield of Achilles; the words άοιδός
and έξάρχοντες are attested in lines 604 and 606 respectively, but μετά δέ σφιν έμέλ-
πετο θειος άοιδός / φορμίζων do not appear in any manuscript of the Iliad; they have
been restored by some editors because Athenaeus (V 180c and 181°) reports that these
lines were taken out of the Iliad by Aristarchus; they also change έξάρχοντες to έξάρ-
χοντος so that it agrees with άοιδοϋ, which should be understood from the previous
4 Chapter 1
line; Van der Valk (1964) 530 thinks that έξάρχοντες should stand as it is. Nonethe-
less, εξαρχος in 24. 721 is a hapax, as it is a noun and not a participle like έξάρχο-
ντες; as far as άοιδός is concerned, I tend to side with West (ad loc. in his Teubner
edition), who does not supply this line since it is not given by the manuscript tradition.
13
Willcock (1984) 321.
14
See Eustathius ad. loc. (van der Valk 977): «έπεί ή γάρ στονόεσσα άοιδή, δ έστιν ό
θρήνος, ούκ ήν και άνδρων και γυναικών, άλλ' αί μέν έστενάχοντο, οι δ' έθρήνουν,
δι« τοϋτο ειπών 'οΐ στονόεσσαν άοιδήν', ô ούκ ήν δ' άμφοϊν κοινόν, έπαναλαβών
κα'ι μερίσας διευκρίνησεν».
15
See 24. 721.
16
Ford (1992) 15 and LfgrE s.v. άοιδή 2:"Gesang als Tätigkeit, wobei... der Charakter
des nom. act. jedoch immer gewahrt bleibt (nicht 'Werk')".
17
On άοιδός in Homer, see Kraus (1955) 65-87; Maehler (1963) 9-34; Svenbro (1976)
18-38; Thalmann (1984) 157-184; Gentiii (1988) 3-23; Goldhill (1991) 56-68; Ford
(1992) 90-130; Segal (1994) 113-141.
'« See Ford (1992) 14.
2. Γόοι and θρήνοι 5
The daughters of the Old Sea-god stood round you with bitter
lamentations, and wrapped your body in an imperishable shroud. The
Nine Muses chanted your dirge in sweet antiphony and you would not
have seen a single Argive without tears in his eyes, such was the clear-
voiced Muses' song.
Heubeck20 thinks that in the light of the Iliadic passage I have quoted
above (24. 719-722), "we should suppose that the Muses act as άοιδοί
(θρήνεον, 61), while the role taken in the earlier passage by the Trojan
women, is here assigned to the Nereids (οϊκτρ' όλοφυρόμεναι, 59)".21
From these two passages we can chart the basic similarities and
differences between the γόοι and the θρήνοι. The γόοι are personal
lamentations uttered by the next of kin, and are delivered in speech (not
sung) by both male and female mourners. The θρήνοι are musical laments,
set-dirges22 sung by non-kin professionals; they probably contain "a praise
to the dead referring to their deeds or a lament in more general terms"23
and are artistic in nature with less improvisation and spontaneity than the
γόοι. They are (like the γόοι) capped by cries or even phrases, probably
in the manner of a refrain, expressed by a chorus. Both γόοι and θρήνοι
represent a marked form of speech reserved only for the great heroes.24 It
is hard to tell whether the θρήνοι were gender-oriented.25 The majority of
" The text of the Odyssey used is the OCT (Allen, 1917-1919).
20
Russo/Fernandez-Galiano/Heubeck (1992) 366-367.
21
Bowra (1961) 5 argued that in Homer the chorus had not found its final duty (apart
from the γόοι, there are other instances in the Iliad where there is a chorus, like the
παιάν in 1. 472-474, the description of the ύμέναιος in 18. 493 ff., the λίνος in 18.
567), and that the leader, for whom a special term is used (έξάρχων), plays a
preeminent role.
22
See II. 24. 720; Od. 24. 60; Pi. P. 12. 6-8,1. 8. 63-64; Plato R. 388d, 398e; Plut. Sol.
21.5.
23
Kornarou (2001) 21 bases this claim on Reiner (1938) 62-63 (who supports the idea
of a θρήνος praising a hero's past deeds), and Harvey (1955) 169 (who argues that the
θρήνος would contain a lament in general terms).
24
The Iliadic γόοι are addressed to Menelaus, Patroclus, Achilles and Hector; the
θρήνοι are reserved for Hector (in the Iliad) and Achilles (in the Odyssey).
23
The γόοι were not gender-oriented. In the Iliad the majority of mourners are female
6 Chapter 1
(Thetis, Andromache, Hecuba, Helen), but there are also male mourners
(Agamemnon, Achilles, Priam).
26
See Sourvinou-Inwood (1983) 39; Hutchinson (1985) 191; Easterling (1991) 149;
Shapiro (1991) 636; Segal (1993) 57-58.
27
In view of Od. 24. 58-62, of some ancient sources (Aesch. Cho. 733, Plato Lg. VII 800e,
Plut. Sol. 21.4) and of the fact that women play a more prominent role in iconographie
representations of πρόθεσις-scenes, Kornarou (2001) 21 recently argued that the
θρήνος was exclusively performed by hired female mourners. I am afraid I have to
disagree with this claim, given that the ancient sources she refers to are much later
than the Homeric θρήνοι and are unlikely to have reflected archaic practice; moreover,
iconographie representations may simply reflect the more prominent role of female
mourners in general, not specifically while performing the θρήνος. In addition, there
is an inherent difficulty in determining iconographically whether we are dealing with
the representation of a γόος or a θρήνος. For Od. 24. 58-62, see main text above.
28
Maronitis (1999) 157-159 rightly argues that it is the linear deployment of the plot in
the Iliad which is partly "responsible" for the lack of internal narratives, internal poets
2. Γόοι and θρήνοι 7
case of Demodocus.29 This may be one of the reasons why the Iliad avoids
the professional lamentations, the θρήνοι or set-dirges of the άοιδοί, but
extensively quotes the γόοι. The latter abound in condensed characteristics
of the inherited θρήνοι (such as those concerning the performance
framework they follow)30 and themes taking their cue from the Iliad itself;
in other words, the typology of the γόοι is combined with the very nature
of the poem that has given birth to them. The γόοι are personal laments
uttered by individuals closely related to the deceased; they are their
friends or relatives, not δημιοεργοί, and so the Iliad "is allowed" to give
them in full." Such a solution, however, does not cater for all cases. For,
if the answer is the typical Iliad vs Odyssey differentiation on most
matters poetical, why then does the Odyssey not cite or quote the content
of the set-dirge or θρήνος for Achilles by the Muses and the Nereids in 24.
58-62? This question becomes all the more pertinent when we consider that
the Odyssey cites or quotes in reported speech the song of other
professional singers such as Demodocus or Phemius.
The suppression of the θρήνος for Achilles in Odyssey 24 is due both
to specific parameters emanating from the immediate narrative and to a
general odyssean narrative strategy. The immediate alternative for the
Odyssey was to have Agamemnon -the speaker in this passage- either give
the θρήνος of the Muses in reported speech or quote it in direct speech;
both options must have been (rightly so) quite unthinkable. It would have
been rather inappropriate to use this scene for such an elaborate lamenta-
tion of Achilles. The internal audience, both the suitors who have recently
arrived in the Underworld and the heroes of the Trojan War who have
been there for some time, must listen to a praise for Odysseus. His ολβος
(in the manner of Demodocus or Phemius) and other songs embedded in the main
narrative.
19
One should note that the first (Od. 8. 73-82) and the third (Od. 8. 499-520) songs of
Demodocus are given in reported speech; in 8. 266 if. his second song starts in
reported speech but very soon slips into direct speech. The references to the other άοι-
δοί, Thamyris in II. 2. 594-600 and the two anonymous singers in Argos (Od. 3. 267-
271) and Sparta (Od. 4. 17-18), are insignificant. Phemius (Od. 22. 330-353) speaks
to Odysseus but his song is not given either in direct or reported speech. For a
comparison between Demodocus in Scheria and Phemius in Ithaca, see Marg (1957)
11 and Pucci (1987) 201-207.
30
This is clearly seen in the case of ritualistic (Iliad 22) or ritual γόοι (Iliad 24).
51
See Andronikos (1968) 12-13.
