Martin L. West - Studies in The Text and Transmission of The Iliad-K.G. Saur (2001)
Martin L. West - Studies in The Text and Transmission of The Iliad-K.G. Saur (2001)
Martin L. West - Studies in The Text and Transmission of The Iliad-K.G. Saur (2001)
BY
Martin L. West
December 2000 M. L. W.
i 5
Contents
BIBLIOGRAPHY . . . . . . . . . . . . . .
"J
INDEXES . . . . . . . . . . . . . . . . . . . .
.= =-Q.- =. \;‘
“xx..
15
$3
1..-I
ii
ii
*»:
=‘.->§_§_§_§\;(
-E
:-1
-‘V
;)'i
;i_-Q
:3
l
-4
.-_:,\;..
~‘.i;;.-_- . .
PART I
THE TRANSMISSION
i
r
i
1. The Pre~—Alexandrian Transmission
The Iliacl that is the object of our studies is a written text. It is the crea-
tion, we presume, of a poet trained in the techniques of oral performance
.I
5
and accustomed to deliver his poetry orally; a poet who composed his ver—
.‘i
i ses, not without premeditation, but with considerable fluency, line by line
I
i rather than word by word. However consistently he kept to one version
3% of the stories he told, there will have been variations of scope and of detail
I
between each performance and the next. Fixity could come only when a
text was written down. y
In principle it can happen that the same oral poem is written down at
I
different times, in different places, or from the lips of different bards. This
results in texts that may diverge substantially in the sequence of episodes,
i‘
in narrative detail, or in other respects. But in the case of the Homeric
é§
§
=2 poems there is no good evidence that such divergent recensions were ever
current. Apparently each epic was written down only once, or if other ver-
i sions ever existed, they disappeared at an early date.
Our Iliad took on its definitive form as it was written down.‘ It is my
6.
conviction that the process of writing it down was intermittent and extend-
2
i
ed over many years. In any case it was the completion of the writing down
that marked the completion of the composition. That is where the story
-s
f of the transmission begins.
':»
1:
e
=§
I';= The first complete text probably consisted of a collection of papyrus rolls.
:1?
5'T;2?
(Parchment is also possible.) We should not imagine them to be twenty-
four in number; the division of -the poem into twenty~—four ‘books’, as they
Eli
5;;
are commonly called, or f)0u|Ito16i0u, as they were called in antiquity, was
made at a later time and for a different purpose, as will be argued below.
E‘?
Q
In any case this original text did not yet contain the Doloneia, which
i
i
§
makes up one of the twenty-four sections of the Iliaal as it has come down
i
IIO US.
-1
.'.i
.-g
.<
is It may be assumed that each verse was written on a separate line, accord-
._,
Iii3
ing to the standard Greek practice attested since the time of the ‘Nestor’s
2-i
5:55
;;.>
iii
iii5:.-i
:k'
0
I?
is
.»;
' This essential point was established once and for all by Adam Parry in his famous paper
._i
ziIii ‘Have we I-Iomer’s Iliad?’ in Yale Classical Studies 20 (1966), 177-216; reprinted in ]. Latacz,
iii
_i Homer. Tradition and Nea-era-ng (Wege der Forschang, 463: Darmstadt 1979), 428—66.
.:3_2,
=2
.-3
25:;
E-ii
Iii
4 1.1. The Pre~Alexandrian Transmission
cup’ inscription from Ischia.2 In the mid seventh century the text would
have been written from right to left or laoastroplaetlorz. The latter is com-
moner in the inscriptions, but the three verses of ‘Nestor’s cup’ are all
written from right to left, and this might reflect a practice established for
books of verse. The use of punctuation to divide words or phrases is also
possible, but we cannot assume any other kind of formatting.
The script used would have depended on where the poet, or his ama-
nuensis, learned to write. I accept the usual view that he was an Ionian
who spent his life largely or wholly in western Asia Minor, moving be-
tween the Maeander valley and the Troad. If so, he is likely to have used
an East Ionic alphabet of twenty-five letters (with eta, qoppa, and omega)?’
Even if there is truth in the ancient and modern theory that certain errors
arose in the transmission through misinterpretation of texts in which E
stood for both e and 11 and O for both 0 and to, it need not follow that
such ambiguities were a feature of the earliest texts: they may have been
first introducedin early Athenian copies made in the local Attic alphabet.
I will discuss the question later in connection with the Attic phase of the
transmission.
If we may argue from the usual practice of early inscriptions, doubled
consonants would have been written single, so that, for example, ’A)(i7\e15c;
and ’A)(17\7\e1.5<; would both have been written Axtkeug. The unvoiced velar
stop was represented by qoppa before a back vowel (o or u): I"7\cx1><Pu>Tu61
Qopnt, QMJG1, i<6Qpu<1>0t7\0v. Elided vowels would have been omitted.
A transcription of the opening lines of the Iliad (with word spacing and
capitalization added) would accordingly have read something like this?‘
um/iv 0:516:-: Bea I'Ir|7\r|iot6ew A)(17\r|oc
okouevnv n pup: A)(U.l0lC odwe e0m<ev
atomic 6 1<i>(-Jipoc qmxac A161 rrpoiouuev
npwtov owtoc be 87\(.0[)1(X revxe Quvecw
oiwvoici ts Tl'0(Ct AIOC 6 8T8}\81ELTO Bean
eE 0 on ta rrpwroc 01(XC‘l'I]TI]V epicowre
Arpeionc re on/at avoptov KOU otoc AXIAEUC
.,.,_,M__.........._.--------------—--
2 CE G I no. 454 (cf. II p. 304), c.735—-20 BC. The format was very likely taken over from the
Near East; see West (1997), 26.
3 Cf. L. H. Ieffery, The Local Scripts offlrcliaic Greece (Oxford 1961; 2nd ed. rev. A. W.
johnston, 1990), 325-7. As ]anko (1992), 35 and 36 notes, the survival of the archaic form [5u'5v
at H 238 is a strong indication that the original text used to and that there was a continuous
tradition of texts using ea.
4 Powell (1991), 65, offers a specimen reconstructed in an archaic script, making a number
of assumptions that differ from mine.
‘E.15
I.1. The Pre-Alexandrian Transmission 5
:3-t'€
':€
1.:
-€.3
é.3
1
Dzffusion
ij
The rolls that made up the primal text were initially the property of the
E.4
-<
ii
poet, unless there was some rich patron who had commissioned them.
Their existence was surely no secret. The poet’s public and his fellow rhap~
E
3
sodes5 knew that he possessed them, and they must have been deeply
{ impressed that he had what seemed to be an unimaginably huge epic
x
stored away in them. (It was indeed, to the best of our knowledge, the
é
2
E
longest text that ever existed in Greece before the fifth century.) In all
ii probability they never had an opportunity to hear it entire (let alone read
-s
-ifl
it), but when he recited episodes from it, they were aware that these were
excerpts from a greater whole that he had created and established in fixed
form.
What happened when he died or became incapable of looking after his
211121:w.
books? We know that they were preserved ~—- not just by luck, I suppose,
ii
but because their value was recognized — and that in time copies multi-
3
plied. Several alternative scenarios can be imagined. The rolls might have
ii
Eli
been kept by the poet’s family, if he had one. They might have been bought
-5!
5’:
‘iiEwe by, or willed or donated to, another rhapsode, or a group of local rhapsodes.
.=2
Ii
Or if the poet had a particular patron, he will have kept them. Finally, they
5 I shall in any case need to use this word before long, and I do not think it is premature to
55-;
introduce it here. It is first attested in the fifth century, and oftentaken to mean by definition
L5.
I$:
a non-creative reciter as opposed to an dzoloog. But the ancients do not contrast the two terms,
=a
and the word pottptolfiog itself implies creative composition, if only by stringing formulae
iii
{ii together. Cf. my article ‘rhapsodes’ in the Oxford Classical Dictionary, 3rd ed.
i
iiifit
ii
‘vi;
J.-
6 I.1. The Pre-Alexandrian Transmission
might have been dedicated in a temple. Let us reflect on each of these pos-
sibilities in turn.
1. Preservation in the family. This is perfectly possible, but it simply
defers the question of how diffusion came about. The ‘I-Iorneridai’ who
later claimed a proprietary interest in the poems of ‘Homer’, their sup-
posed ancestor, were not a real kinship group and are not to be identified
as the Iliad poet’s descendants?
2. Preservation in a community of rhapsodes. This is my favoured op-
tion. We have no idea whether our poet left any family, but we can be sure
that there were other rhapsodes in the neighbourhood and that they would
have been keenly interested in acquiring his great written epic. This would
lead naturally to copies being made, and to knowledge of the poem being
spread about as individual members of the group moved away to other
places.
3. Preservation by a patron. A wealthy man who appreciated a good epic
song, but was not himself a performer, would have prized the manuscript
not because it promised him a feast of private reading but because he
would be able to engage other rhapsodes to learn episodes from it and per-
form them. I-Iere again a group of rhapsodes would gain knowledge of the
poem, and this would favour copying and diffusion.
4. Dedication in a temple. This is a theoretical possibility, as we hear
tales (true or not) of books being occasionally so dedicated.7 It would pro-
vide, at least for a time, some security against loss and some public access,
though extensive copying would not be easy unless the sacristan was wil-
ling to lend out the god’s property.
The locality from which the Iliad began to be disseminated was, I would
suggest, the Troad. The poem was named after Ilios not, surely, because its
action takes place during the Trojan War, for that was true of a great deal
of epic poetry in the seventh century, and the title should serve to distin-
guish this poem from others. The Iliad is not, after all, the story of the
sack of Troy. More likely it was named after Ilios because it was perceived
as coming from that district, just as an epic ascribed to a Cypriot poet (Sta-
“~*~i
Z.t
2
§
1
1
$
:>'25
§ I.1. The Pre—Alexandrian Transmission 7
1.i
rides of Phocaea was called the Plaocais, and a third attributed to Carcinus
of Naupactus was entitled Tot N01mtd1<'r101 §Ttn.8 It is clear from his de-
tailed knowledge of the landscape around Troy that the poet of the Iliad
1?
E
was well acquainted with the area and probably composed at least part of
ii
-<
::
the poem there.9 Ilios was in his time no longer a living town; according
'§
to the current director of excavations, Manfred Korfmann, it was a vast
Fi
rubble-field crossed by footpaths, with the boundaries of the old city
ii
.1
"é
clearly outlined and the great walls of the Bronze Age citadel still standing
ii
above ground to a height of six metres. But it was not an empty landscape,
devoid of human activity. There remained a functioning shrine of Athena
Ilias, a holy centre that attracted visitors from surrounding settlements
§ such as Sigeion and Rhoiteion. There may have been people living here and
there in the more habitable parts of the ruins, and some - even then - offer-
-s
:5
.5
=5 ing services to tourists. Most of the local populations consisted of Aeolic
5%
i
in the Hymn to Aphrodite" that there was an aristocratic family some-
'/.
5?:
E;$:
.:§;
where in the region that claimed descent from Aeneas and suzerainty over
.-.,<.
§§E
‘Trojans’, and that our poet was one of a number in contact with this
--4;
a
a family.
J
-is
-.v'
The original Iliad manuscript may have been conserved in this milieu,
among these poets. From here knowledge of the poem could spread to
other parts of Greece. Clear evidence of such knowledge first appears in
.;.q1 the last quarter of the seventh century, in Argive and Corinthian art and
it
2
.'/ : in the poetry of the Lesbians Sappho and Alcaeus.” Lesbos, at least, was
engaged in the Troad at the time. When Phrynon of Athens, who had won
.--4.
as:
Archeanax of Mytilene fortified it using stones from the ruins of Troy.”
.._,,
E555
.2?-
ii.
at
£12’-
8 Cf. West (1999b), 365, where I note that certain Old Norse texts were likewise identified
5;:
.==a_-e
..=;'. by their place of origin, such as Atlakvifla in Groenlenzleo (the Greenland Lay of Atli) and the
92 Flateyjarbéle (Flat Island Boole). According to one view, the title Edda comes from the place-
name Oddi in south—west Iceland, the home of Snorri Sturluson.
9 Cf. Schadewaldt (1938), 125 n. 0; West (1995), 217 n. 43.
1° Strab. 13. 1. 38 (referring to the late seventh century) Acofiitov érrloticodjopévmv oxeoov Tl
"rfig oupationg Tpt010’160c_;' tiw an 1<o1‘1 icrioutxtd e’1o1v on 1r7\e'iot011 T1311 1<0rro11<103v. oii pi-av oup—
31% pévovoon K051 vfiv. oii 6’ 1jc|>(xv1opévo11.
_,.;i
1‘ T 306-8; Hymn. Aplar. 196f.
12 See West (1995), 206-7.
13 Strab. loc. cit.
-‘$
.:\\\:§
8 I.1. The Pre—Alexandrian Transmission
Alcaeus himself fought in the ensuing war, and perhaps mentioned the
shrine of Athena at Ilios.“
When we speak of knowledge of the Iliad spreading, what is involved?