8 Chapter 1
surpasses that of both Achilles, who won κλέος but did not gain his νόστος
since he died at Troy, and Agamemnon, whose νόστος was practically
nullified by his own murder once he arrived at Mycenae.32 Moreover, the
narrative conditions (based on a typology of odyssean internal narratives)33
are not fulfilled. In particular, the placement of the scene in the Under-
world annuls all the temporal and spatial specifications which the Odyssey
typically employs when offering an internal narrative. This is compounded
by the masked absence of a request for such a speech, which the internal
narrator only somewhat grudgingly utters if specifically called upon to do
so. Within the above narrative parameters a threnodic encomium for
Achilles would have been rather unfitting, to say the least.
3. Selection Criteria
32
On the supremacy of Odysseus and of the Odyssey vs Achilles, Agamemnon and the
poetic traditions they represent, see Danek (1998) 486-487 and Tsagalis (2003) 43-56.
33
For a typology of internal narratives in the Odyssey, see Maronitis (1999) 162-164.
34
Fingerle (1939); for speech classification in the Odyssey, see Larrain (1987).
35
Bezantakos (1996).
36
Derderian's (2001) 33-34 classification of γόοι is problematic since only four of the
formal laments she cites occur in a ritual context and one is expressed by a male
mourner, Achilles (23. 19-23). Derderian includes in her discussion four other laments
(34, ft. 76) which she does not consider as γόοι because they are not labeled as such
(which is true) but exemplify more or less the same typical themes, motifs and
features of the formal γόοι. What Derderian has failed to see is that the γόοι "are bound
more by structure and content than ritual performance context" (Hame 2001) and that
they are not a female-genre but a female-dominated genre. Derderian has not consulted
Fingerle who offers a constructive, albeit not complete, guide to the use of reliable
criteria for classifying Iliadic speeches in general. To use a typical example, Achilles'
speech to Patroclus in 23. 19-23 is not a γόος, despite of the standard introductory
formula that is employed before the γόοι. The content and structure of this speech are
very different from any other lament; the use of the typical introductory formula is rather
due to the funereal context of Book 23 that "invites" the use of lament vocabulary.
3. Selection Criteria 9
other textual markers such as those often found outside the personal
laments, indicating that a speech to follow or one which has just been
completed is identified as a γόος; 3) thematic (and partly structural) typology
-by this I am referring not only to the actual content of a speech, but also
to a broadly defined set of motifs that the Iliadic γόοι employ and arrange
in a more or less stereotypical manner. In my view the formulation of
these criteria calls for some justification.
37
See Parry (1971). For typical scenes in Homer, the scientific incipit belongs to Arend
(1933).
" A number of scholars (Bekker, Fick, Robert, Dilntzer, Ellendt, Hinrichs, Witte,
Meister) had partly anticipated the Panyan "discovery" of the formula, but they did
not work out a general, cohesive theory, nor did they realize the far-ranging implica-
tions of such an approach. See Holoka (1991) 456-481. Recently, Hummel (1998)
55-71 has drawn attention to the fact that certain of the main arguments of Parry had
been already expressed (though not in a systematic manner), by Pierre-Antoine Grenier
in 1861. These early studies in no way undermine the revolutionary contribution of
Milman Parry to Homeric research and the field of Oral Poetics. From the very
beginning, Chantraine (1929) 294-300 correctly observed that Parry was the first
scholar to study noun-epithet formulas in a systematic manner, and realized that his
approach "renouvelle la philologie homérique" (294).
" See Lord (1948) 34-44; (1951a) 57-61; (1951b) 71-80; (1953) 124-134; (1956) 301-
305; (1960); (1968) 1-46, (1969) 18-30; (1970) 13-28; (1976) 1-15; (1981) 451-461;
(1986a) 467-503; (1986b) 19-64; (1986c) 313-349; (1991); (1995).
10 Chapter 1
40
Foley (1991) xii.
41
I am here referring to Lord's initial studies.
42
The most noteworthy early efforts to modify the Parryan definition of the formula
were made by Hoekstra (1965), Nagler (1967) and Hainsworth (1968).
43
The bibliography on oral-formulaic theory is immense. The most comprehensive
survey for the research carried out, that I am aware of, is that of Foley (1985). See also
Edwards (1986 & 1988) for the formula and (1992) for the type-scenes.
44
Foley (1991) xiv.
45
Foley (1991) xiv.
« Finnegan (1977).
3. Selection Criteria 11
56
According to Nagy (1996b) 110, 112 text-fixation designates texts in the sense of
scripture "where the written text need not even presuppose performance".
57
Nagy (1996a) 42 and ( 1996b) 110 proposed a five-period Homeric transmission model
based on performance, not on text (but text in the traditional meaning of the word)
"with each period showing progressively less fluidity and more rigidity" (1996b) 109.
" There were also early advocates of this approach, like Bassett (1938).
59
On time and space, see Hellwig (1964) 4-22. On foreshadowing and suspense, see
Duckworth (1933); Reichel (1990) 125-151, (1998) 45-61; Nünlist (1998) 2-8;
Rengakos (1995) 1-33, (1999) 308-338. On characterization, see Van Ε φ Taalman
Kip ( 1971 ); Andersen ( 1978). On presentation of the story, see de Jong ( 1987b); Rabel
(1997). On the Homeric narrator, see Richardson (1990). Cf. also the twin narratological
approaches of Morrison (1995) and Doherty (1995) for the Iliad and the Odyssey
respectively, and de Jong's narratological commentary on the Odyssey (2001).
60
See de Jong (1987b) 29-40, who in turn employs Bal's model of analysis (19972
[1985]) "which incorporates and partly refines that of Genette (1980)".
61
The terms diegesis and mimesis are used by Plato in the Republic (3. 392d), where he
is concerned not only with the form and content of poetry but also with its ethos. By
laying emphasis on how (ώς λεκτέον 3. 392c) poets and mythologers say what they
say, Plato distinguishes between three types/classes of poetry: single-layered narrative
recounted by the poet himself (άπλη διηγήσει), narrative effected through
impersonation of a character (διά μιμήσεως γιγνομένη) and narrative effected through
both diegesis and mimesis (δι' άμφοτέρων). Aristotle followed Plato in classifying
poetic genres according to the way one imitates, but further elaborated his
classification of poetry in respect of epic: he distinguished between a brief non-
3. Selection Criteria 13
for apart from our own ideas, aesthetic principles or interpretative desires,
we should attempt to decode what the poem itself says about the
speeches' identity. When we pass from narrator-text to character-text, the
external narrator recedes into the background, withdraws and lets the
internal narrator express his ideas, feelings or arguments. Before doing so,
he uses an introductory line which lies on the border62 between diegesis
and mimesis·, this line functions as a "marker", a "literary label" that
describes the identity of the ensuing speech:
The external narrator indicates here one speech-act with two distinct
aspects:63 a locutionary (μετέφη) which describes the utterance of the
speech and an illocutionary (ευχόμενος) that refers to the form of the
speech-act. In this way, he clearly sets a limit, marks a stop in the diegesis
and indicates not only that a speech will follow and that someone will
speak, but more importantly how the speaker will speak.
That is to say, despite the fact that he will withdraw from the scene
when the speech is uttered, only to return after its completion, the external
narrator smiles behind the curtain he himself has just drawn, since he is
mimetic proem (where the poet speaks as poet), a mimetic narrator-text (where the
poet speaks as narrator) and an equally mimetic character-text, the speeches (where the
poet speaks as the character he impersonates in any given situation). For a detailed
analysis of the two passages discussed above, see de Jong (1987b) 1-9; Ford (1992) 22.
62
Chatman (1990) rightly treats both diegesis and mimesis as separate narrative modes
and subordinates them to a wider category he calls Narrative. At the same time, the
two narrative modes often intermingle. In light of the above observations, the verse
introducing a speech stands between the diegetic and mimetic modes on the one hand
and comprises a part of the broader category of Narrative on the other. See Ricoeur
(1984-1988) vols, i-iii; Onega & Landa (1996) 3. I owe these bibliographical
references to Markantonatos (2002) 2, ft. 4.
63
See Searle (1976) 1-23; Prince (1978) 305.
14 Chapter 1
the one who has decided that it is now time for a character to speak. When
he uses a closing formula, he indicates that the actual speech is over and
that the diegesis will now continue.
In addition to the introductory and closing formulas, there are other textual
markers -such as γόος-related terminology- which may indicate that a
speech should be identified as a γόος:
This line functions as a "marker", for it denotes either that a γόος will
follow (first case) or that it has preceded (second case). A textual marker
can be very helpful, but should be used with great caution and always in
combination with the other criteria I have referred to; its use of lament
vocabulary may be simply due to its immediate environment. In the case
of 23. 106, Achilles uses lament vocabulary to refer to the speech the
ghost of Patroclus has uttered because of the general funereal
environment it has been placed in and so Patroclus' speech does not, in
fact, meet any of the other criteria for being a γόος-speech.
64
The fact that line 6. 373 is expressed by the external narrator, whereas lines 23. 106
and 23. 157 by Achilles, who functions as an internal narrator, is not of particular
interest to me; I am using any textual indication, whether placed in narrator or
character-text, as evidence that the Iliad recognizes a speech as a γόος.
65
This is the R. Lattimore translation, which renders the same word (μυρομένη) into
English using two different adjectives, namely tearful and mourning.
3. Selection Criteria 15
66
I have deliberately lumped together typical motifs and structural features of the γόοι;
the thematic typology of the Iliadic personal laments is inextricably interwoven with
its internal arrangement, i.e. the use of the same motifs in the same order by different
speakers. In other words, both patterns (thematic and structural) are the different faces
of the same coin. Moreover, the use of structural features of broader attestation, e.g.
ring-composition, shows how the Iliad has both "absorbed" and adapted the sub-genre
of the γόοι to its subject-matter, applying the same structural techniques it has used
for various sorts of speeches.
61
Nagy (1996a) 18.
16 Chapter 1
4. Types of Γόοι
68
See the personal laments of Andromache in 6, Briseis in 18 and Hecuba in 24.
69
See also Chapter 5 (5. 1 Introductory remarks) and Table 2.
70
By textual environment I am referring both to what precedes and follows a personal
lament and to the positioning of the γόοι within the narrative structure of the whole
poem. Most speeches of this kind are situated in important narrative junctures that
designate a significant shift in the course of the epic plot.
71
The γόοι at the end of Book 22 are triadic (Priam, Hecuba, Andromache) and
ritualistic (for, unlike Book 24, they are uttered at the absence of the deceased's body,
since Hector's corpse is retained by Achilles).
5. Genre and sub-genre 17
72
Bakhtin (1986) 66.
73
According to Suleiman (1980) 35, the horizon of expectations is "the set of cultural,
ethical and literary (generic, stylistic, thematic) expectations of the work's readers in
the historical moment of its appearance".
74
Sourvinou-Inwood (1995) 3.
18 Chapter 1
75
On intertextuality, see Culler (1975) 139-140; Kristeva (1981) 170; Culler (1981) 12,
38, 100-118; Eco (1981) 21-22, 32.
76
In contrast, such cries do occur in Greek tragedy. See Koonce (1962) 77 if.; Stanford
(1983) 59; McClure (1995) 35-60; Kornarou (2001) 87-93; Loraux (2002) 35-41.
5. Genre and sub-genre 19
This does not mean that communicative purposes do not exist in such
cases. It is simply a question of distinguishing between marked and
unmarked elements. The more we adopt marked forms of expression, the
more we need to specify minute communicative purposes in order to
establish distinctions between genres and sub-genres.77 Hence the
distinction between the θρήνοι and the γόοι within the Iliad, which is
necessary for the designation of particular sub-generic subtleties.
Finally, genre analysis has shown that genres, situated as they are in
specific socio-cultural contexts, have always been considered as "sites of
contention between stability and change".78 The crystallization of genres
seems to be at odds with a propensity for innovation.79 The latter is often
exploited by the expert members of the community in order to create new
forms in response to new historical contexts. Genres based on conventio-
nalized and institutionalized discoursal practices of specific communities
are imbued with the rigidity of canonizing, but also with the fluidity
typical of tropes. Consequently, they adopt new forms and refuse to abide
by rules of integrity, whether generic or other. While language moves
towards less organized forms, the socio-cultural context in which it
evolves tries to classify and tame it through the creation of numerous sub-
categories and sub-branches.
This conflict between rigid stability and fluctuating change will be
observed in the case of Iliadic γόοι. Divergence from the norm does not
mean that the norm doesn't exist, but rather that by using established
generic knowledge, the poet handles generic resources and conventions in
such a way that they are able "to express private intentions within the
framework of socially recognized communicative purposes".'0 Beyond
the poet, the expert tradition within the discourse community works in the
same direction. This line of thought is greatly reinforced by the fact that
such innovations are realized within generic boundaries and, by recourse
to modified material from the collective treasury of resources accumulated
by generic convention.
This conflict can be seen in Saussurian terms as a slightly modified
version of the well known dichotomy between langue versus parole and
77
Bhatia (1997) 634.
78
Berkekotter and Huckin (1995) 6.
79
See Bhatia (1997) 634-635.
80
Bhatia (1997) 635. The italics are mine.
20 Chapter 1
83
Cairns (1972) 99.
84
The Iliad bears traces of an established usage of lamentation, but voices only one form
of this lamentation, the γόοι. See Smyth (1906) cxxi.
85
Cf. Nagy (1979) 112.
22 Chapter 1
It has been maintained that "the theme and diction of lament appear to
have shaped the Iliad and can even be found embedded in the name of
Achilles, 'grief of the fighting-men"'.86 If this is true, then the fact that the
Iliad represents only the γόοι becomes even more significant, since it
poses another kind of question: do the γόοι represent a "better genre" than
other forms of lament? This may seem a naive question," but if sub-
genres found in epic (such as commanding, flyting and recollection)
demand treatment as "poetic" performances, then one can assume that an
ancient audience accustomed to traditional oral-poetics is "naive"" in the
sense that "it has internalized the conventions of the overarching genre (in
this case, epic) to the extent that it can focus more than we can on the
primary, sub-generic level..."."
7. Scope of Research
16
Nagy (1979) 69-71 quoted in Martin (1989) 86.
17
On the same kind of question see Martin (1989) 89 ff., with revealing comments on
the function of speech-genres as primary types of performance.
" Scodel (2002) proposes a reassessment of epic performance by challenging the view
that Homeric audiences were minutely competent. She rightly argues that epic poetry
"speaks" as if everything is familiar; we do not need to postulate superbly informed
and experienced audiences in order to account for the understanding and appreciation
of the technical complexity and sophistication of Homeric poetry. Positive reception
of this poetry was, no doubt, a reality and, one might say, one of the basic factors
contributing to its diffusion. Therefore, it is not a question of competent or trained
listeners but of positively responding audiences.
·» Martin (1989) 90.
90
Fingerle (1939) classified all the speeches of both the Iliad and the Odyssey into
categories using various criteria. As far as lament is concerned, he was the first to
distinguish between two distinct types of lament speeches in the Iliad: the Totenklagen
and the Klagereden. As Totenklagen, Fingerle classified lament speeches addressed to
a dead person and as Klagereden general laments, not addressed to the dead. The
importance of Fingerle's study lies in his perceiving of the existence in Homer of
introductory and closing lines that initiate and cap the Totenklagen and in his using
them as criteria for his speech classification. In 1970, Lohmann attempted a sensitive
approach to the composition of all the Iliadic speeches. He examined both their
7. Scope of Research 23
internal and their external mode of composition and pointed to the existence of three
basic structural patterns: ring-composition, parallel composition and free expansion.
The most recent monograph on the speeches in the Iliad is by Bezantakos, which
modifies the pioneering work of Fingerle and offers a complete catalogue of all the
Iliadic speeches divided into categories. As far as the Odyssey is concerned, the most
recent comprehensive account known to me, is that of Larrain (1987), with rich
bibliography on the previous work done on this topic. Generally speaking, speeches
in the Iliad have received more attention than those in the Odyssey.
" Monsacré (1984) examined the expression of suffering in Homeric poetry, the
different attitudes of men and women in respect of the externalization or
internalization of grief, and the wider consequences these attitudes imply for our
understanding of Homeric society. Monsacré is influenced by the Paris school and the
emphasis in her book is on what Vidal-Naquet describes in his preface as Homeric
anthropology, which is actually an anthropology of the text.
92
Reiner (1938) examined the lament for the dead in all ancient Greek literature. The
wide thematic scope of his study did not allow him to focus his attention on the lament
theme in the Homeric poems; nevertheless, his survey is the first momentous
contribution to the research of the way lament operated in ancient Greek literary
tradition. A true landmark in the research concerning lament in Greek literary heritage
was the work of Alexiou (2002 [1974']). Alexiou studied lament not only in ancient
but also in byzantine and modern Greek culture, showing the existence of common
motifs, conventions, themes and even formulas in all phases of Greek literature, thus
pointing to its continuity. As far as Homer is concerned, Alexiou analyzed the
personal laments at the end of Book 24 of the Iliad and argued for a three-part form
following a ring-composition pattern (ABA). Her observations (2002,131-160) on the
antithetical thought and antiphonal structure of the laments as well as on the use of the
allusive method broke new ground in the study of lament in Greek literature.