Some knowledge of it could have spread without any diffusion of written
copies. If a rhapsode who had learnt some episodes from the poem tra-
velled to another area and recited them, explaining to his audiences there
that these were parts of a great unified epic that was to be found elsewhere
in manuscript, that would at least create an awareness that such an entity
existed. Indeed, anyone who had heard such a rhapsode perform was in a
position to spread such awareness. The curiosity of another rhapsode or
of a literary-minded tyrant might then be stimulated, prompting him to
seek out, if not the manuscript itself, at any rate those who knew more of
the work.
But there must have been some diffusion of written exemplars too. To
copy the entire poem by hand was a lengthy task, though we know it was
done many thousands of times in the course of antiquity and the Middle
Ages. However, many of the early copies will have been partial ones, par-
ticular episodes that a rhapsode wanted to have and to study. Recitations
were normally of single episodes or sequences of episodes. A complete
recitation of the whole Iliad was hardly possible without organizing a
team of rhapsodes to do it in turns, which would only make sense in the
context of a major festival where a crowd was present for several days
together. We do not know whether this was ever done before Hipparchus
established the practice at the Great Panathenaea.‘5 The challenge was
comparable, perhaps, to that presented by Wagner’s Ring. One may hear
at a concert the Ride of the Valkyries or Siegfried’s Rhine ]ourney, or even
a whole Act; occasionally there comes the opportunity to see one of the
dramas in the opera house; but a complete cycle is necessarily a rare event.
When an operatic ‘highlight’ is performed as a concert item, it is some-
times necessary to adjust its beginning or its end so that it can stand on its
own. The same must have been the case with certain episodes excerpted
from the Iliad for a rhapsodic performance. The rhapsode might well make
his own ‘edition’ of the piece, and it might presumably happen that
someone finding such an exemplar, perhaps with some extra verses added
to make a ‘concert ending’, used it to supplement the continuous text of
l‘ Fr. 401B D1010 Ktémov ipov; Herodotus, however, locates this at Sigeion (5. 95). It may be
remarked that the Iliad’s title would be especially apt if the original manuscript was ever dedi-
cated in the shrine at Ilios.
15 There was no complete performance of Bach’s B Minor Mass until the 18305.
._l
ii
2:3
like a conclusion added by a rhapsode who had recited the Atbq d1Tr0‘t'rn as
-Ii
-2
Z
a separate item. I discuss the passage in more detail in Part II.
isti
The evidence of Attic art confirms that before about 520, when the
:§
Panathenaic recitations began, some parts of the Iliad were better known
Ix
'2 than others, reflecting the popularity of particular episodes with rhap-
sodes.“ Even later in the Classical period it must be assumed that episodic
=§
=s recitation was the norm, and that partial texts were more easily come by
than complete ones. When Socrates in Plato’s Ion (535b) asks the famous
Li
rhapsode how he feels when he is reciting certain dramatic passages -
Ei
:-; Odysseus revealing himself to the Suitors, Achilles rushing at Hector, and
i
i
sias
so on - he is surely not referring to complete performances of the Iliad
-E
E2
-.<
:2 and Odyssey. Aelian actually gives a long list of episodes from the two
-.>
:.-t
1%
epics that the ancients used to recite separately.” This has no historical
ft
ii Cynaethus.2°
13;;
.;_,(,_
iii
;;..<$<
FE
Hg=-is
.3‘
5°‘
ii
it “’ Friis Johansen (1967), 224-30.
iats
‘iii
‘7 Ael. V.H. 13. 14, rfiv :’srr’1 roiig votutfl poiprnv. 1<01’1 Aoatévetdv twot. 1<01’1 tip1.o'tei01v ’Ay0t—
ii3*,
:.v<
-.<<
néuvovog. 1<0t’l veclfw xaroikovov. 1<t7\.
i
"3 Heraclides Lembus, lIOAlT61(Xl 10 (Aristotle fr. 611); Plut. Lycarg. 4. 5; Dio Prus. 2. 45;
i
jig Ael. loc. cit.; cf. Timaeus FGrI-Iist 566 F 127, Ephorus 70 F 149 §19; Apollodorus 244 F 63b;
Merkelbach (1952), 31-3 = (1997), 9f.; Burkert (1972), 77f. = (2001), 141f.
Ea
gt
:13
.1;
'9 [PL] Hipparc/1. 228b rrptiitog éxépioev sit; tfiv yfiv towtnvi; Hippostratus (FGrHist 568
213
'.-:s=
F 5) ap. sch. Pind. Nem. 2. 1c.
2° For I-Iipparchus see below; for Cynaethus at Syracuse cf. West (1999b), 368.
-5-2%
.=;<
Eiti
-i
lg
5%
.2
é
t~.,.-..
_ -¢; §,_.,:-
10 I.1. The Pre-Alexandrian Transmission
ii‘.
diverge substantially in what they contain.” We are fortunate that the tra-
$21.4;
dition of the Homeric poems is so unified. If seriously divergent recen-
:5?
5;:
a.=x-
sions ever existed, they have disappeared without trace.
X Minor interpolations
I’
The line-numeration that we use today is that of Wolf’s edition (1804). He
ii
sari»: included in his text some lines that are in fact very poorly attested and that
A?-
- <<
may reasonably be omitted as being not part of the parad0sz's.2“ There are
=..<
6-.»-
15;?
many others, attested only in one or two sources each, which Wolf did not
W53
=.-.,..
know, or at any rate did not include, and which have to be labelled with
.1“.
,\§-
an a or b after the line number. All of these must be regarded as interpo-
lations. But even of the 15,693 verses of the ‘Wolfian vulgate’, some 166
lines or passages (apart from the Doloneia) are identified in my edition as
6%
Iéiifii
,6.
".':-=><:
W? w6
interpolated on grounds of content, language, contextual coherence,
it
Q"
it
and/or external evidence (absence from part of the tradition without
2/
.-<' obvious mechanical cause).
%
§<
?;‘=¢£
géz
iii
:.%_a§
23 I take the examples from Merkelbach (1952), 34 f. = (1997), 11f. Many others could doubt-
-94¢.
ma».
less be cited.
I;\‘R
24 Cf. Bolling (1925), 3, 9-13. I have excluded the following: A 265; B 168, 206; E 901; E)
52%
:;;.<E
224»6, 277, 466-8, 548, 550-2; A 543; M 219; N 255; E 70, 269; n 381, 614-15; P 219, 455, 585;
..!=<v::
T 177; r 312, 447; tn 434, 460, 510; x 121; W 565, 264; n 558, 693, 790.
215$;
12 I.1. The Pre—Alexandrian Transmission
The passages so diagnosed fall into the following seven categories (with
occasional overlaps):
1. Verses reflecting local or political interests that do not seem to be
those of the original poet. Most of the instances are designed to promote
the status of Athens, or refer to Attic myths or customs. One couples the
Phocians with the Boeotians, which may reflect the historical situation
after the First Sacred War. One gives Sicyon a place in the heroes’ world;
its absence would have been felt most keenly in the sixth century, when it
grew great under Cleisthenes. Another shines a momentary light upon
Pheia.”
2. References to Cyclic material that is otherwise unknown or ignored
in the Iliad. The prime examples are two passages that introduce 1\Ieopto-
lemus, Achilles’ son by Deidameia, and the one that refers to the Judgment
of Paris?“
3. Explanatory glosses. I use ‘gloss’ here in a wide sense, of lines added
to clarify an expression or a reference that might be unclear or ambi-
guous.”
4. Rhetorical expansions, that is, lines or passages added to enhance the
dramatic effect or graphic vividness of the narrative.”
5. Syntactic complements. As in tragedy, an elliptical construction or a
pronoun whose reference might not be immediately obvious was some-
times filled out by adding a line or two to make things more explicit.”
6. Redundant speech introductions. This rubric covers two different
phenomena: (a) interpolation of a line with a verb of speaking immediate-
i.
25 Athens: B 547--51, 558; 1" 144; H 334—5, 466-81?; M 372. Phocians: B 525-6. Sicyon:
B 572. Pheia: H 135. I have discussed some of these in West (1999a), 185-8, others in Part II
below.
26 T 326-37, Q 466—7; Q 29-30. The lines about Phoenix’s youth known only from Plutarch
(I 458~61) may be another instance, if Stephanie West is right in suggesting that they were inter-
polated from a speech that Phoenix made at Achilles’ funeral in the Aetloiopis (Scripta Classica
Israelica 20 [2001], forthcoming).
27 B 535 (explaining which Locrians are meant in 527), 742—4 (genealogical-rnythological
appendix); E 415 (making it explicit that Aigialeia is mentioned as Diomedes’ wife); ® 528
(expanding xqpsooicboplitovg, which the interpolator misunderstands); M 426 (expanding
Boeing); ‘P 92 (giving specificity to the ooptig); Q 763-4 (pedestrian explanation of why Hec-
tor was Helen’s fiotrjp).
2“ Under this head I include B 491-2, 674;\I" 18-20; A 159; E 398—402, 449—53, 487-9, 887;
1'1 196-9; O 73—4, 199, 421--4, 535--7(—41?), 548 + 550-2; I 320, 523; A 540-3; M 14; N 731;
E 49-51; O Z14, 291-—3, 610-14; II 158-65, 242-5, 591, 661--2; P 273; Z 26—-7, 34, 39»-49, 272, 381,
461, 535-8; T 365-8; r 26942; <1> 126-35; x .436; \l1 806, 878; Q 54, 232, 586.
2° F 41; Z 334, 388; G) 183; I 416; K 147; N 114-15, 316; E 40; P 244; T 374; T 312; (I) 480;
‘P 628; Q 45, 558.
E?
i
ii
II
.1
I I.1. The Pre-Alexandrian Transmission
ly before the speech, where the act of speech had already been sufficient-
13
2%
iiit ly indicated; (I9) interpolation of a vocative as the first line of the speech
.5
itself.3°
7. Concordance interpolations. This useful term, introduced by Bolling,
Z>.N't1n 0l
refers to the frequent tendency to insert a line (or two, or at most three) that
§§
It
‘.9
Q
the case with the Doloneia). An interpolation dating only from the fourth
or third century, on the other hand, would be less likely to infect the whole
tradition. Now, when we examine the interpolations assigned to the seven
categories described above and inquire to what extent they are present in all
sources and to what extent their attestation is defective, we find remarkable
ii
8 differences among the categories, as the accompanying table indicates.
3
5a
Number of Number absent from
i
2 interpolations part of the tradition
1. Local-political
2. Cyclic material
3. Explanatory glosses \1-P-\O
4. Rhetorical 44
5. Syntactic complements 16
2:11-8
6a. Speech introduction 12 r-B
'1-21.-'= .1:I<rim»
6b. Prefixed vocative 3 l\J(j)'\.l'\Il\J*- ‘(D
7. Concordance interpolations 70 62
==':?§fi
3° (a) F 389; A 337, 369; K 191; N 480; P 219, 326, 585; T 82; (D 73, 471 (this can also be con-
sidered a gloss); X 81. (b) H 234; (:) 185; N 255. On (a) cf. G. M. Bolling, CPI? 17 (1922), 213-21;
;=._g.
E53?
Apthorp (1980), 150-2.
_ *5
;5=a_;;
31 Bolling (1950), 3 n. 6. Apthorp (1980), xxiv, modifies ‘scribe’ to ‘scribe or reciter’. The lines
. .%_
-‘Jig.
in question are: B 168, 206, 703; F 78; A 60-1, 504; E 42, 57, 457, 540, 783, 808, 901; H 315,
368-9, 371, 380, 439; O 6, 59, 123, 224-6, 277, 315, 383, 458, 466-8, 557-8; I 221, 694; K 386,
13%
409-11, 531; A 662; M 6, 219, 449; N 131, 749, 832; E 70, 269, 420; O 266--8, 481, 551, 562, 578;
II 381, 614-15, 689-90; P 455; E 200-1, 427, 441; T 177; I’ 135, 316-17, 447; ‘I3 158, 434, 510; X
121, 316, 363; qt 565, 864; o 519-21, 693, 790.
_.
14 I.1. The Pre-Alexandrian Transmission
It must be observed that a special factor in involved in the very high pro-
portions of defective attestation given for types 6 and 7. These types of
interpolation hardly betray themselves except through defective attesta-
tion. A few of the concordance interpolations have only recently been
revealed as such by the appearance of a new papyrus that is free from
them. There are no doubt many more that have not so far been detected.
Even so, it is clear that many of the speech introduction and concordance
interpolations do not go back to the earliest stage of the transmission, and
that these types of interpolation continued to proliferate at a period when
there was no longer much chance of their dominating the tradition. By
contrast, the local—political interpolations seem on this evidence to be
early, as we should expect on general grounds.” So do most of those in
categories 2-4; in certain of the instances counted here as ‘defective attes-
tation’, it consists only in absence from Zenodotus’ text or one of the ‘city
editions’, the verses being present in all other sources. In category 5 the
proportions do show a shift, suggesting that this type of interpolation
began in the archaic period but continued for some time afterwards.
wi-
32 One or two later ones, such as the Megarian version of B 557 f., are recorded in the indirect
tradition but never had any currency in the direct, and found no foothold in the Wolfian vulgate.
33 Cf. Haslam (1997), 69, ‘the entire text was unstable, showing a degree of volatility more
characteristic of texts whose transmission is oral’.
3" See Haslam (1997), 74-7.
I.1. The Pre-Alexandrian Transmission 15
i-E
ii
How are we to explain this textual instability in the book texts? It is dif-
"i5%i
=2
ficult to account for on the assumption we might most naturally make,
that every copyist of the Hind had a previous copy in front of him, and
that the sequence of direct copies led back to the original. Evidently early
Homeric manuscripts were not always the work of scribes dependent on
Iii
'§
‘.:2-Ii
what they saw before their eyes. Normally a copyist would have had a
Eli
good familiarity with the poems, and it is easily imaginable that he might
in places take his eyes off his exemplar and Write out whole stretches of
the narrative from memory, introducing echoes of similar passages else-
ii
where, whether unconsciously or because he thought it would be just as
good so. Some manuscripts were probably written by rhapsodes, whether
for their own use or at the request of others. A rhapsode held the text of
entire episodes, or a form of it, in his head. When he had occasion to write
_é
'-:1?
._§
2
it down, he might well rely on his memory and write it in the form in
ii which he had most recently recited it, with the oral variants and extra lines
2-é5%
=;§;
that had come to him while reciting; perhaps making a few further
i embellishments or unconscious changes as he wrote. This hypothesis, I
-iii-'26
:..,
_:_¥
.2
believe, accounts realistically for the prevalence of the ‘wild’ texts down to
f
early Hellenistic times.”
12/
The Homeridai
air
P2:
:;5§
Hesiod (Op. 25--6) remarked that
i i‘?
E.-5
icon icepauebg 1<eporu:-;'i Koréei K061 TéKTOVl TéKT(.0V,
:-2;"
icon rrtwxog it'rw)<o'51 cbflovéei Kori doiobg 61016651,
522
£- and it may reasonably be supposed that a community of any size would
.-:-:<.:._;.,
,¢.
3:§;
have had more than one rhapsode competing for its attention. In the fore-
going pages I have postulated the existence in the archaic period of local
iii
21¢ groups of rhapsodes, without committing myself to anything in the nature
fr;
of organized corporations.
§
335$
ii?‘
From the fifth century on, however, there are references to what has often
been understood as some sort of rhapsodes’ guild: the Homeridai. They are
II-\I1I
1-ii‘:
P-.i::
particularly associated with Chios. We gather that they recited hymnic prooi-
Eh/.
mia and epics, which they attributed to their supposed ancestor ‘Homer’.
is
.\'.,-w_.)~ They also told stories about his life, and were concerned to spread his fame.“
z-.:“é-
51$
.=_:
3'5 We find variants of similar character in the transmission of Babylonian poems between the
.____,,.
=-we14:;
i
3
Q.
late second and the mid first millennium. There too the reason is presumably that the copyists
T
€~ knew the texts well from hearing or actually singing them. Cf. West (1997), 601i.
i=Dfi.
.»2*= 3“ Acusilaus FGrH:'sr 2 F 2; Pind. Nem. 2. 1-2; Hellanicus FG'rHist 4 F 20; Pl. P/oaedr. 252b,
3";'&»‘§
alga Ion 530d, Rep. 599e; Isoc. Helen 65.
my:
=-1.4:
.o\$:£.
..~9 .
1§E"é1'
.3.
2%; -
ai=,§:::. -
I
16 1.1. The Pre-Alexandrian Transmission
l
Chian Homerid who composed the Apollo hymn and pretended that it
was by Homer fits well with the epilogue of the Delian hymn (166-76),
where the poet appears to present himself as a blind Chian who roams
from city to city and whose songs are acknowledged as supreme after he
has once sung them. That he was not in fact blind seems reasonably clear
-2
-.¢
. <. free-standing records of the past, but as the artistic creations of an indivi-
Erf:
it.
Is:
..,,
3E
dual, a Xiog do/lip (Simon. eleg. 19. 1). Homer becomes an object of histo-
.5.
E-i ‘;
:-
rical curiosity, literary criticism, and biographical romance.
5“
5
iii‘-
iii:
-x.
W
Hipparclms and the Great Pamztloenaea
é._.>,.
Bi
=5:
-.~;.
~‘-“§:
When I spoke of the Chian Homeridai being associated with an event of
2%
‘f<'s" !i:'-i1i>i
outstanding importance for the transmission of the Iliad and Odyssey, I
was referring to the inauguration of the Athenian custom by which both
=2;
:v
:5
=;<._._,i
.-5
»..
poems were recited in their entirety every four years at the Great Pana-
ea thenaea. According to the author of the pseudo-Platonic dialogue Hippar—
-23;:
clans, the practice was instituted by Hipparchus, the son of Pisistratus. This
§§£=I=
3'-.5;
=-5;
2
account is more deserving of credence than the version which involves
i
=e
Pisistratus himself, for three reasons. Firstly Pisistratus was the more
.3.
{-
famous figure, and great things inherited from the era of the Pisistratids
£
were liable to be credited to him. Secondly Hipparchus is known for his
.2 interest in poetry. Anacreon, Onomacritus, Lasus, and Simonides all had
s‘
-~
;Ys=
,/-M
=.,,.:
11;.-2
18 I.1. The Pre-Alexandrian Transmission
him as their patron. Thirdly, and most tellingly, it is during his rule,
around 520, that Attic vase-painters suddenly start to show an equal
acquaintance with all parts of the Iliad instead of just the last third.“
Everything points to this being the time when the Panathenaic recitations
started. As 520 itself was not a Panathenaic year, the festival in question
was most likely that of August 522.43
The 522 Panathenaea came a little over a year after Polycrates’ Delian
festival. Polycrates had died in the meantime. One poet at least who had
enjoyed his patronage in Samos, Anacreon, went to Athens and attached
himself to Hipparchus. If a Chian Homerid proclaimed the greatness of
Homer in 523 at a pan—Ionian festival which must have been attended by
many Athenians, and in 522( P) Hipparchus augmented the greatest Atheni-
an festival with recitations of the entire Iliad and Odyssey, identified
henceforth as the poems of Homer,“ it is hard to believe that the two
events were unconnected. It is a plausible hypothesis that the Homeridai,
being in possession of texts of the Iliad and Odyssey, claimed these two
poems for their pretended ancestor, and that, following Polycrates’ death
and the eclipse of Samos as a centre of poetic patronage, Hipparchus
invited them to Athens to assist at what was possibly the first complete
performance of either epic.
The recitations called for the collaboration of a team of rhapsodes who
took turns to perform successive sections of the poems.“ Clearly it was
necessary to assign each rhapsode’s portion in advance, and this is surely
the origin of the twenty-four f>cxt|:w16iou, literally ‘recitations’, into which
each epic was traditionally divided.“ The prestige of the Panathenaic reci-
tations and the dominance of Athenian books in the pre-Alexandrian
transmission led to these divisions being regularly marked in the margins
of rolls and acquiring a lasting status.
,,Mw
i
21
.;;1. I
I
Why were they twenty-four in each poem? There must have been some
E
stronger reason than that this happened to be the number of letters in the
‘€
Ionic alphabet (which was, after all, not yet in common use at Athens).
i
1
The governing considerations were that each segment needed to be of a
E convenient length for a single rhapsode to manage, and that each poem had
.
ii
to be distributed in an orderly way over the festival time available.
E21
I3
According to two late scholiasts, the Great Panathenaea lasted four days.”
5%
:33
53
55
If the Homeric recitations were spread over all four, as seems likely, the
:-3
.§
The establishment of the Panathenaic recitations was an epoch-making
i
event in the history of the Homeric transmission. It enabled people for the
I
2’ first time to become familiar with the whole Iliad and the whole Odyssey. It
2
i
fixed once and for all the attribution of these two epics to Homer.
It imposed on both of them the canonical division into twenty-four f)ou|Jw1—
i
Siou. And it was one of the factors that made Athens the great centre for read-
2
Z
3'
ing, copying, and debating about Homer for the next two hundred years.
.§
:;=§~
Athens: the emergence of a reading pahlic
1:11.
-"-":1
='-'-I21 Formal schooling was available for Athenian boys from early in the fifth
Iii century, if not before. Once they had learned the basic skills of reading
and writing, they practised them by studying excerpts from poets, above
5 1%
iia
i
all, it may be assumed, Homer.“ A famous cup by Duris, dated to the
4818's, shows a schoolroom scene with a boy reading an epic verse from a
:14: roll held by the teacher.” A character in Xenophon’s Symposiorz, Nicera—
..,,;
tus, had a father who held such a high opinion of the moral value of study-
ing Homer that he made him learn the whole of the Iliad and Odyssey by
I3-0'
5-..;
EEE=>.
-.1.»
--'.-‘C
=1-3 !?
=:...,.;'§l§
heart. This was evidently an unusual accomplishment for laymen, though
commonplace for rhapsodes.“) But most people will have had a general
III €
familiarity with the epics, not just from school but from the frequent
.-I5 €
-'=.?.5.
.1.-g
e
opportunities of hearing rhapsodes recite. Niceratus says he listens to them
=1-1;
i
ji practically every day: a colloquial exaggeration, no doubt, but we must
take it that they could indeed be heard on many occasions.
1:5?
1§=§='i§
--'2:
iii5;.-:1!
;¥§1,l ‘*7 Sch. rec. Eur. Hec. 469, i. 336 Dind. (not in Schwartz); sch. Aristid. 98. 31f. Dind., cf. 99.
-E'§I': £l
6, 196. 30f.; L. Deubner, /ittische Feste (Berlin 1932), 24.
ii
gig
48 Cf. Pl. Prot. 3256.
1" Berlin P 2285, often reproduced, e.g. in West (1992), pl. 11. Cf. Burkert (1987), 56-7 =
(2001), 216-17.
5° Xen. Symp. 3. 6.
@mm_i
'§
-a
IE
-e
‘iii.-.-Y
. ._i
:':§
-5
‘F2?
Homer in the Attic alphahet?
i
§'§
xi
xii
The Ionic alphabet which used H and O as long vowels became increasingly
'.'.'\¢
familiar to Athenians in the course of the fifth century,” and it was adopt-
5::
21E
--<-.
¢>n»: 2\'-w m-1<~
ed officially on the urging of Archinus in 403/2 (not 404/3 as sometimes
stated). I argued at the beginning of the chapter that the Iliad is likely to
vii
i-3%
if-Bi
F
have been Written from the start in this type of alphabet. We must, how-
iii
ii ever, reckon with the possibility that in the period between Hipparchus
.I,_
and Archinus Athenians who were accustomed to use E for 6, if (81), and
3%
I], O for 0, 5 (OD), and co, and H for the aspirate, may have used this Attic
system when they copied texts of Homer. This would be particularly
_.-5:=,.?_ understandable if What they were doing was not so much slavishly tran—
scribing an existing (Ionian) exemplar as using it as an aide-mémoire to
is
"‘in
::1~IzF' '
..4.
recall the sound of the Well-known verses and then Writing these down as
;':'
553:.
-:=:'<
§§E1t.2.
they heard them in their heads. Certainly verses depicted on book-rolls in
'='&Ii
vase paintings are Written in Attic (or mixed Attic~—Ionic) script.
...,.
=;-_~ {I
2:-it
Some critics in antiquity cleverly argued that certain errors in the text
$25.
'.-I= :-
:5};
had arisen through a p6T0t)((1pa1<'mp1op6g from Attic to Ionic script, when
2-,,
:4; an ambiguous E or O Was wrongly interpreted.” The argument has been
.-47;
was
r
.-:5 i
553i
=-; 2
2211i
57 Alexis fr. 140 K.~A.
58 Eupolis fr. 327 (K.—A.); Ar. At). 1288; Aristomenes fr. 9; Nicophon fr. 10. 4; Theopompus
..,
..<,
Com. fr. 79; Pl. Apol. 26d; Aristotle fr. 140; C. Dziatzko, RE III 974i.; E. G. Turner, Athem'-
15:3:
iii
an B00/es in the Fifth and Fourth Centuries B. C. (University College London 1952), 21; Pfeif—
-...=
:§;'=‘§
fer (1968), 27f. There seems to have been overseas trade too. Xenophon’s men saw a consign-
1V ment of books among the cargo of a ship wrecked at Salmydessus (Anah. 7. 5. 14).
=5-ii
2:2:-5
5" Threatte (1980), 26-~45; (1996), 688i.
iii
6° Sch.T H 238c2; Ariston. A 10421‘; sch. (Porplm?) E 2410; Porph. ad (D 127 (287. 25 Schr.);
sch.“/h94 (I) 363; sch. 0: 52, 254, 275; sch. Pind. Nem. 1. 34bc; sch. Eur. Pho. 682; Galen In Hipp.
De officina med. xviii(2). 778.