93
Petersmann (1973) attempted an analysis of five speeches which he called "lament-
monologues". His analysis is not as pervasive as one would have expected. Not only
24 Chapter 1
did he fail to see the problem of the placement of the personal laments, he even included
Achilles' initial speech in 18. 6-14 in his category of "monologische Totenklage"
despite the fact that it is addressed to his own θυμός. Nevertheless, Petersmann made
an important distinction between ritual and non-ritual personal laments. He
convincingly argued that only the three final γόοι in II. 24 should be regarded as ritual
in the true sense of the word, since they are uttered during a funeral (that of Hector), and
show traces of a latent pattern of internal composition and sequential performance upon
which they are based. Pucci (1993) 258-272 was the first to analyze the personal dirges
of Briséis and Achilles in II. 19, by laying the emphasis on the importance of the re-
presentation and placement of the laments within the framework of oral performance.
Derderian (2001) 15-62 devotes one chapter to the study of Iliadic personal laments.
The rest of her book deals with the archaic epigram, the study of lament in the
classical period (Pindar, Simonides and Sophocles) and the epitaphios logos.
" Cf. Fingerle (1939).
95
Mawet (1975) turned the focus in a new direction scarcely studied before: the
connection between lament poetry and funerary epigrams. Later on (1979), extending
Anastassiou's study on the Homeric vocabulary of mourning (1971), she explored the
functional oppositions in the vocabulary of pain in Homer; though not concerned
solely with lament, she added much to our knowledge of the differences between
words pertaining to the semantic groups of πήμα and άλγος. On the relation between
the genre of lament and mourning in general with the inscribed epigram, see also
Derderian's detailed examination (2001) 63-113. Derderian convincingly
demonstrates how the ambiguous σήμα of epic is transformed into a μνήμα, with the
resultant increase in its communicative capacity by the use of writing.
7. Scope of Research 25
96
See Martin (1997) 138-166, who suggests the same thing for the Homeric similes,
arguing that they are "genre imports", swallowed by the "ambitious supergenre" of
epic poetry.
97
Martin (1989) 88.
26 Chapter 1
and Hector; the future destruction of Troy; Achilles' suffering and his
future death: in a nutshell, the kernel of the entire poem98 is mirrored in
the Iliadic personal laments." By using external analepses and prolepses,
the γόοι draw a circle that surpasses the epic and encompasses larger parts
of the Epic Cycle. In terms of poetics, the γόοι reproduce the fabric of the
Iliad and, by and large, summarize it.
98
See Griffin (1980) who studied the polarization between life and death, a theme of
central importance for the interpretation of the Iliad, of which lament constitutes an
integral part.
99
See Foley (2001) 44.
Chapter 2
In the Iliad there are twelve (12) γόοι100 quoted in direct speech, of which
ten are situated in the second part of the poem from Book 18 to Book
24101, but only two in the first part, namely in Books 4 and 6. The full list
contains the following γόοι:
100
By using the term γόος I refer either to a speech introduced with a phrase containing
this word or to a speech introduced by some other expression of lament and followed
by the formula έπί δέ στενάχοντο γυναίκες / γέροντες / πολΐται / δήμος άπειρων.
Although in the second case the word γόος is not used, the thematic development and
internal structure of these speeches place them within the category of γόος-speeches.
By thematic development I am referring to the selection and presentation and by
internal structure to the order and disposition of elements contained in γόος-
speeches. See also section 3 (Selection Criteria) in chapter 1.
101
Introductory and closing formulas are extremely important for classifying the γόοι.
Petersmann (1969) 113 and (1973) 4 makes a distinction between ritual death-
laments (rituelle Totenklage) and those which are more personal and directly function
as a form of monologue. He only considers the three γόοι in Iliad 24 by Andromache,
Hecuba and Helen as ritual, arguing that all the rest differ significantly from them.
On the other hand, Reiner (1938) 12 ft. 4 maintains that Thetis' speech in 18. 52-64
is a ritualistic γόος rather than a monologue. Petersmann's classification is closer to
mine. Ritual laments are those having both a formulaic introduction and a typical
closure. Formulaic introductions and closures are rather the by-product of the ritual
nature of a personal lament. As ritual γόοι we should classify those fulfilling the
following three criteria at the same time: (a) they are introduced and capped by
specific formulas (see chapter 3); (b) they are placed within a purely funerary
environment and (c) they contain a number of basic elements (which I examine in
this chapter). According to these criteria only the three laments in Book 24 are ritual
γόοι. Those in Books 19 and 22 are ritualistic, i. e. they share the personal laments'
basic properties but are not ritual, for one essential prerequisite is not met: the body
of Hector has not yet been returned by Achilles (cf. also Seaford 1994, 154-190). I
do not consider the speech of Achilles (23. 19-23) as a personal lament. The use of
the formula τοΐσι δέ Πηλείδης άδινοΰ έξήρχε γόοιο in 23. 17 is a reflection of the
28 Chapter 2
influence of the funereal context. There is nothing that recalls a personal lament in
this speech. For a different classification of the γόοι, see Bezantakos (1996) 154-156,
who classifies laments and complaints in the same category and considers certain
γόοι as general laments. Table 2 offers a classification of the Iliadic personal laments.
102 By "Comparison" and "Death-Wish" I am translating the German terms "Vergleich"
and "Todeswunsch" respectively. These terms were employed in the analysis of
Homeric speeches by Lohmann (1970) 99-11 who also draws on the groundwork of
Fingerle (1939).
103
Cf. Reiner (1938) 116-120. See also Garland (1985) 146, with a useful note on περί-
δειπνον; Garland says that according to Prof. L. R. Rossi in a talk entitled "Poem and
2.1 Introductory Remarks 29
fee in Greek Archaic poetry, both choral and monodie" delivered to the Hellenic
Society on Nov. 15th 1979, the following fragments may be extracts from long lost
perideipnon songs: ΡMG 419, 485(?), 894, 896, 907, 911; Pi. /. 2 (?); Anaxandrides,
fr. 1 K.-A. See also Cie. De leg. II 63. In Homer we hear about something probably
linked to the perideipnon in 23. 29, after Achilles' personal lament for Patroclus; on
this occasion the eating takes place before the cremation of the body (23. 29 ff.),
whereas in Hector's funeral after the cremation (24. 802). Keeping this in mind one
could formulate an interesting hypothesis: since the γόος was the informal, improvised
lament uttered by the next of kin and was later on restricted to the house (due to
Solon's legislation), it may have been replaced by some sort of praising lamentation
uttered during the perideipnon. Both Zenobius V. 28 ("είώθεσαν oí παλαιοί έν χοϊς
περιδείπνοις τον τετελευτηκότα έπαινεΐν, και ει φαΰλος ήν") and the proverb "οΰκ
αν έπαινεθείης ούδ' έν περιδείπνψ" seem to support this claim. It is precisely this
praising element -so often encountered as an integral feature during my analysis of
the Iliadic personal laments- which makes me think that this praise might not have
been lost. Moreover one should keep in mind the belief of the Greeks that the dead
was present during the perideipnon in the capacity of a host. This fits in perfectly
with Patroclus' funeral in II. 23, since his dead soul visits Achilles and talks to him
after the funeral feast (23. 29).
104
Ferrari (1984) 264-265.
105
Fen-ari (1984) 265.
106
Alexiou (2002) refers to the same motif by using the exactly opposite term (!), as she
talks of "the antithesis between mourner and deceased". For all my respect for her
groundbreaking work on Greek lament, I think that all the Iliadic examples she refers
30 Chapter 2
to (the same as my study) do not separate mourner and deceased, but rather attempt
to create a link between them, in the world of suffering. To use a well known example
(also mentioned by Alexiou), when Andromache laments Hector in 22. 477-478, she
begins her speech with the following expression: Ifj άρα γεινόμεθ' αϊση / αμφότεροι
.... She then employs first and second person pronouns -not to differentiate her fate
from that of Hector, but to emphasize their unbroken bond, their common suffering.
See also 22. 485 when she refers to Astyanax: δν τέκομεν σύ τ' έγώ τε δυσάμμοροι.
The antithesis is at work, but rather as an opposition between past and present than
between mourner and deceased. The motif of common fate is also connected with the
expression of a "Death-Wish" by the mourner.