22 L1. The Pre—Alexandrian Transmission
.i
6‘ Cobet (1876), 289-98; Lehrs (1882), 350; K. Brugman, Ein Prohlem der ham. Textkriti/e
(Leipzig 1876), 20 n. 1; Wilamowitz (1884), 301--7, 323-7 (strongly negative); R. Herzog, Die
Umschrifl der alteren griechischen Literatar in alas ionische Alphabet (Basel 1912); Cauer
(1921—3), 99--110; Chantraine (1942), 5—13;]. Irigoin, L’histoi2'e da texte de Pindare (Paris 1952),
22-8; G. P. Goold, TAPA 91 (1960), 272-91; A. Heubeck, Arch. Hom. X 165—7; ]anko (1992),
34-7.
62 West (1978), 61f. The example which I cited there as the only convincing one in Homer,
i<0upoo(o)étov at T] 107, must indeed be a case of mistranscription, but it need not involve Attic
spelling, as an early Ionic text too would have had O for 6 < *0(F)e.
‘*3 Cf. Leaf ad loc.; Chantraine (1942), 458.
64 Wackernagel, Bezzenhergers Beitnige 4 (1878), 266 = (1979), 1519; West (1998b), xxxiii.
3
i»
Trtepét vivet’. Contracted niire for fiiire is not otherwise found, and we
2
:5:
-A should expect lip:-30g to be trisyllabic. It seems therefore that eiite is cor—
J‘
-1:
1- g.
rect in these two passages, but that its unusual use as a comparative rather
-e
=;-".-
than as a temporal particle led some scribes to make it into the expected
-I I91 nvte. However, they might have done so without an Attic text being
§
involved.
At Z 291 all sources give §;7rtIr7\d><; siipéa Trdvtov, although Apollonius
=15.-ij
Dyscolus (sch. F 47a) saw that -7rJ\o15g would be the predicted form. It
52%
:9 would have been written ~flAOZ in either an early Ionic or an early Attic
exemplar. The interpretation ~1r7\t6<; ‘peut étre dii a l’analogie des autres
formes en t0’.65 But someone copying from an Ionic text would have been
.-g
Sophistic criticism
From the time when the Chian Homeridai reinvented Homer as a figure
2:-ii; of flesh and blood, he became an object of consuming interest. There was
much comment on his works, some of it admiring, like that of Simonides
..r
zzii
.
i }-52
and Pindar, some critical, like that of Xenophanes and Heraclitus. The
negative criticism focused on the treatment of the gods in the epics, and
steps were quickly taken to try to reconcile Homer’s apparent deficiencies
in this regard with his recognition as a poet of the highest standing and
authority. Even before 500, Theagenes of Rhegium wrote a book discus-
sing Homer’s theology and justifying it by means of allegorical interpre-
tation. He is said to have lived in the time of Cambyses: that means in the
time of Polycrates, just when the Chian Homeridai were promoting their
claims about Homer. It is not impossible that Theagenes actually lived in
Polycrates’ Sarnos, like his fellow-Rhegine Ibycus.
The allegorical approach was carried on in the fifth and early fourth cen-
turies by Stesimbrotus of Thasos, Metrodorus of Lampsacus, Diogenes of
Apollonia, Democritus, and Anaximander of Miletus (the younger)."7
Various other sophists discoursed on Homeric topics, identified problems,
and offered solutions. We can mention Protagoras, Hippias of Elis, Hip—-
pias of Thasos, Glaucon (of Athens?), Antisthenes, Alcidamas, and later in
the fourth century Zoilus of Amphipolis, Heraclides Ponticus, and Aris-
totle.“ Rhapsodes themselves, including the Homeridai, lectured on
Homer as well as reciting the poems.°9
We are not concerned here to write a new history of Homeric philolo-
gy, but to assess the relevance of these developments to the transmission
of the text. One effect of them must have been to stimulate further inter-
est in the Homeric poems and thus the further multiplication of copies.
Another was to promote criticism that could lead to alterations of the text.
In the anecdote about the schoolmaster who had ‘corrected’ Homer on his
own initiative, Alcibiades’ response implies that what was in question
was not simple correction of obvious copying mistakes, but changes such
as a sophist might propose on the basis of critical discussion of the con-
tent.
On the whole the sophists seem to have resolved their own difficulties
by the arts of interpretation rather than by questioning the text. The alle-
gorists found hidden meanings that alleviated the apparent improprieties
of the gods’ conduct. The Thasians Stesimbrotus and Hippias tried to solve
supposed problems by over-ingenious rearticulation of the written words,
reading irrou/Tot in O 189 as Troivt’ ii, Sioouev in B 15 as an imperatival
_,,,,,_,i._..-...---.-----i-
X
.3;
.:
.4
Ht
:,;
I.1. The Pre-Alexandrian Transmission 25
El
.-_-:
.35
1-;
infinitive ‘oifiduev, and 01> in ‘P 328 as 013.70 Metrodorus dealt with the
-.4
puzzle at K 252f., where Odysseus appeared to say that the night was
more than two-thirds gone but a third remained, by declaring that 1T7\8LOV
meant not ‘more than’ but ‘full of’. '
The schoolmaster, however, ‘corrected’ his Homer, whether by altering
the wording or simply by deleting verses that he found unacceptable. The
i verb is o1op0o13v,’and the implied ideal is op66Tng. The critic who required
i
<5p66'rng did not necessarily have in mind a restoration of the text to its
historically original state; he may have been thinking rather of ‘correct-
. _j
ness’ of subject matter or language. When Plato makes Protagoras state as
551
.-_.:g
his manifesto
-E:l
fivofauou évtb dvopi xouoeiotg uéviotov uépog e’ivou ‘trepi z’-zrrtfav oeivov elven‘
'-Ii
..-;.-.-3
Eonv at Toiito T0! {mo rtbv rroinnliv Aevduevcx oiév T’ elven ouviévon d T6
dp0o3c rrexoinrou KO.l éi mi. icon értiotoiofion oieheiv re Kori i:pwrt£5pe—
{iii
1-ii
vov Aévov 6o13vou, e
i
i
-1 :_;§
Ii-1:1?
iii
he is not speaking of textual criticism as we would understand it. Similar-
ly, Socrates sees it as a matter for informed judgment whether or not
E:';§
Fifi
iii
ii
1\lestor’s advice to Antilochus on tactics in the chariot race is ‘correct’ or
i
=.=.:t
not, cite opfitfnc 7téYE.l "Ounpoc dire ufi." "Opnpov 51op6oi3v, therefore,
p
i might well mean at this period not to correct the received text of Homer
i
if-'-ii
with the aim of restoring the original text (which is what it means in Alex-
11:; andrian and later times), but to correct the errors of the poet himself.
1.1;
»':I:.§:
When Timon advised Aratus that for a reliable text of Homer he should
ii
£512;
consult rd dpxaia dvtivpaclnot and not T0£ fion oiwpfitopéva, we cannot be
‘;=:=§
Elli?
sure which type of oidpfitooig he had in mind.” Nor is it obvious what
sort of ‘correction’ Aristotle practised on the copy of the Iliad which he
1:;
supplied to Alexander.”
-.--a=
;;;.>.~<><
It was the rhapsodes and schoolmasters rather than the sophists who
5:291
iii‘ were liable to make actual changes in the text in response to the difficul-
:=.;<-.
55-ii
-~-‘.=
‘.=1-.‘--551;;
vii
1~»
E???
.=.¢».
ézii
7° It is a point of great interest and significance that this approach takes the written traditi-
8.
2:53;;
aw:
on, not the oral, to be primary.
7‘ P1. Prot. 339a; Ion 537c. Protagoras invented the catchword épfioérreici (Pl. Phdr. 267c),
which Democritus also used (DK 68 B 20a); cf. Ar. Nah. 659, 679, Pl. Crat. 391e. Prodicus too
. .,.
.5
gave instruction Trepi ovopdttov opfldrntog, Pl. Crat. 384b, Eathyd. 277e; Hippias of Elis Trepi
|5u8t1t3v 1-toil dcppovitfiv Kori vpouiuoittov épfiotnrog, Hipp. min. 368d. Cf. Pfeiffer (1968), 39f.,
74f.
an
=;= ;
72 D.L. 9. 113. It is often assumed that the reference is to Zenodotus.
r.-.l,.
H.‘ .
... .
;=.-,:;
73 Plut. Alex. 8. 2 Kori rfjv pisv ’I7uo'z6a. tfig flO7\BplKfiQ dpetfig ixbéorov soft vopifiwv K051
2:9;
évopoiltov. €7toi[3e pisv ’Ap1ototé7t0v<; Slopfltiioowrog iiv ex toii voipfimcog |<0O\oi3ow. eixs 5’
dei permit T06 éyxeipioiov xeipévnv (mo to xpooxetboikaiov. dag ’Ovnoii<p1toc_: iorépqxe
(FGrH.ist 134 F as). cf. Pfeiffer (1968), 71f.
852:
3 535215
'-
26 1.1. The Pre-Alexandrian Transmission
ties and criticisms that were being raised; they were the sort of people who
made new copies. In any case, for a sophist it would have been a confes-
sion of failure if he could not find a solution to his critical problems except
by altering the text, whereas a rhapsode, persuaded by the sophist that
something was unsatisfactory, was more likely to feel that the text was his
to edit as he pleased.”
We have no clearly documented cases of pre-Alexandrian emendation
on the basis of moral or aesthetic disquiet. However, among the hundreds
of readings recorded from Zenodotus’ text, which (as will be argued in the
next chapter) are to be regarded as in the main inherited from older exem-
plars, there are a certain number that may represent emendations of this
category:
A 47. Phoibos goes down vuxti i:7\uo0eic instead of vvicti i:ou<o5<;. A
deliberate change to avoid assimilating the god of light to the darkness?
Or just a prosaic clarification of the idea that he came under cover of
night?
B 111-18 (or 112-18?) omitted, perhaps to avoid the imputation that
Zeus had given a false promise and contrived a I-<(XK1’] duttitn.
F 423-6 reduced to the single line on3tfi 6’ dvriov "iCev ’A?teEoiv6po1o
éivomroc. According to Aristonicus, who is presumably reproducing Aris-
tarchus’ discussion, Zenodotus did this because he thought it drrperrég that
Aphrodite should carry a chair for Helen. But it is unlikely that Zenodo-
tus left behind any discussion of things that were absent from his text/5 If
the motive for the abridgment is divined correctly, it was probably the
work of an older rhapsode. Criticism of Homeric passages on grounds of
dUTpéTf81.0t goes back to the fifth century.“
The same applies to A 88-9, where, instead of Hdvootpov dvrifieov
Btcflpévfi, éi trou écbeupor I niiipe Auxdovog vibv dtuiuovoi Te icportepév re,
Zenodotus’ text (like the second—century BC papyrus p41) had simply
Hoivooipov dvti0eov oifinuévn, siipe 5?; rovoe (or perhaps 'r6v ye). Aristo-
nicus explains that Zenodotus thought it inappropriate for a goddess to
‘look for’ Pandaros ‘in the hope of finding him’, as if there were some
chance that she would fail. Again, if this was the actual motive behind the
variant, a pre-Zenodotean rhapsode will have been responsible.
i”
74 Note, however, that Aristarchus’ athetesis of T 269-72 is reported to have been anticipa-
ted by ‘some of the sophists’: rtponfletofnvto nap’ évioig to3v oocblotiiiv (Aristonicus(?) ad T
zesty
75 Even where he athetized, we find Aristonicus having to guess at his reason (B 553-5).
7“ Democritus DK 68 B 23; cf. Theagenes DK 8 A 3.
1.1. The Pre-Alexandrian Transmission 27
=;;=a.=
;===€,
-==.<,=
A 123 and 138 ’AvT1u0't)(oi0 1<0u<6<|>povo<; for ooiidppovog. The variant
denies the laudatory epithet 5(x'i<1>ptov to Antimachus, who had accepted a
a
4: bribe from Paris (124) and advocated killing Menelaus When he came into
5 Troy as an envoy.
7 M 450 omitted, perhaps for the same reason that led Aristophanes and
i
r Aristarchus to athetize the line, namely that Hector’ feat in lifting the
2.
§~
Q‘
mighty boulder is diminished if ‘Zeus made it light for him’.
._.,{__
1"-‘§‘
where he was on the Trojan plain, not on Ida.”
;v_;~= '
77 This is the interpretation presupposed by Aristarchus’ criticism (sch. U 666ab, cf. 677). Cf.
.=\'-<K-.-
‘=12? .
Nickau (1977), 141i.
--:'.-W
.-.\<...
i>.¢EEI '.
-'.\~:§~_'_-- . .
§.l..a. -
"V5554-
$1.. .. .
I .' 3.
-mi
=a,-';;= I.--3 1.";
7;
_ -.1. _-_
.4: :-:-'. -
Iliz.--
¢},~ ~_._ ._ .'
»»a?.2:. --
28 I.1. The Pre-Alexandrian Transmission
7\ 245 omitted. It may have been thought absurd that Poseidon, after
having assumed the form of the river and submerged Tyro, should undo
her girdle and leave her asleep, apparently under water.
u 290. Eurylochus is afraid that storm-winds may wreck the ship Bediv
dérnrt civoixttov. Zenodotus’ text had ¢>i7\tov déicnti étcxipwv, which avoids
(at the expense of introducing banal sense) the suggestion that storm-
winds might blow contrary to the will of the gods.
Modernization
I have emphasized that the great majority of the new readings that ap-
peared in the pre-Alexandrian tradition were not deliberate alterations but
casual corruptions, mainly ‘oral’ variants, and I have described the pro-
cesses by which they arose. One type of casual change that is character-
istic both of oral and of written traditions, if they extend over many
centuries, is modernization. Archaic vocabulary and morphology tend gra-
dually to be replaced by more familiar equivalents. Of course an enormous
number of archaic words and forms survive in the Homeric vulgate. But
sometimes they survive only as variants in older sources, showing the
modernization process at work.
If an archaism appears as a variant in one place, we should not only take
it to be the original reading there (so long as there is no reason to object
to it), we should go further and infer that the poet is likely to have used
the same form in parallel circumstances elsewhere, even where only the
more modern alternative is attested. For example, in a papyrus at A 559
and in Zenodotus’ reading at B 4 we find residual evidence for a mascu-
line accusative plural Tro7\t7g; otherwise, where the metre requires an iambic
form, the tradition offers trokéotg or 'Iro7\eig. Tro7u3<; is linguistically the
oldest of the alternatives (< ’i'-tivg) and, once lost, could hardly have been
restored; -éotg and -61; will be later analogical forms.” We must regard
Tr07u3c_; as the original reading at least in those two places, and as it makes
little sense to suppose that the poet sometimes used J1'o7\i3g, sometimes
Tl'O7\60tC_; or Trokeig with the same metrical value, we should restore 1T07\i3f;
throughout.” Another instance is the word énioowrpov, ‘tire’, literally
‘what goes on the felloe (ou3'rpov)’. It occurs in six places, always with the
less well supported variant otrloo-. eni is an old form of the preposi-
i*
78 See Schulze (1933), 316; Chantraine (1942), 221; B. K. Braswell, A Commentary on the
Fourth Pyt/aian Ode of Pindar (Berlin-New York 1988), 135f.
79 Except perhaps in interpolated verses. A later poet may have used a later form.
6
tion erri, well attested in Mycenaean. ofrioototpov must be the proper form
of the Homeric word, as there is no reason why anyone should have intro-
.,,,,.
duced it in the course of the tradition; érrioourrpov is a modernization.
E81
59f
Modernization continued gradually throughout antiquity, and the depth
.,,I?
to which it has affected the tradition varies accordingly.8° Some changes
took place so early that no traces of the original forms survive. Examples
‘Nil
=:‘=<:
K‘
at
are the replacement of genitives in -oo, which are required by the metre
in a number of places,“ by -ov, and of Koitieoov, Kafiéoootg, Koifieoav by
r K6616-.82 Another case may be the replacement of ’ii§ev ‘was’ by {iv in over
thirty places, unless this development had already occurred when the [Ziad
i
,.
I->': was first written down.83’ In other cases, as with Tro7\1'3<;, the older form is
5,5:
.34:
~_§;:
‘$85
second half of the sixth century. In the fifth, the practice of representing
5%.
lengthened 8 with the ‘spurious diphthong’ 81 started to become common,
,.r.-
and on for 6 followed a little later. Spellings such as eiv, E:-;'ivoc;, [3c>'i?t3\e1v,
éiveicot, oiivz-:1<or, Antofig. oiivouot. Offluiurrov, [3ou7\r‘|. date from this peri-
od: for the first couple of hundred years of the Ili¢zd’s transmission
they had been spelled ev, ljevog, [3oz7x(7\)ev, 8V6K0t, OVEKCX, Anrog, ovouot,
O7\uu1ro, Bohr]. Also in the fourth century (rarely earlier) the contraction-
diphthong so came to be written av, the two diphthongs having fallen
together in pronunciation. Hence the received text offers 611613, 1<oO\e13vTo,
.,@.
i,
412 .
Is
:5 .='= '-
/
./5
/""
30 I.1. The Pre-Alexandrian Transmission
wi
given by papyri correctly preserve old zero-grade forms where the main
tradition has éP\671e1TrTo etc.8” There was also uncertainty over how to spell
the old poetic adverbs in —ei.
The other major Hellenistic change of vocalic pronunciation, on >- [e],
had little effect on the text in general, but at B 749 the correct Aivlfivecg
has been replaced in the medieval tradition by ’Ev1five<;, which the metre
seemed to call for.
it
65
J‘
Atticisms
3‘-
-I=:-3%
cisms.9° It is doubtful whether any of them has to be attributed to the ori-
If:§;§:f
ginal text of the Iliad.” As with the other modernizations, some of them
='£-§‘é
i -a-I‘. I?":'I"i
iztreiooiv for erreit’ éiv. Some are attested unanimously but only in certain
3.:::§';-.,.:..
passages, like ufiv for uév or uoiv. Some are dominant without having com-
=;:§:
pletely ousted evidence of the Ionic form, as with 7I'07\8l<; (accusative) for
"Ia-~>-»f'I’i".‘j"I'i
5%
'==I=;\r- -Tc, )(i7\10l for Xei?uoi.9" Some, like <I>6[)06Cl)0\/8161 for Hepo-, eipvto for Epvw
=1-,=,»»,,
8%:
or éépvw, Téoocxpeg for Té008[)6Q,”’ are established as ancient variants but
fziifii
not dominant, or not until the Middle Ages. Others again are found only
1-gig;
sporadically in the tradition, like Téuvuo for Tciuvw.
-5.2"‘:
This situation in all probability reflects the major role of Athens in the
i
transmission of the Iliad and Odyssey from the time of Hipparchus to that
of Alexander. The number of copies produced at Athens during those two
55-‘ii: hundred years was, it may be surmised, substantially greater than the num-
ber produced in any other part of Greece. The texts current in the Hellen-
istic and Roman periods were therefore statistically likely to have Attic
saa>§%‘;
“F
ancestors, and divergent variants tended progressively to be eliminated as
.=.=_;;%-I
;;.;\%;. Caner (1921—3), 121-6; Chanttaine (1942), 15f.; Janko (1992), 35.
9‘ The most obstinate is perhaps Boreas’ name at I 5 and ‘P 195. It is transmitted in most
§?§;gez
zEI~?:fi='.- :_ sources as Bop:-Eng, Bopém, but the metre demands the Attic Boppékg or an equivalent. Cf.
below, pp. 205f.
»~> . 92 West (1998b), xxxiv, xxxvi. Another case may be oiiooto, cf. ibid. xxx with the addendum,
(2000), v; below, p. 263.
I'£ai§': I
'~
32 I.1. The Pre-Alexandrian Transmission
However, there must have been non-Attic lines of tradition in the fifth
and fourth centuries, above all, one would expect, in Ionia. We have in fact
some evidence of it.
The pattern of aspiration and psilosis in the received text led Wacker—
nagel to the conclusion that the epic language was at one time consistent-
ly psilotic and that at a later stage aspiration was selectively restored on
words which were familiar in Attic in aspirated form.” If the aspiration is
of secondary, Attic origin, it should have been absent in any separate (east)
Ionian line of tradition. Now, Stesimbrotus of Thasos proposed to solve a
problem that he found in O 189 by interpreting 1T0tVT(X as rrdvt’ ii, or ra-
ther Troivt’ d. His text, therefore, was psilotic.”5 So was the exemplar of the
Ephesian Zenodotus. At A 589 it had At0tvT0<; for A’i'cxv6’ fig, at M 75
averwg for éivefi’ dig, at E 249 (JO\7\0Tt-3T]1 probably for 6120106’ éfil, and at P
368 Toooov probably for 0’ 5oo"ov. This is a small body of evidence, but
it provides some basis for the generalization that Ionian texts were psilo-
tic. Everything seems to point towards this having been true of the origi-
nal text too.
Zenodotus’ text also diverged from the common (Attic) tradition in its
admission of various late Ionic forms (see below, p. 43). These were not a
personal fad of his own, nor inherited from the primal text. They were fea-
tures that had developed in Ionian manuscript tradition, just as Atticisms
did in the Attic tradition. No doubt analogous secondary features devel-
oped in other regional traditions too, only they did not affect the main tra-
dition and have left no record.
9‘ Wackernagel, Beri. Pia. Wot/a. 11 (1891), 8 = (1979), 1571; (1916), 40—52; cf. Chantraine
(1942), 134-s.
95 Stesimbrotus FGrHist 107 F 24. I owe this observation to A. C. Cassio, who notes that
the Thasian dialect was psilotic.
?
k.
5
2. Zenodotus’ Text
‘K
Homeric textual criticism. Eduard Schwartz complained that Zenodotus’
3‘
‘<€
title to this effect was unjustly overlooked.‘ Giorgio Pasquali declared that
€~
€
‘la filologia metodica, scientifica comincia in certo senso solo con Zeno-
doto’.2 Rudolf Pfeiffer wrote that Zenodotus ‘initiated Homeric studies on
=:»- a grand scale and in a methodical way both as editor and lexicographer’,3
._;»;
an
2%
while according to Klaus Nickau his achievement was ‘dafi er die Textkri-
figs
Isl;
I8: 1.
tik in Alexandreia ins Leben gerufen hat’."'
“i
i Since 1788, when Villoison announced his epoch-making discovery of
the Iliad manuscript Venetus A with its rich scholia, we have been in pos~
£3»?
Q.
session of a considerable quantity of information about Zenodotus’ text.
We know his readings in over four hundred places. Mostly they are quo-
ted because Aristarchus disagreed with them, which means that eccentric
.5,_
<->~1=
==§5:
5§EE
252-‘:-
525?: or obviously false readings in Zenodotus’ text are being cited in dispro-
1-.3
portionate numbers. Nevertheless, it is clear that this was an eccentric text
:':>-=3
?»~"E
iii; in many respects. Like the wild papyri of the third and early second cen~
turies BC, it had some inorganic additional lines with respect to our vul-
gate and some differences of wording (substitution of formulae, etc.). But
unlike the papyri, it more often had an abbreviated text: in over fifty pla-
ces it omitted lines or passages which are present in the rest of the tradi-
tion known to us, sometimes with reformulation of the surrounding lines.
Most of the omissions are of just one or two lines, or not more than five,
but there are five or six omissions of over ten lines and one of over twen~
.9,
ty.5 In four places he had lines in a different order; the most notable case
21$.-=
was at the beginning of ('5), where we are told that in his text the appear-
ance of dawn came not before but after the scene on Olympus, that is,
some fifty lines later than it does in our vulgate. In half a dozen places
-£2 3
,2-<¢ - - -
-
i Schwartz (1908), 4, ‘at Zenodotum principem nostrae artis fuisse inique nunc tacent’.
2 Pasquali (1952), 223.
3 Pfeiffer (1968), 92.
4 Nickau (1977), 88.
» - ,- 1
‘:l'§,-‘= -
5 I should emphasize that I am not speaking of Zenodotus’ atheteses, the passages which he
marked with the obelos as suspect, but of lines that were actually absent from his text.
_ _
'\’3§Q.
. .
/{"511-:jl::..
/‘S -=
/\4
~< |
34 1.2. Zenodotus’ Text
:12
;_;§
:§‘
ii
-r
Quum tot Homericorum carminum editiones tantis sumptibus ubique gentium com-
paratas Ptolemaei collegissent, quis credat, Zenodotum, Aristophanem, Aristarchum,
his editionibus in emendandis divini poetae operibus non usos, vulgares editiones his
vetustate et summa accuratione insignibusmagriisque opibus emptis praetulisse?’
Wilamowitz, likewise, speaks of Zenodotus’ ‘Ausgabe’ as being underta-
ii
i ken ‘mit den Mitteln der Bibliothek von Alexandreia’ and representing for
i that reason a decisive step forward. Zenodotus inspected the manuscripts
2
‘iii available and chose from among them what he judged to be the best wit-
;§-i nesses.‘° In Pfeiffer’s account this is only slightly modified:
125:2; We can appreciate Zenodotus’ problem when we realize that he was confronted with
E
such a great number of more or less differing copies It is not improbable that
ii
Zenodotus, examining manuscripts in the library, selected one text of Homer, which
i
E seemed to him to be superior to any other one, as his main guide; its deficiencies he
may have corrected from better readings in other manuscripts as well as by his own
2-ii
ii
conjectures.“
Nickau too makes the riches of the Alexandrian library a decisive factor.
§;=§'i§
2 2% In his view, Zenodotus, as the first librarian, discovered that the various‘
costly exemplars of major authors had many divergent readings, and this
realization meant that the question of what was the best text had to be put
in new terms. The Sophists and Peripatetics had practised literary criticism;
‘i -1 1:
now the Greeks became aware of the need for textual criticism, a textual
criticism based on study of the Ulaerlieferzmg. Thus Zenodotus took the
step that led to the development of philology as an independent disci-
Iii
§§:,iI
pline.'2
-.-5.1;:
E‘.-j=j;§
int;
I.Ԥ_- E: Pasquali gives us a slightly different picture, in that he allows Zenodo-
5 5% I- tus to put his nose out of the library, but he has essentially the same con-
ception of the nature of his work. According to him, Zenodotus wanted
:';}I :5;
553:1: E5
to prepare an edition worthy of Homer, but he realized that the interpo-
Iiiii lated texts circulating in Egypt were worthless, so he based his work prob-
i~%'I.‘?%
'i'=‘='§ E!