107
See Alexiou (2002), in which the author rightly claims that the origins of the tripartite
structure of the Iliadic γόοι must be sought in the very structure of primitive rituals
(the same applies, Alexiou argues, to the hymn, the encomium, and the funeral
oration). The alternative scenario, according to which the tripartite structure of the
Iliadic personal laments is due to structural techniques pertaining to Homeric speech
at large (since the γόοι are coherent, uninterrupted speeches), seems to me highly
unlikely, given that the tripartite structure is typical of other genres of ritual poetry.
108
Garland (1985) 21.
2.1 Introductory Remarks 31
cannot exclude the possibility that a primitive belief in the magic power
of number three may have survived in ritual lamentations, especially since
the chief mourner (εξαρχος γόοιο) was initially regarded as the equivalent
of the later magician (γόης),109 who communicates with the dead through
songs and spells and is considered to be the "necromantic counterpart of
the poet"."0
g. Iliadic personal laments show a remarkable preference for ring-
composition as opposed to parallel structure or simply free development.
The first question one should ask is what this compositional choice re-
veals about the origin of the γόοι; ring-composition is both a composition-
al aid and a rhetorical device used in order to achieve a specific stylistic
effect, often that of emphasis. By repeating, in more or less the same way,
themes previously expressed in the speech, the speaker aims either to
stress their importance or to encircle a digression or short episode (as in a
"Comparison"), by weaving a larger fabric. On the other hand, since the
structure of a speech -its compositional pattern- reflects the poet's choice
or the tradition's preference, the ring-composition technique must have
some other meaning connected to the way personal laments were uttered
in the Greek world, one that "scarcely changed between the Bronze Age
and the Hellenistic Period"."1 This argument remains valid regardless of
whether we favor a single authorship for the Iliad·, the tradition in which
epic was forged knew of a specific ritual for the lamentation of the dead,
part of which would have been personal laments. Emily Vermeule argues
that "the two great art forms of early historical Greece, Ionic epic poetry
in the east and Geometric painting in the mainland provinces of Attica,
Boiotia and the Argolid, focus on burial and mourning in styles so similar,
109
See LSJ s.v. γόης: "a wizard, a sorcerer".
110
See Reiner (1938) 20 ff., 27-28 where a list of expressions of incantation in tragedy
is given; also Rohde (1925) 14, 198; Bickel (1926) 22; Burkert (1962) 36-55;
Romilly (1975) 13, 31; Vermeule (1979) 16-17. Burkert (1962, 45) refers to 24. 720
ff. and notes that this may be a reminiscence (Erinnerung) "an magische Sänger bei
der Bestattung ..., eine Erinnerung, die darum gleich unterdrückt wird, weil das
homerische Epos an Stelle des Magischen das Menschliche entdeckt hat". The most
recent account of the origin and roles of the γόης in relation to the songs of mourning
is offered by Johnston (1999) 82-123. She convincingly shows how a basically
female-restricted genre such as the γόος was gradually transformed into a male γοη-
τεία in the fifth-century polis.
111
Vermeule (1979) 11.
32 Chapter 2
that scholars understood they both share an older tradition long before it
could be proved. The themes of the Homeric Iliad are precisely those of
Attic painting, battle and sea and the ceremonies for the dead. These were
the old themes of the Mycenaean Greeks, and of most other Bronze Age
cultures around the Mediterranean"."2 Thus the personal laments also
forming an integral part in the ceremonies of the dead must have followed
some speech conventions, which would have in turn developed over the
centuries since they were practiced for a long time. Like the other art forms
(e.g. painting), these must have exhibited some consistency in the way
they were performed and composed as well as in the themes they dealt
with. The typology of their structure and thematic material is probably a
reflection of the lamenting conventions that our Iliad is aware of.
h. The antiphonal nature of the funeral lament is typical in all eras of
Greek tradition."3 In Homer, each phase of the singing of the ritual lament
is antiphonal: it begins with the hired mourners singing the θρήνος, while
a chorus "responds" to their grief. The next of kin then utter a personal
lament, the γόος, accompanied by a chorus of anonymous mourners utter-
ing a refrain of cries."4
112
Vermeule (1979) 11. The italics are mine.
113
For a detailed analysis of this concept in ancient Greek, byzantine and modern Greek
tradition, see Alexiou (2002) 131-184.
114
I cannot agree with Garland (1985) 30 when he says "In Homer... a chorus of women
utter a refrain of cries"; this is not always true (see 19. 338; 22. 429; 24. 776).
2.2 The Praising Address 33
turns out to be far greater than before. This is a way of expressing the
intensity of grief exprienced by the mourner; the more important the dead,
the greater the suffering the mourner feels and tries to express. There are
four typical introductory lines preceding the Iliadic γόοι:
All of these lines contain a vocative referring to the deceased and a dative
of either a personal pronoun referring to the mourner or a possessive
pronoun referring to the speaker's θυμός. In addition, there is a dative of
reference (usually modifying the word θυμός), an epithet in the
superlative degree and "an intensifier" of the superlative (such as the
adverb πλείστον or πολύ that may or may not accompany the superlative
epithet). Finally, we have a genitive expressing the second term of the
comparison, if it is not an absolute one. Therefore, it seems that the personal
lament has developed a particular form of introductory address to the
deceased stemming from its thematic preoccupation with the dead's special
impact on the life of the mourner. The particularity of the person results
in the particularization of the mourner's feelings and his specialness
effects a special verbal form of treatment, the expanded-praising address.
In Briséis' lament for Patroclus the expression κεχαρισμένε θυμψ used
when addressing a close friend and therefore showing endearment and
familiarity, acquires an intensifying tone lent by the superlative πλείστον
placed next to it."5 This particular case thus looks more like the other
initial addresses of the personal laments than those which use the phrase
κεχαρισμένε θυμψ; the superlative functions in such a way that it virtually
changes the scope of the address. Briséis is not only addressing Patroclus
but with him all the others who are dear to her or better were dear to her.
She is also addressing her three brothers (293) and husband (295) slain by
113
The expression έμφ κεχαρισμένε θυμφ is used five times in the ¡liad, for Patroclus
by Achilles (11. 608), for Diomedes by Sthenelus, Athene and Agamemnon (5. 243,
5. 8 2 6 , 1 0 . 2 3 4 respectively) and of course for Patroclus by Briséis (19.287). See also
Edwards (1991) 269. For the peculiar scansion of the vocative, see Pucci (1998) 99-
100.
34 Chapter 2
Achilles. This initial address is not simply a typical way to begin a speech;
it acquires a specific functional role, that of encompassing a wide scope
of people whom the speaker, through the comparison of the superlative
form that she uses, first implicitly and then explicitly includes in the
personal lament.
Briséis' introductory address to Patroclus contains the only Homeric
example of an enclitic following a vocative (Πάτροκλε μοι); the form εμώ
would have been perfectly possible, in which case it would refer to the
dative θυμφ and would follow the pattern of the other lines where the
expression κεχαρισμένε θυμφ is used. It seems that this unusual
expression creates an intense rhythm, reflecting the emotional involve-
ment and frustration of Briséis. Being a slave-girl, she is not connected to
Patroclus through family ties and is thus unable to address him using a
family term,"6 as often happens in the Iliadic funeral laments. Her speech is
actually looking ahead, towards the address (19. 315) by which Achilles'
γόος for Patroclus is initiated. It is to the dative μοι used there that the
personal pronoun μοι in 19. 287 responds, showing the level of consistency
between the initial address of the Iliadic γόος-speeches.
In Achilles' γόος-speech Patroclus is addressed by a line ending with
the phrase φίλταθ' έταίρων. The vocative of the word φίλος when used as
an adjective in the superlative degree is attested four times in the Iliad
three in the first line of the three personal laments I have quoted above
(19. 315, 24. 748, 24. 762) and one more at the beginning of the speech by
Idomeneus to his comrade Meriones in 13. 249. It is followed by the
genitive plural έταίρων in 13. 249 and 19. 315, since the genitive de-
termines the class of people among whom the speaker is placing the
deceased. Yet a closer look at the two passages shows that the meaning of
the genitive έταίρων is not the same in each case. When Idomeneus
addresses Meriones, the rest of his speech indicates that he refers to the
latter's identity as second-in-command, i. e. a front line warrior. This is
why he expresses his surprise on finding Meriones, a man of arms par
excellence, away from the battlefield. In Book 19 Achilles paints a picture
of Patroclus preparing and serving dinner for the Achaeans before the
battle; it is true that in both passages the word κλισίη is used to describe
the place where the person addressed either is present (Meriones) or
116
For the use of family terms in the funeral laments of the Iliad, see Ebbott (1999) 17.