='I':':';‘:'
I'=';'§": ably on Attic texts, perhaps also on some from outlying parts of the Greek
world.” Some more recent scholars, while still seeing Zenodotus, in his
capacity as Ptolemy’s first librarian, as faced with the task of bringing
2'52-I3 E.-I
E"-'.< =5
===r 2--. order into the chaos of the Homer manuscripts, have emphasized ‘that we
I='?3'é:
=5.-‘
===;i=?; have no idea how many manuscripts Zenodotus consulted, how he eval-
=,¥_i-=.;z. r;,.
5_E j
Iiiiif
5'3ii
5;?_=j§§:
E~‘.-.§FI
ii _________________,.,_.___,.,..,.......i....-
-="_;=_"a:z;:_a;I!=§' .
<=::F-
9 Diintzer (1848), 43.
:1
‘O Wilamowitz (1916), 6f.
>33
é “ Pfeiffer (1968), 110.
>1:-5%-_ .. :. -
if
‘I
2?.’-=3 E:
‘2 Nickau (1972), 44; (1977), 259.
{E
=.4§'55-"51:
EF;
Z
E
'1-'
ifQ‘’-
3'‘II
53 'I‘J-I:'{
:1'.--‘\='5.R:-I'§-.,!'I’-:'(\<''I 51
-4,;_--/xi4.1§>.-_<:-=s.,->vi=-
£3‘.-Q .~\, . <: I.\'>$ <2-¢’><w
36 I.2. Zenodotus’ Text
uated them, or whether he was consistent in his use of them’.1“ Yet the
basic assumption that he did consult different manuscripts remains as
strong as ever. Let me finally quote Franco Montanari:
I believe that selection among textual variants discovered through collation of differ-
ent copies and recourse to conjecture to emend a univocal text are jointly present in
the Alexandrian otcipfioaoig, starting from Zenodotus.”
It is time to challenge this assumption, inherited from Wolf, that colla-
tion of different copies was a normal and essential part of what Aristar-
chus and his predecessors did. Clearly Aristophanes and Aristarchus were
not dependent on Zenodotus’ text but followed another source or sources
much more similar to the vulgate. But there is no evidence that they act-
ively sought out a plurality of different manuscripts for comparative pur-
poses. Arthur Ludwich begins his bulky work on Aristarchus’ Homeric
criticism with a section headed ‘Aristarchs handschriftlicher Apparat’. It
opens with the assertion,
Als Aristarch die Homerischen Gedichte herauszugeben unternahm, schuf er sich
zunéichst durch Vergleichung einer Anzahl iilterer und jiingerer Manuscripte die
nothwendige Grundlage.“
The same section-heading, ‘The apparatus criticus of Aristarchus’, appears
in the important (and often overlooked) Appendix to D. B. Monro’s editi-
on of Odyssey 13-24, with the same mistaken assumption that Didymus’
citations of manuscript sources in his work Trepi Tfig ’Ap1oT0tp)(eiou
81op6u5oew<;, as excerpted in the scholia, were taken over wholesale from
Aristarchus.” We shall say more of this in the next chapter. In fact the first
scholars known to have cited manuscript authority for variant readings are
Aristarchus’ contemporaries Gallistratus and Crates. Didymus is the first
known to have compiled anything in the nature of a ‘critical apparatus’. It
is entirely unjustified to project his methods back onto Aristarchus or Zen-
odotus, or to assume that all the various copies available to Didymus in the
time of Augustus were already part of the library’s holdings in the early
third century. In the first decades of the library’s existence the priority was
surely to acquire the works of as many different authors as possible, not to
accumulate multiple copies of standard classics such as Homer.”
i
. _,
._ 2,
Consider what we know of Aristarchus’ methods, for which we have
plenty of material in the scholia. Of course he had the text of his teacher
I,
t
Aristophanes before him. He also kept an eye on that of Zenodotus, and
2
._-5.
._.,
took up critical positions against it. But the arguments he used, as report-
Eli;
iii ed by Aristonicus and Didymus, were always based on the internal evi-
dence of contextual coherence or general Homeric usage. Not once does
he appeal to the authority of manuscripts.” It is Didymus, not Aristar-
iii
i chus, who makes all the references to oti 7T7\81OUQ, (sc. éxodoelcg), (xi Jtéioou,
ui pgocpiéorepou or on doteiérepou, and their antitheses, I’| icoivfi or on
='='§‘.;
roivai or icoivérepou or 0I’|l.l(13081(; or e’u<0u6tep0u or <1>otv7\6'repou; it is
25 15
Didymus who cites the texts of Antimachus, Rhianus, Sosigenes,
1531:
i
Philemon, and I) 7ro7u$o1'1)(o<;, and the six ‘cityeditions’, even though his
1
S
knowledge of some of these sources was indirect. Aristarchus’ approach to
textual criticism was based on a detailed knowledge of Homeric language
.5;
.:»,<.
and style. I-Ie appears to have been simply unaware of the possible benefits
of extensive manuscript collation; or perhaps he realized only too well that
.i
it was likely to yield little but a bewildering variety of aberrations.”
-.1311'
. ,.
§'.=-‘J5:
Aristophanes’ text seems quite often to have agreed with Zenodotus’,21
§=:-"i i
.-1-'-iv:
II-Iii:
and this must reflect his use of it besides his main source. Again it is a mat-
.‘<.
1.
:':'.‘.:<‘EI
ter of one scholar paying attention to another’s text, not of his looking
around for a wider range of manuscript sources.
:-:-1). As for Zenodotus himself, the text associated with his name is so eccen-
rug;
tric that it is impossible to regard it as the product of any rational process
:iEl»'.
%‘;1;¥ti-
1.1; of selection of readings from alternative sources - or indeed of any ratio-
nal thought at all. I have already mentioned its several offences against
-..<:.
E-j metre, virtually incomprehensible in a critical text produced by a man who
25.93:
;::*r.-
,1i?i'§l§Ii
"’ Pointed out by A. Roemer, Ab/0. cl. Bay. Ale. d. Wiss. 17 (1885), 678. At I 222 Didymus
::-'.j ;:.'
21$:
,;,.F.. reports that Aristarchus had remarked that éitp or otilp’ itxdioavto would have been preferable
to ti iipov Evto. But, says Didymus, {mo nepirtfig €-07\0£[3810£Q oufiiav petéflnxev. év rro7»7\o€i<;
oiittog eiaprbv qmpouévnv rfiv ypocqafiv (sc. £25 i-ipov 'évto). This does not mean that Aristarchus
§z§=>,_I
i,2
E
-.= 5.1 5;:
2.2.2.21 .2
said ‘I find this reading in many manuscripts, and therefore do not venture to alter it’; it is Didy-
mus’ way of saying that Aristarchus refrained from putting the reading he would have preferred
-:-<;-.
:..<,.
=I=I-'3
:::1;I
I11 PI
. .E".2 into the text because he had no manuscript authority for it. Didymus knew that Aristarchus
E5552 E-R"
consulted more than one text, because he cited different scholars’ readings on different occa-
. ;.
F-' sions, but it is just his own assumption that Aristarchus systematically checked ‘many’ copies
before discussing any reading.
2° Cf. Schwartz (1908), 4, on the Alexandrians’ use of their predecessors’ exodueig: ‘neque
;‘53;§:-i i
;:.-.;_
in librariorum uitiis eis erat desudandum, sed inter lectiones doctorum hominum qui ante eos
Homerum ediderant, eae seligendae quae non inuentae sed traditae uiderentur.’
;=:-3 :1
z:=-<£= 2’ Cf. T 126, A 137, Z 121, 9 290, I 158, K 306, A 94, M 66, 79, 127, 138, N 242, 246, 551,
5-5-5-I 22
E 36, 177, 208, 259, 276, 299, 310, 340, 412, 505, O 139, 301, I1 223, Z 198, 400, 466, 502.
;::.<,.
:.;._., 5:
::=.-:5
sf
I3EE'<"<-.
:{-Si -.
2:221 '.
-;::g -.
=11‘! :I
serf;
E1
;.\::zt-:
-:\- F.-
1-,:'=': >1.
. f.
L
38 I.2. Zenodotus’ Text
.%-i
22 The only evidence is the Sada article, where he is called érrorroiog K01’: ypotppomxdg (II
506. 14 Adler), and the existence of a couple of epigrams ascribed to a Zenodotus, who may or
may not have been the Ephesian (cf. Gow-Page, HeZlem'st:'c Epigmm-5, II 557).
23 Diintzer (1848), 21f.
24 Cobet (1876), 251f.
25 H 127 uezpéusvocg = ii’ sip-, A 589 Atowtoc; == -own Sq, M 75 uysrux; = éiyefi’ dag. E 95 oe
= oe’ (= P 173). 249 o<7\7\o'ren1 = dikkofi‘ i-:r“|i?. Tl 202 pnnototofie = p’ rim-, P 368 toooov : 8'
6o0ov?, ‘P 886 pqpoveg = 6’ ijpoveg? At y 444 Zenodotus had Bocuviov for 6’ éipviov and listed
it under 6 in his Uidboocxl. At <1) 95, where he must have intended ouxi oyoiotpiog, Aristarchus
read his text as 013K iovoiotpiog.
ijiji
--<.
1 ‘E:
"i
vi;
iii
Ejii
I.2. Zenodotus’ Text 39
Ii.53;i:
Should we then suppose that Zenodotus wrote out his text from a single,
IM’.
:8;
....
:'-9'.
-§ defective older source, introducing his own conjectures where he thought
-‘ii:
;_.~<.
fit and marking suspect lines with the obelos? But the question must first
be asked, did his text actually contain conjectures of his own, apart from
r his atheteses? For if it did not, there is no reason to think that he wrote
1'2
‘-3-:
.|-,
out a new copy at all: what was later known as Zenodotus’ otdpfitooig
might simply have been an old exemplar which he had owned and marked
up with his obeloi. Even if it did contain other emendations by him, they
might have been written in the margins; there is still no need to postulate
‘-iii an autograph text?" If we assume that he took an old existing copy and
E33
used it to register his atheteses, without subjecting its text to any close cri-
{Ii
tical study, it becomes very much easier to understand how it could have
.- I-
2?;
had the erratic character that it evidently did have.
K
Scholars have been much too ready to diagnose the peculiar readings of
.'l\
.5 '
4» Zenodotus’ text as wilful alterations by him, even though they recognize that
.:,._
:<§'
Egg
.J_
many of these readings must have been taken over from earlier exemplars. The
:.:.
-.25:
fit:
,.~._. contradictory approach goes back to Wolf. In the Prolegomena he writes:
Many of his readings are, indeed, so improbable, and result from such rashness of
13.1,. .
1
judgment, that even a beginner today would be ashamed to make such emendations
‘Jig;
=.»;.
in general he treats the Iliad as if it were his own composition.”
ii?
But in the footnote (71) where he lists a selection of Zenodotus’ aberra-
Iii;
zii ié tions, he says, ‘I call these readings, not corrections. Who could believe
that all these were first introduced into the text by Zenodotus?’ Again in
.,. the following note the two interpretations are in conflict:
5.1;:-I
it;
Zenodotus’ most ridiculous attempts, however, involve transposition and the supple-
_: 12.
_.é§_:
,. ,.
ments he invents for lacunae which he created at will But in many of these
i
instances, particularly in those where he completely omitted something, anyone might
1-Bi:
£9“-
E suspect that part of the blame is to be placed on the poor quality of the ancient
"*9;
{.I;i£I manuscripts, not on Zenodotus’ invention alone.
--2:.
}':'fi:_
We find the same ambivalence in the monographs of Diintzer and Nick-
au: on the one hand the recognition that many of Zenodotus’ readings
;-:1F
must have been inherited, on the other hand the conviction that he made
._it
Ii ? many changes suo Marie, while no clear criterion is offered by which to
1<
‘<3
=.5.1
distinguish between the two cases.
..,..t_.
55,1.-
24?!
_
E5?
-5 :=
26 Cf. Montanari (1998), 4-9.
Iii-1
;=;;..
~;.;_¢,; .
2’ Wolf (1795/ 1985), 174. Many later scholars have been content to take this dismissive atti-
?:¢;.
tude to Zenodotus; e.g. P. M. Fraser, Ptolemaic Alexandria (Oxford 1972), 450f., ‘His emenda-
tions at times reveal considerable childishness and irresponsibility his treatment of the text
seems to have been quite arbitrary at times: he omitted consecutive series of lines which he
regarded as superfluous’, etc.
5}. . ..
ii; 2:
\-.35125.. . . .
-.-..s=
1.-4... -
:_\;:._.-_.
55
= 2 II ti = -1 1
3 ;’:
The majority of his variant readings are of a sort that We should not
expect to find in a scholar’s text but rather in a rhapsode’s. In dozens of
places they consist in the substitution of one formulaic expresion for ano-
ther; for example, in A 73, where the vulgate has ii ocbtv opovéwv
oivopriootto 1<ot‘1 petéerrrev, Zenodotus had ESQ 111V ciueifiéuevog Em-:01
Trtspoevtot Trpooq 15601.28 Today We understand that this kind of variation is
typical of the tradition; Diintzer, however, felt obliged to invent a reason
Why Zenodotus should have been dissatisfied with the received text and
rewritten it.” In many more cases the effect oi his readings is to replace
Words or forms with more banal equivalents, the kind of thing that was
likely to happen in any casual line of tradition. Thus at A 203 he had the
active ionic; in place of the poetic middle ’i5r|1, at B 1 and K 1 11670101 for
éi7\7\o1, at E 132 otitoioou for ofrrdapev, at H 153 Boipoet éptfii for Boipoei (.31,
at E 274 utiptvpsg for 110ip"rvpo1.3° In several instances some grammatical
irregularity or harshness is smoothed away,“ or asyndeton is replaced
by connection.” There is no justification for taking any of these as a deli-
berate change by Zenodotus.
The plus-lines reported from his edition or inferred to have stood in it”
are not interpolations from his hand, as supposed by Diintzer and still by
van der Valk,“ but something typical of the texts current in his time, as
We know from the Ptolemaic papyri and from early quotations. Of the
_,,,,._i_....__...------i------
28 Cf. also A 91, 204/212; B 55, 119, 681, 718, 727; F 100, 244, 364/O 192;A139, Z82. 339;
Z 71; 0 166, 501; I 36, 88, 660; K 291, 306; A 86/730 (fisinvov and §0p’J1'OV interchanged), 480,
492; M 230, 444; N 148, 229, 246, 609, 610; '5 40, 177. 310; O st, 134, 307; 11 92, 223. 710; P
214, 456;Z154. 400; T 76; X 378; ‘P 533.
2° Diintzer (1848), 145, ‘quia Calchas Achillem alloquatur, non totam concionem, quod illo
versu significetur’.
3° Cf. also A 24 (case), 34 citxéwv, 129 Tpomv, 351 Tdvomtoig, B 435 tafitot, 532 Bnoctv for
Bfiooow, F 155 tlixot for |i|1<ot,A 3 évwtvoxoel, 478 Bpcrrtoi, E 128 fifi’ éivfiptorrov, Z 54 oivtiov,
H 32 \'>p”w dzfiowoiroiot, 127 péyoz 6’ Eotevsv, 428 (case), (-9 349 Fopvévog, 378 rrpoooweioag, 448
Ktxpétnv (cf. K 545, A 782). 562 pupi’ dip’, I 158 r<oq.|q>9fitto (gl.), 506 cpfiavéct, K 10 c|>o[3éovt0
for tpop- (cf. Z 247, T 14), A 528 1<e'1o’ for K5161 (cf. M 348, 359, contra 368, \P 461), 841 dtpa—-
Miow (gloss), M 11 Em-;. 231 Houmifiozpotv (E 470), 340 éaruitxero, 423 07\1Yl'|l évi Xuipnl, N
172 Br; voie. 237 Bin for dpetfi, 702 )(0'iCet’ rim’ fofiiotozr’ din’, E 276 omission of epic t’, O 139
T086:-3, 179 trohapiflsw, 207 ofiolpot éirmt, 277 teitog, 470 rrptétnv, H 10 zwpoofiépxetoti, P 392
|<t54<7\w1, 2 198. 0u3tb<;, 499 dnroxtocpévou, Y 273 Saétepov, (D 335 (tense), W 759 €1<8op’.
31 A 251 (concord), 404 (case), B 36 (singular verb after neuter plural), I" 211 (case), E 898
évéptottoc (cf. O 225), Z 511 prucb’ E01, A 838 Em, N 315 éooépsvov 1‘TO7\E}.l1C81V, E 208 (case),
H 281 (concord).
32 B 187, Z 155.
33 A 69419? F 3380, E 808, G) 52a—a'? K 349? N 8084, E 1364, P 456.4, Z 156a.
34 Diintzer (1848), 157-62; van der Valk (1963—4), II 36, 40f.
1‘
it
x
1.2. Zenodotus’ Text 41
i
22.}.
omissions, one or two must have been merely accidental, like that in A
.552
E553;
._-=4
446-7,
(Tag eimbv <i;v Xepoii 110,21, E5 6%: Z‘>é<E0tro Xotipwv
1;;
551??
I152
.=-=
...,,
1:
1'
rroiiéot q>i7tnv'> T01 6’ Gard 08031 iepfiv i~;1<orr6u{3nv,
><
?
or that at 491, where there were two successive lines beginning oiiré Trot’
<1
and one fell out. Some of them may represent the true text, where inter»
.-.'-:;?,
-_=.f.-.|5
Z‘-‘£18
553:5?
polations have occupied the vulgate.” The majority, however, must be
1555?;
-1:;
E out twice in full, is reduced from eleven lines to two. Some of these abridg-
=E:;ii
'=§-E-i i ments involve some re—writing.3" If they were made by Zenodotus to suit
his own taste, then he was indeed an arbitrary and frivolous operator,
_.:.:.:E
:5:-‘,5
555;‘-€
:':';;4 treating the Iliad, as Wolf said, as if it were his composition. But if he was
I_E§.f.é
,;._=_,-.;
so free in eliminating verses which he thought the poem would be better
without, it becomes difficult to understand the purpose of his atheteses.
Why did he not omit those passages too? Abridgment is not a scholarly
but a poetic operation. Why not, then, accept that it was done by a pre~
Zenodotean rhapsode or rhapsodes?
The same point may be made about other aspects of Zenodotus’ text
=' :-.='-;.=-;. =. '=;- =3
where divergences from the vulgate seem to be not casual but show some
element of principle. A number of his readings have the effect of eliminat-
ing hiatus or mending a oTi)(o<; 7\otvotp6<;.37 We should not conclude that he
,
had a personal aversion to hiatus and devised these variants, or preferred
them to others, in order to avoid it in a few places. The probability is
rather that post—classical rhapsodes, in line with the trends of later
Q-1%:-.f,='.f: ‘,§_:_:_
'5.-_;J.-=:_,E=-,I_E;=,'._-§; _ ;_;
5,
versification, occasionally made easy adjustments in the text to achieve a
better synapheia.
::;I.=.;E
55;;-.=,;_=
:=;§i:‘;.
:5=.. ;_.
:=;== g: -
; =;=.=;1r-;
=:;;.3-
:'='=':?j?:
;';';';:',§'.
:'ai 'f-11;:-I
35 In this category I would put B 674, 6) 528, 535-7 (or 532-4 ?), 557--8, I 416, 694, O 64-71,
1
266-8, 610-14, (1) 195(7).
31’ Here is the full list of passages: A 219-20, 446-7, 491, B 60—70, 111—-18, 156-68, 226—8,
674, 726?, F 334-5, 423-6, A 88-9, E. 53-4? H 255-7? 482, 6) 37, 284, 371-2, 385-7, 528, 535-7
.near,-:
__;_, ._j?_. '
(532-4?), 557-8, 1 14-16, 23-31, 416, 694, K 240, 253, 497, 534, /\ 13-14, 78-83, 179-80, 356,
.»,..,,,
»
~t»'.~:;.. » .
515, 548-57? 705, M 175-81, 450, E 376-7? O 18-31, 33, 64-77, 266-8, 377-80? 610-14, II
89-91, 93-6, 141-4, 237, 677, P 134-6, 404-25, Z 174-7, 528-9, T 77, (D 195, Q 269.
§i='?'§R if
37 B 571, Z 112, 6) 503, M 138, Z 287, 385 = 424, Y 230, 0 70, 379, 0 251, 1| 209, 297.
_=;i;,-4;. ._
=z:=1-I-E :2
'$(-'<:'-
-:;3:;_=§,;. -_ . -.
'.'.;:j§
,¢;-:e, 4eaz4;>:4->'
42 L2. Zenodotus’ Text
3“ A 567 iévre, B 297 pipvovre, F 459 oirrotiverov, A 348 pévovte, M 412 ecbopaprsitov?
(after (5) 191, ‘P 414), N 627 0’1')(eo0ov, O 347 émoosrieofiov, W 753 rrelpfioeofiov (after 707), or
38 rrépqxotvre (with hiatusi), 0 251 arotioottov.
3° Hymn. Ap. 456 (see Allen—-Sikes ad loc.), 487, 501; 6. 12?; interpolations in E [487], G) [74];
Hom. epigr. 4. 9; orac. ap. Hdt. 7. 140. 3; Emped. B 137. 6; Arat. 968, 1023; [Theoc.] 25. 72
(dub.), 137; Ap. Rhod. 1. 384 (v.l.), 3. 206 (with Campbell ad loc.); Dion. Bass. 19 v 3 and 22
Livrea; [Opp.] Cyn. 1. 72, 144--6, 2. 165, 260, 494, 3. 464, 4. 358; Q.S. 4. 506; Orph. Arg. 1091;
Diintzer (1848), 46. '
4° Schwartz (1908), 4f. Cf. Nickau (1972), 32, ‘Konjekturen bei Z. zu finden (wenn wir ein-
mal jede unoriginale, aber sinnvolle Lesart so bezeichnen), ist nicht schwer, wohl aber, zu zei-
gen, daii sie won Z. sincl. Der Nachweis, daii bestimmte Konjekturen erst zu Z.s Zeit méjglich
waren liiiit sich kaum je fiihren’; id., (1977), 45--8.
x
r
2
ii
Q
<2
8
I.2. Zenodotus’ Text 43
is
Z 155 (al.?) ’E?\?tspo<1>6vrnv" §7t7\epa voip (1)0101 KCXT01 f>tci7\e1<rov T01 1<0u<0i
.2:
155% (Eust. 289. 38, 632. 7, 635. 6). Zenodotus’ text was not accompanied by
555$.
3-E-3 1:3;
explanations, and that statement about i§7\7tepot perhaps derives from his
.;_.<.
.:.%
=-.;t.
=2}
153;;
I"7\t6oo0u, where he may have used it as the basis for an etymology of Bel-
=.‘=‘=;‘i§
Ilgfii i
-.1 lerophon’s name, Be7t7\epoq>6vtng oiov ékkepoqnévrng, §7\7\ep0t vétp tcit 1<0u<E5r
55.5%
E-5-it? KOLTOt <310i7\e1<Tov. If the ’E7\7\epo— form appeared in his text, it may have
I:::l~$3
':;.§
:-:8 been only in the margin. It is hard to believe that it had any existence in
older manuscript tradition.
$1,?
5258
553$
T 224 irrrrwr 6’ eiooiuevog éuiyn (|)l7t6Tl']T1 K011 sin/fit for xapehéfiaro
1<UC(VO)(Ot1TT]1. This variant is not explicitly ascribed to Zenodotus, but it is
252%
55.,8
54%
pre-Aristarchean, and if it was inspired by the defect of the vulgate read-
5':'='fi
ing noted by Aristarchus (irrrroc vétp 015 Tf0l[)0tK01|.101T(X1), it would be in
;::;;_,g keeping with the zoological awareness attributed to Zenodotus in a few
:_5;I;:§
other passages.“ On the other hand, the variant may be nothing but a
banal formulaic substitution.
This is very little. It may be allowed that Zenodotus did make conjectures
on occasion, though they may have been confined to the margins of his
i ?-$5
;_ -.
copy. It is absurd to start with the assumption that every variant reading
recorded from his text is likely to be a conjecture by him. Overall and
...,¢
rhapsode’s copy.
;:5:3--.
55;-1%;
-5-::-:%i
. .
:E;E;g>;;. _ ._
in which he marked his atheteses: that was his 516p6t0o1c;.
==a:;$:
as-. " -. -
One of the ways in which this text diverged from the main tradition
was in its admission of various neo-Ionic or hyper-Ionic forms, such
as éuwvrév and éwurfiv (A 271, E 162); Kpnrcic for 1<p0rr6c; (A 530);
Trerroréorrou and érriotécxrcxr (singular: Z 58?, H 243); Emit, Euvénfie,
Kotrerjfioruev (H 270?, N 166, 257); [3ouvfi‘i'e for Bouvoiie (N 824); 06V0[)61
for fievfipéwl (F 152). These reflect, as I argued in the preceding chapter,
not the arbitrary doctrines of a grammarian, but transmission in a late
- -- I '- :
5:543:-. '
iséii
.'
Ionian milieu. This was also the opinion of Wackernagel, who wrote that
iiizzg
._, ‘derartiges beruht nicht so sehr auf gelehrter Theorie, sondern auf Benut-
:_
.- 1, _ _ .- . .
-_ -. -. -.
1
;>f<'_vv_-. . '- . . 3 .
_-;,;.;@. -. ' -_ . _ .
4' N 198; P 134-6; Anacreon PMG 408. The criticism of Anacreon’s antlered doe seems to
E:
go back to sophists before Zenodotus; see Arist. Poet. 1460b31. Cf. Pfeiffer (1968), 1171. Ari-
3 "-1 -. -- -
:;=.5Q<- '--- -. .