2.2 The Praising Address 35
depicted (Patroclus),117 but the situation is very different. In the first case
both interlocutors are alive and present in the hut; the speech belongs to a
long series of exchanges between Idomeneus and Meriones. On the other
hand, following the death of Patroclus only Achilles can possibly be
physically present in the second case. Moreover, Idomeneus interprets
Meriones' coming to the hut either as a result of his wounding or because
he is bringing some message (ήέ τι βέβληαι, βέλεος δέ σε τείρει άκωκή, /
ήέ τε' άγγελίης μετ' εμ' ήλυθες;),"8 whereas Achilles is lamenting Patroclus
and recalling a familiar scene when his friend was alive, in the hut, pre-
paring dinner before the battle. One can thus see how the personal lament
shapes the meaning of the genitive έταίρων. The intimacy element that
emerges here to highlight the peaceful side of Patroclus is consonant with the
situation we encounter when we look at the initial addresses inaugurating
the personal laments of Hecuba and Helen in Book 24. There we hear the
genitives δαέρων and παίδων which, as other terms pointing to the
relation of the deceased to the mourner, show that the personal laments
use terms of familiarity and endearment. As a speech form uttered by the
next of kin or close friends, they stress first the mourner-deceased re-
lationship and then the impact of the loss on the entire community the
dead once belonged to. Patroclus' preeminent position among Achilles'
friends is, of course, a key-theme for the entire epic. Yet here it is treated in
a special way, being explained via personal memories such as the deceased's
gentleness and almost maternal role as one who cared, probably more
than anybody else, for his fellow comrades."9
The two initial addresses in the personal laments of Hecuba and Helen
in Book 24 share a number of similarities: they both begin with Hector's
name in the vocative and continue with the expression έμφ θυμφ. Metrical
reasons determine the positioning of the genitive to denote the class of
people the mourner considers the deceased to have been part of. In the
personal lament by Hecuba, this genitive comes at the end, whereas in that
by Helen it appears immediately after the phrase έμφ θυμφ. Consequently,
the word πάντων modifies the genitives παίδων and δαέρων.
Another common element is the adverb πολύ, which intensifies the
117
Cf. 13. 253 (ένί κλισί,-ησι) and 19. 316 (évi κλισίη) respectively.
118
See 13. 251-252.
See Shay (1995) 44-49.
Exploring the Variety of Random
Documents with Different Content
„Goed, en dan gaan we meteen thuis het pakje aanhalen.”
„Da’s al lang in orde. Moeder heeft voor alles gezorgd. Voor ieder een stapel
kadetjes met ham en kaas.”
Dan gaan de beide jongens op stap. Ambro koopt een pond goudreinetten.
„Wat een kanjers,” zegt Paul. „Me broekzakken puilen er van uit.”
Intusschen hebben ze Paul’s huis bereikt, waar [162]ze het pakje afhalen.
Boven aan de trap roept mevrouw: „Ambro, zul je goed op Paul passen?”
„We gaan eerste klas, hoor!” zegt Paul. „Dan zitten we fijn vooraan. En anders
krijg je al de rook in je facie.”
„Chique!” zegt Ambro. „Kijk het bankje bij de bel is vrij, laten we daar gaan
zitten, dan mogen we bellen als ie vertrekt.”
„Ik ga een appel schillen,” zegt Ambro en hij haalt zijn mes te voorschijn; het
mes, dat alle jongens hem benijden, want het is groot, en gevuld met alle
soorten van mesjes en priempjes en het is overal voor te gebruiken.
De machinist van de boot, die vlak bij de jongens z’n pijpje staat te rooken,
kijkt er met welgevallen naar.
Ambro, zeer gevleid, geeft hem het mes om ’t van dichtbij te laten bekijken.
„’n Effetief messie, best staal, zal nog heel wat gekost hebben.”
„Op de zomer-kermis, boven in een mast, die met groene zeep was
ingesmeerd. Ik had een heele sjouw voor ik er was en toen ik eenmaal boven
was, wist ik niet wat ik pakken zon, een ham of het mes.
„Maar ik koos toch ’t mes en nou ben ik er maar blij om, want ’t heeft me al
heel wat diensten bewezen.”
„Mot je een stukkie,” vraagt Ambro, hem de helft van zijn appel voorhoudend.
„As ik je niet ontrijf. Maar, ’t is zoo’n groot brok, je houdt temet zelf niks over.”
Vóór de machinist vertrekt, heeft ook Paul hem zijn aandeel geofferd.
En Paul, die op zijn beurt gehoopt had de bel [164]te mogen luiden, werd
beloofd bij de aankomst aan de eerste aanlegplaats zijn schade te mogen
inhalen.
Juist zou de loopplank naar binnen gehaald worden, als van de kade hevig
met een parapluie wuivend, een oude juffrouw aan komt loopen, zoo hard, als
haar onderdanen het toelieten.
„Wacht even, schipper, die juffrouw moet nog mee,” roept Ambro.
„Ja, ik zie ’t” zegt de schipper en terwijl hij de juffrouw, die intusschen de
loopplank bereikt heeft, ridderlijk de hand toesteekt, zegt hij goedig: „Kalm an
maar, moeder, me hebbe de tijd.”
De juffrouw is nog niet in staat een woord uit te brengen en zit amechtig naar
lucht happend op een bank, terwijl ze het bezweete welgedane gezicht met
een netjes opgevouwen, hagelwitten zakdoek afveegt.
„Blij dat u zit, juffrouw,” toetert Ambro haar in de ooren, zoo hard, alsof ’t
vanzelf sprak dat de juffrouw doof was, hetgeen ze wederom beantwoordde
met de noodige „hè hè’s” en „mense mense”.
„Dat scheelde maar een haartje,” schreeuwt Ambro, nòg harder dan den
vorigen keer, in de meening, dat zijn eerste gebrul niet verstaan was door de
juffrouw. [165]
Dit blijkt tè kras voor haar trommelvlies en ze keert zich verwoed naar Ambro
om, terwijl ze nijdig zegt:
„Ik bin niet doof! je hoef niet zoo te schreeuwe!”
Na dezen uitval kijkt Ambro haar even beteuterd aan en wijdt dan weer zijn
volle aandacht aan een worm, dien hij in een van zijn mooie goudreinetten
ontdekte.
„Ik bin anders altoos op tijd,” opent de juffrouw eensklaps het gesprek. „Me
klokkie gong sekers na.”
„Wat zegt u, juffrouw?” zegt Ambro, z’n oor vlak bij haar gezicht houdend.
Als ze hem verbaasd aankijkt, zegt de deugniet, terwijl hij veiligheidshalve een
eindje van haar af gaat zitten:
En onmiddellijk laat hij er op volgen: „Moet u een stukkie?” en houdt haar een
stuk appel voor.
Dit verteedert de juffrouw en met een vriendelijk: „Dank Uwes wel, jongeheer,”
neemt ze het partje aan, dat ze luid smakkend verorbert.
Paul, die zich al dien tijd stil hield en zich kostelijk vermaakte met het gesprek
tusschen het tweetal, proest het op eenmaal uit.
„Wou je ’n oud mensch nemen?” zegt de juffrouw lachend tegen Ambro. „Me
jongste soontje is al vijf en dertig. En die heit er krek zoo eentje als deuze.”
„Nou, dan kon ’t uw kleinzoontje zijn. Paul, daar zit je opoe. Lach es tegen d’r.”
[166]
„Je bent me d’r eentje,” zegt de juffrouw, maar ze heeft toch schik in hem.
„’t Is toch … wat e rakker, hè? Sal me ’n lekker dier thuis sijn …! Soo twalef!!”
Dan licht ze zorgvuldig haar japonrok op en haalt uit een grooten witten zak,
een knipbeurs van enormen omvang en grut er met beverige vingers het geld
voor een „karetje” uit.
„De gewone reis, moeder,” zegt de schipper die haar reeds van lange jaren
kent. [167]
„Verkoopt u retourtjes?”
„Ja, ja,” lacht Ambro. „Ik mag vandaag met oome uit.”
„Ja, juffrouw,” zegt Ambro. „Opkomen voor ons nummer, we zijn d’r allebei
ingeloot.”
„Nee,” zegt Ambro. „Ik hou van commandeeren en niet van gecommandeerd
te worden.”
„Maar hij,” zegt Ambro, op Paul wijzend. „Hij is er gek op. Hij kan een heelen
dag met een houten geweertje en een kurk spelen.”
„Nee, hoor juffrouw,” valt Paul hem vlug in de rede. „Hij jokt maar wat, ik moet
er ook niets van hebben, me broer zegt ook dat ’t een nare zooi is en zonde
van je goeie tijd.”
„O, is uwes broer in dienst?” zegt de juffrouw, bij voorbaat ja knikkend met ’t
hoofd.
„Al zeven maanden zit ie d’r in,” zegt Paul. „En nou heeft ie straf; daarom gaan
we naar Delft om ’m wat lekkers te brengen.” [168]
„Heit ie straf?” zegt meewarig de juffrouw. „En mag ie nou niet naar huis?”
„Hij heeft veertien dagen, hoe heet ’t ook weer, o ja, kwartier-arrest.”
„Sa-je gebeure,” zegt de juffrouw, ofschoon ze volstrekt niet weet wat kwartier-
arrest wil zeggen.
„O, daar hei je die pejas ook weer. Van jou geloof ik toch nooit niks meer.”
„Nee juffrouw,” licht Paul haar beleefd in. „Dat is het cachot wat u bedoelt, dat
is voor erge feiten, zegt me broer.”
„Wat heit ie dan wel gedaan, zonder nieuwsgierig te zijn,” zegt de juffrouw.
„Hij heeft aan den luitenant z’n sik getrokken,” zegt Ambro. „’t Was een
aangeplakte, want hij hield ’m in zijn hand.”
„Och malle,” lacht Paul. „Niks van aan, hoor juffrouw. Hij was niet geschoren
op een inspectie en toen heeft een sergeant hem er bij gelapt.”
„Wat een stuk sjagrijn! Ja, se kenne die jonges soo koejeneere,” zegt de
juffrouw vol medegevoel.
„Ja,” zegt Ambro. „Ik wist dat ’t iets met den baard te maken had.”
„Nou, ik bin d’r,” zegt de juffrouw, terwijl ze haar parapluie en mandje ter hand
neemt. [169]
Onderweg moeten zij onder twee bruggen doorvaren. De pijp wordt omlaag
gehaald en ook de ijzeren hekken van het bovendek moeten plat liggen opdat
de boot zonder stooten onder de brug door kan.
De schipper, die rustig zijn pijpje zat te rooken, verlaat nu ook het dek, want de
brug is in aantocht.
Maar Ambro heeft schik in het geval en stelt Paul voor boven op het dek te
klimmen.
„Je zal heusch je kop niet stooten! kom Paul, we gaan plat op onzen buik
liggen, net als de stoompijp.”
Meteen is Ambro al naar boven en Paul, hoewel een beetje angstig voor zijn
hoofd, volgt hem.
„Wat een rare knullen,” zegt de schipper lachend. „Als een verstandig mensch
voor zijn gemak naar beneje gaat, motte hullie juist naar bove.”
„Nee hoor,” stelt de schipper hem gerust. „Als je maar plat blijft liggen en niet
je neus in den wind steekt.”
„Hè fijn,” juicht Ambro. „Hij schuift net langs mijn pet.”
„Tòch gevaarlijk,” zegt Paul. „Hij moet maar eens ’n beetje hooger liggen, de
brug zal heusch [170]niet uit den weg gaan en dan zouden we toch netjes
gekraakt worden.”
„Bê-je nou!” is de verontwaardigde uitroep van Ambro. „De pijp ligt immers nog
hooger dan wij.”
Maar ze zijn er al veilig onder door en Paul herademt. Hij vond het een
benauwde bedoening.
„Straks nog een brug,” zegt Ambro. „Jammer, dat er geen twintig onderweg
zijn. Ik vind het wàt emmes, ’t is weer eens wat anders.”
[Inhoud]
AMBRO REDT EEN SECTIE SOLDATEN.
Delft nadert.—Het weêr is heerlijk, het lijkt wel een zomersche dag. De
jongens zitten weer in de laagte bij den voorsteven van het bootje.
„Kijk es, Paul,” roept Ambro triomfantelijk. „Ben ik niet handig?… ik pak
me daar zoo maar een visch uit het water,” en hij houdt Paul een klein
witvischje onder den neus.
„Maar niet heusch,” zegt deze. „Hij is zoo dood als een pier. Kijk, er
dobberen er nog veel meer. Tsjonge,… kijk es wat een zooi! En allemaal
dood.”
„Da’s van de fabrieke … die vergiftige hier het water met d’rlui
uitwaseminge en dan mot, al wat leeft, sterreve.”
„Nee jonge, da’s niet goed voor je maag, maar de poes, die zou d’r an
smulle.”
Ze nemen afscheid van den schipper en roepen hem een tot-straks toe,
want over vier uur gaan ze weer terug naar Rotterdam.
„Hij is alwéér reusachtig!” lacht Ambro. „Zeg, weet jij den weg?” vraagt hij
dan.
„Goed, burger Paul, alweer gelijk. Ik had anders wàt graag eens een
kijkie genomen in zoo’n soldatenhuis. Ik vraag het ijskoud aan den
generaal.”
„Jôh! de generaal! Die is er niet altijd, een hoogst enkele keer,” zegt Dirk,
„en dan moeten ze poetsen tot ze groen zien en een pluimpje [172]krijgen
ze nooit, alleen straf als er wat aan mankeert.”
„Nou, ik wil en zal de kazerne in,” zegt Ambro. „Wedde, dat we d’r in
komme?”
„Hier heen,” roept Paul. „Dit straatje in, dan zijn we er.”
„Wat zien ze d’r raar uit,” roept Ambro. „Dat is zeker hun daagsche
pakkie, ’t lijken wel clowns met die malle mutsie’s.”
„Ja zeker,” zegt Paul. „Aan den schouder, geweer!” schreeuwt ie. Hoor,
nou roept ie weer „zet af.”
Meteen klinkt weer een harde schreeuw, dat „geweer” moet verbeelden,
waarop alle geweren netjes, tegelijk weer op den grond worden geplaatst.
„Kom mee,” dringt Paul aan. „Laten we nou niet aldoor hier blijven kijken.”
[173]
„Nou, misschien is Dirk er wel bij,” zegt Ambro, die pret heeft in al dat
gedoe. Hij had nog nooit zooiets gezien. Eenmaal in Den Haag, maar dat
was parade, toen waren ze allemaal netjes aangekleed en moesten
wachten op den generaal, die op een paard kwam aanhollen en al de
mannetjes monsterde.
„Kom je nou, of niet,” zegt Paul ongeduldig. „Anders zal ik wel alleen
gaan.”
„Fopspeentjes,” roept Ambro terug. „Vóór jullie naar bed gaan krijg je ze.”
Vóór de poort zit een sergeant op een kapotten stoel een krantje te lezen.
Hij kijkt bij het naderen van het tweetal van zijn lectuur op.
Paul is zoowaar geschrokken van den harden toon waarop hij wordt
toegesproken, maar al gauw merkt hij, dat het niet zoo kwaad gemeend
is.
„Meneer, we zouden graag dit pakje aan Dirk brengen. Dat is zijn broer,
ziet u. Mag dat?”
„D’r zijn hier misschien wel honderd Dirken. Maar waar leit ie? Welke
compie, welke sectie, welke kamer?”
„Dirk Vermeeren, 3de compagnie, 2de sectie, kamer 17,” antwoordt Paul
vlug.
Hij heeft nu weer allen moed teruggekregen nu hij merkte dat de bullebak
geen „kwaje” was.
„Dan gaan jullie ’t zellef maar brenge, hij zal d’r wel zijn, alles is
vanmorrege thuis. Hier de gang in, één trap op en dan de eerste deur
rechts. Ingerukt, marsch!”
„Zie je nou wel,” zegt Ambro zegevierend. „Ik wist het wel, dat we naar
binnen mochten, die kerel was immers veel te lui om dat pakkie zelf te
brengen. Hier Paul, kamer 17.”
„Moet je kloppen?” vraagt Paul een beetje bevreesd in dat nare, donkere,
groote gebouw.
„Bê-je wel wijs?” zegt Ambro. „Van nette manieren weten ze hier niks,” en
meteen doet hij de deur open.
„In orde, staat!” klinkt het op luiden toon en de beide jongens zien
plotseling overal kerels opduiken, die op stroozakken lagen te rooken, te
lezen en te luibakken.
„Hallo,” roept Dirk, die hun die poets gebakken heeft. „Daar zijn de twee
hooge oome’s, heeren, [175]mag ik u voorstellen, mijn geachte broeder,
kolonel Paul en Overste Ambro, het grootste beest van Rotterdam en
omstreken.”
Paul, blij, dat hij eindelijk een bekend gezicht ziet en een bekende stem
hoort, komt met Ambro op Dirk af en reikt hem het pakje over.
„Nou, maar jullie zijn beste kereltjes, hoor!” zegt Dirk die het pakje begint
te openen. „Hoe vinden jullie ’t hier? Gezellig, hè? Ambro, is dat geen
lekker bedje, je komt maar es logeeren, plaats en eten genoeg.”
De jongens kijken hun oogen uit; zoo’n rare boel hebben ze nog nooit
gezien.
„Ha!” roept Dirk blij. „Sigaartjes en Kwatta’s! da’s toch maar alles. Ambro,
moet je een stukje?”
Ambro kijkt met glundere blikken naar het verleidelijke stukje chocolade,
maar hij weigert het en vindt, dat je een gevangene zijn lekkernij niet
moet ontnemen.
„Heb je nou heusch straf omdat je niet geschoren was,” vraagt Ambro
ongeloovig. „En als je nu je baard laat staan; hij kan toch niet in één nacht
lang zijn?” [176]
„Daar houden ze hier geen rekening mee,” antwoordt Dirk. „Of een glad
gezicht òf een flinke baard.”
Ambro voelt eens over zijn wang en kan slechts constateeren, dat er nog
geen spoor van haar te bespeuren is. „Bij mij is gelukkig alles in orde,”
zegt hij. „Laat de hooge-oome maar komen, hij mag me over mijn koonen
aaien.”
„Wat een lefschoppertje,” roept hem een landweerman toe, die evenals
alle anderen, lui op zijn krib ligt uitgestrekt.
„Had ik maar een beetje rook in mijn mond!” Deze verzuchting was voor
Dirk bedoeld en deze gooit hem een van zijn pas ontvangen sigaren toe.
„Hier, ouwe dief, jij ook wat,” lacht Dirk, en dan tot de jongens: „Dat is nou
onze Manus, de grootste lijntrekker van de heele compie.”
En Dirk legt uit: „Lijntrekken is net doen of je druk aan ’t werk bent en een
heeleboel te doen hebt, maar in werkelijkheid voer je niks-niemendal uit
en als ze mannetjes zoeken om het een of andere karweitje op te
knappen … kamers zwabberen, stroozakken vullen of zoo iets lekkers,
dan ben je in geen hoeken of gaten te vinden.”
De jongens snappen het wel, maar zien niet in, dat kamers-zwabberen en
stroozakken vullen zulke nare bezigheden zijn.
„Ja,” zegt Manus. „Eventjes is aardig, vooral voor zulke jonge snuiters als
jullie, maar een heelen middag zou zoo’n grappie je gauw de keel uit
hangen.” [177]
„Zijn jullie nou allemaal lijntrekkers?” vraagt Ambro. „En kan niemand je
vinden?” en meteen wijst hij naar buiten, waar, op de binnenplaats een
sectie recruten een looppasje in ’t rond maakt.
„Oh, dat is de ooievaar,” lacht Dirk. „Ik hoor het aan zijn commando, hij
kraait net als een oude juffrouw, zie je wel, jongens, wat een lange nek
die kerel heeft, dat is „de ooievaar”.”
„Mogen jullie hier nou zoo niks doen en luieren?” vraagt Paul, die bang is,
dat er wel eens ’n hooge oome kon binnen komen.
„Ja, hoor Paul, dat màg, we hebben nachtdienst gehad in de duinen bij
Den Haag en daarom hebben we nu rust, ja, ze zijn wàt bezorgd voor
ons, als het regent mogen we ook thuis blijven, kijk maar …” en hij wijst
op een broek, die stijf is van natte modder.
„Ik heb de jicht in m’n rug van dat nachtelijke tochie,” klaagt Manus. „En
de volgende keer vertik ik ’t, dan piep ik ’m en kruip onder de wol … ze
lijken wel gek,… als een fesoenlijk mensch maft, gaan hullie met beeste-
weer door de duine renne.”
„Ik het toch moar twie kenijntjes te groaze hàad,” [178]klinkt van de
overzijde de stem van een boerenzoon.
„Hei ’j nog tebàak?” roept hij tot Dirk, maar deze, wel goed, doch niet gek,
antwoordt hem: „Een ander keertje, broer, anders hou ik zelf niks en ik
moet nog negen dagen brommen.”
De jongens zijn bij elkaar op de leege krib naast Dirk gaan liggen en
voelen zich langzamerhand thuis in de vreemde omgeving.
„Ik vind ’t toch wel aardig hier,” zegt Ambro. „Alleen een beetje saai op
den duur.”
„Toe, maak geen spas, hou op met die flauwe kul, leelijke kevers!”
schreeuwt Manus, die den kapitein niet in de gaten krijgt
Maar nu hij de anderen doodstil voor de bedden ziet staan kijkt hij op en
ontdekt dat inderdaad ditmaal de kapitein op de kamer is.
„Kop dicht,” roept de sergeant die hem begeleidt en Manus glijdt van zijn
krib en neemt „de houding” aan. [179]
„Wat moet dat,” vraagt de kapitein verbaasd. „Wat doen jullie hier? er
mogen geen burgers in de kazerne.”
„Zie je wel,” fluistert Paul in doodsangst. „We zijn burgers, ik heb je wel
gewaarschuwd.”
„Hoe komen jullie hier?” vraagt hij, terwijl hij naar de jongens toe gaat
Paul weet geen antwoord te vinden, hij is bang geworden door die
doodelijke stilte in een kamer waar een dertigtal mannen als beelden
staan te staren. [180]
„Hij moest een pakje voor z’n broer brengen en we mochten van dien
mijnheer beneden aan de deur het zelf gaan brengen,” antwoordt Ambro
op helderen toon.
„Zoo—en waarom riep jij daar straks „kaptein”, had je me wat te zeggen?”
„Ja kaptein,” luidt het antwoord van Ambro. „Nu Manus,” en hij wijst op
den landweerman naast Dirk, „nu Manus z’n kop moest houden, wou ik u
maar even vertellen, dat ze allemaal vannacht gerend hebben in de
duinen met dat beestenweêr en dat ze daarom, zooals u zei, niks doen
en luieren.”
„Zoo, nou maar dat is wat anders,” zegt de kapitein op zachten toon. „Dus
jullie hebt nachtdienst gehad,… nou, blijf dan maar hier … je kunt je gang
gaan.” [181]
Meteen ligt alles weer op de geliefkoosde krib, als werden ze door een
electrischen schok getroffen, en de kapitein, die de jongens toeknikte,
verlaat even dreunend de kamer als hij binnengekomen was.
„Heere, Heere, kan dàat jong proate,” is de lof van den boerenzoon en
Ambro, die hen allemaal redde van den gevreesden looppas in de zon,
stond als held van het oogenblik midden in de kamer, blij, dat alles goed
afgeloopen was, want Paul had gelijk gehad, ze waren tòch burgers.
„Ja,” zegt Manus. „Had ik de burgerpet op gehad, dàn had-ie naar mijn
ook wel geluisterd, maar as-t-ie je in je werkpakkie ziet, wordt-ie
dienstklopper en dan is ’t „kophouwen en doen”!”
„’t Was tòch fijn,” zegt Ambro, als ze weer op straat staan.
„Ik vind ’t een nare boel, ik zou niet graag zoo behandeld worden,” is
Paul’s antwoord.
„Hoorde je niet hoe ze allemaal het land er aan hebben? Dirk kan
gelukkig weer op zijn kantoor terug komen na den dienst, maar er zijn er
veel die hun plaats bezet vinden.”
„Ja,” zegt Ambro. „Van dien kant bekeken heb je gelijk, maar zoo’n
nachtelijke tocht door de duinen zou ik toch wàt graag meemaken.” [182]
„Nou, dat kan best eens gebeuren. Karel z’n vader gaat van den zomer
kampeeren, je zult zien, dat we dan mee mogen.”
„’t Zou tijd worden,” zucht Ambro. „Onze nachtelijke tocht is toen mooi in
de war geloopen. Maar, over wat anders gesproken, me maag jeukt.”
Het was een fijn dagje geweest en met voldoening kon Ambro terugzien
op zijn heldendaad in de kazerne, waar hij het waagde een „hooge oome”
te trotseeren en de soldaten te redden van den gehaten „looppas in de
zon”.
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade
Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.
ebooknice.com