-"tr-<1‘-'. - -_ '. . ' .
583$:
r-.=:X~'
stophanes (despite being the author of a work T1601 Ctéttov) vigorously attacked Zenodotus’
.,_,:,,‘_:
.. .1“! . _. ._ ._ (unmetrical) emendation (Ael. H./1. 7. 39; Ar. Byz. fr. 378 Slater).
-,-.,<»<_ . . . _ _
-\'§;:. ><
Eififv E‘. 3
\/
1-.6» »-
5; :‘/
=4-22 :.'; I-_-I'j':;
;;2:=; 2 .. .-
£»‘$Q,,=_:_I I
‘I:'-.':i.:'; -2 I-:1‘-5:‘-'
I :
.: 2-:1-2.: 3.'-
44 1.2. Zenodotus’ Text
zung von Handschriften, die etwa aus ionischen Stizidten kamen’.42 The evi-
dence that Zenodotus’ text was psilotic supports the view that it derives
from an Ionian tradition separate from the Attic tradition which was the
basis of the Alexandrian vulgate in general. We can modify the description
‘rhapsode’s copy’ to ‘Ionian rhapsode’s copy’.
And where should Zenodotus have obtained this copy but in Ephesus,
his home town? Is it not obvious what happened? The studious young
man acquired his own copy of Homer as soon as he could afford one, and,
having acquired it, he continued to use it as his working text throughout
his career, just as many a modern scholar continues to use the text he
bought himself as a student.
If this account is correct, Zenodotus will have come into possession of
his text sometime before 300. But it may have been written twenty or thir-
ty years earlier, if not more. This would take us back well into the fourth
century, a time when the practice of competitive rhapsodic performances
at festivals was still very much alive in Ionian cities. At the Eretrian festi-
val of Artemis from about 340 BC, rhapsodes competed for a first prize of
120 drachmas, a second prize of 30, and a third prize of 20.43 Zenodotus’
city, Ephesus, had been the home of the famous rhapsode Ion, known to
us from Plato’s dialogue, and of a Panionian festival which Thucydides
compares with the old Delian one described in the Hymn to Apoilo.“ In
the later fifth and early fourth centuries, then, it had been a major centre
of Homeric tradition. Ephesian texts of that time no doubt figured among
the ancestors of Zenodotus’ exemplar.
Of course this does not mean that Zenodotus’ text takes us straight back
to fifth-century Ionia. Its many errors included a number of modernisms,
forms which on linguistic grounds could not be so old. But no matter how
many corruptions his text contained: if it came from a iine of tradition sub-
stantially independent of the Attic paradosis that was the principal source
of the later vulgate, there must be a fair chance that it preserved some
good, old readings that were otherwise lost. And in fact, if we consider
Zenodotus’ readings with an open mind, not determined (as some scholars
are) to dismiss them wholesale — a curious attitude to take towards what
m_
"2 Wackernagel (1916), 73; cf. ibid. 61, ‘zweitens ist einiges geradezu neuionisch und unat-
tisch, scheint also einen ionischen Nebenstrom der Uberlieferting zu erweisen’; janko (1992),
24, ‘perhaps he [Zenodotus] had a patriot’s preference for Ionian MSS’; Wilamowitz (1884), 257,
‘Zenodots Homer war ja wirklich sehr viel ionischer als der des Aristophanes, weil Zenodot ein
Ionier war.’
43 [G 12(9). 189. 15--20.
*4 Thuc. 3. 104. 3. For the importance of the Ephesia in later times see M. P. Nilsson, Grie-
cbische Feste won religzifiser Bedentzmg (Leipzig 1906), 243-7.
~§
4‘
I.2. Zenodotus’ Text 45
>4
{um
ti?
is in general our oldest source for the text - we find a not negligibie quan-
tity of excellent variants. Some are shared with part or all of the medieval
tradition against Aristarchus. Others are unique to Zenodotus, or virtual-
ly so. I have already listed a number of omissions which seem to represent
not casual abridgment by a rhapsode but the true original text. There are
A
$‘
3
1
‘l
also choice readings such as the following:
1
515%;
4‘
n" A 5 00(iTOt. Although I do not accept this as the true reading, there is lit-
:=;:z;
;=;:4;
2:215:
Iiijili
----4.
tle doubt that it goes back at least to the fifth century. See below, Part II.
B 4 'n'o7ti3g. See above, p. 28.
-8"Z48 '?<*%¢‘>~:'I .i :.';.:'Ii"
::_._,,_.
.2».
its:
B 144 ¢rj (cf. E 499). Zenodotus alone preserves this archaic compara-
.8
way»- tive particle, the main tradition having replaced it by the banal dag. As
$1
i
i .5
_
Antimachus knew it (fr. 121), he presumably read it in his Homer: a fur-
-I§§;..- :
. ther pointer to an Ionian tradition.
4
54%;
,‘§'§'§§j: Z 112 dwépec €ote 600) K011 dtutivsrov dorsal‘ 7\u3[inv. The incorrect dual
must be rejected as a hiatus-saving device (cf. above), but if riruuvete is
3% .
restored (Leaf) we have an individual verse that is preferable to the vulgate
civépeg §o're, ¢>i7\o1, uvfioaofie 5%: 6o15p16o<; dtkxfig, which repeats a banal
formula from elsewhere.“ The exhortation Boo) §ore is paralleled at H 422.
Z 285 ofituo Kev <|)i7\ov iitop. See below, Part II.
A 831 fiefidiotofiou. See below, Part II.
E 37 iiw’ oiiévrec. See below, Part II.
E 427 T’. See below, Part II.
E 485 (év) 118Y01{)O1G1V) oipfic d7\1<rfi pot. This seems to be the correct form
of the phrase; cf. West (1966), 345.
II 156 Tf01VTI]1. See below, Part II.
P 54 dvafiéfipoxev. See below, Part II.
Z 142 oivopeiioou. See below, Part II.
5'2‘.-1?:
Z 576 513 r‘)0<50r7\<‘>v 0OV0tKfi(X. See West (2001).
,,,.$_- . . . .
'¢.-<r-.- (I) 95 013K) 0YO1()'TplOC;. The ofmi is not acceptable, but évoiotptocg must
be an old reading. See West (2001).
..,.>,.
-'<»9.
=12-1-5.3: .\rz~e§4a»:=_s-?.~>\
'<f&§‘=a :
1
/-“‘ -_
-.
1:‘:-':-I-':.":::':-5.:=":.':-1' 1'.‘
3. Didymus and his Sources
,i.._
<
11 1.3. Didymus and his Sources 47
<1
Z.
manuscripts that he has not seen himself and other scholars’ discussions
€
that he did not find by reading their books from beginning to end but by
taking a reference from a previous commentator. Ancient scholars behaved
.§:;,'<_,j
likewise, except that, whereas the modern scholar will (if he is prudent)
I-':‘..’:E
':'5E-K
-::3¢*_:
:"-5-5.4. look up the work he is citing to check what it actually says, our ancient
I-;-I-8
;=.=.¥§
:5:
colleagues were not always in a position to do this, as a work that they
iii?
found cited by a predecessor might not be available. In any case it was
-15551:
.:-;-rs
='.§.§; ;
troublesome to unroll a volume and locate the desired passage, especially
E5592
>< in the absence of numbered sections. In the present chapter I shall review
the sources cited by Didymus, noting such hints as there are of indirect
4
5‘
transmission.
i
:-;-;-a;-
:._./.:
Didymus was one of four grammarians from Alexandria who were all
- _
.:_;:.Y;:._
..,,._.
.
. .
_
active in studying the Homeric text at about the same period. The other
three, who all practised at Rome, were Aristonicus, Seleucus (nicknamed
6 'Ounp1|<6c;), and Philoxenus. The first question is how Didymus’ work
related to theirs. No biographical testimony makes any connection be-
.4581
::;EZE3
'
' grammarian ‘of our time’/1 He had, therefore, established his reputation by
-=;:.;,-
::_ai :¢z 3 - 1 .
.:;.\%~, . , . .
:;»:rF:,
fig:
;;§> ,, . the Augustan period. Besides the work on Aristarchus’ onueitx he wrote
I=I=,§=§§E: - - -
commentaries on the Iliad and Odyssey, and it is probably to the latter
I;';';j;‘l":; . . .
;=5;‘i‘é'_ . ._ -_ ._
;;;;a-_ -. -_ . that Strabo refers. He had a son called Ptolemy who was also a Homeric
12418 - ' '- " '
scholar.5
::.-:8"
1:2,-241:: . _ . _ .
As Philoxenus’ treatise trap) Tfig toliv ‘Pwuotitov 81ot7\.é1<Tov is not men-
tioned by Varro in De lingua Latina, it is inferred that he wrote after
Varro. But he cannot have been much later, as he is cited by Didymus and
._ . .. Tryphon.6 Of his many works, those concerned explicitly with Homer
were Trepi onpeiwv Tt0V év 17110151 and a commentary on the Odyssey.
---- .- .
=\~.~,~g,-‘-- . -- . 3 A fifth important grammarian of the time, again from Alexandria, was Tryphon. He fo-
ii‘-‘=52’: cused his attention on the language itself (parts of speech, accentuation, dialects, metre, etc.),
and while much of what he wrote was relevant to Homer, he did not produce critical or
exegetical work keyed to the text.
4 Strabo 1. 2. 31, 6 K016’ fiuéig vpappatucég. On him cf. L. Friedlander, Aristonici Hep)
=E;aE§,=
onpeiwv ’I7\1o't6og reliquzkze emendatiores (Géttingen 1853); La Roche (1866), 105-8; L. Cohn,
:-.::2.= RE II 964-6 (Aristonikos no. 17).
5 Herodian on A 423; Ath. 481d. Sada rt 3036 wrongly makes Ptolemy the father of Aristo-
nicus, but may be correct in stating that both of them érrefieixvuvto iv ‘Poipm.
"’ See Theodoridis (1976), 3-6.
48 1.3. Didymus and his Sources
**Hi
7 Suet. Tib. 56; cf. Sada 0 200, éoorbiotauoe 6%; év ‘P0311 n1. On Seleucus cf. La Roche (1866),
83-6; M. Miiller, De Seleuco Homerico (Diss. Gottingen 1891); B. A. Miiller, RE IIA 1251-6
(Seleukos no. 44). I have profited also from an unpublished doctoral thesis by Elizabeth Duke,
The Gmmmarian Seleukos of Afexandria. An edition of the fragments (Oxford 1969).
8 It is curious that these three Alexandrians who went to Rome, Aristonicus, Philoxenus,
and Seleucus, all wrote Works about Aristarchus’ critical signs.
9 Cf. La Roche (1866), 82.
1° Sada 6 872, 1 399. On Didymus cf. M. Schmidt (1854); La Roche (1866), 82, 99-105; L.
Cohn, RE V 445-72 (Didymos no. 8); U. von Wilamowitz-Moellendorff, Emleitrmg in die grie-
chische Tmgodie (Berlin 1907), 157-68; Pfeiffer (1968), 274-9.
1' Cf. E. Rohde, Kleine Schrzften (Tiibingen-Leipzig 1901), I 114-484, esp. 176f.
*2 Plin. H.N. praef. 25.
.;.
iéii
:/:-
.3,
?
scripts. This must have been work that Seleucus had done at Alexandria,
using the resources of the library, before he settled in Rome. There is no
difficulty in Seleucus’ book having been available to Didymus, who so far
as we know remained in Alexandria‘.
=s
Didymus does not cite Aristonicus by name, but there are several indi-
BI
Eé/I cations that he made use of his work. At (E9 535-7 and O 86c Aristonicus’
as
.,,
material was repeated so exactly by Didymus that the excerptor responsi-
ble for our scholia excuses himself from writing Didymus’ note out. At H
-a 256--7 the situation is the same, except that the excerptor has copied Didy-
mus first and excuses himself from writing out Aristonicus. In each case
_.,¢.
E-‘231 Aristonicus must be regarded as the original, as the information in questi-
2%
:'.-x=
=y.
on fits precisely with his manner and objectives. That he did not use Didy-
i-i
mus’ work appears at B 111 (cf. at I 18ab), where in a lengthy note Didy-
¥
W9?
mus refutes as a o;(o7\1i<ov dtvvonua, an error of the school tradition, the
A
2
Y belief that Zetig us péycxg Kpovioncg was Zenodotus’ reading and Z615; tie
é
uéyoi Kpovionc, Aristarchus’. He may be pointedly correcting Aristonicus,
who makes precisely the statements to which he objects. At any rate Aris-
,3.
tonicus is unaware of Didymus’ observations. I therefore agree with Lehrs
:5:
._.g,;. ._
and Ludwich, against Moritz Schmidt, that Didymus wrote after Aristo-
nicus and sometimes used him.”
This conclusion receives further support from a series of cases such as
the following:
K‘
.:/'.:
ta‘;
.><».
5"Yv: »
» H 5 éirsi K82 (Ariston.) (ti 5irr7\fi) E511 Ev not vpdcbetat errfiv”.
(Did.) oiirtog ’ApiotocpXog, éi7\7\o1 62-: “izrrfiv Kc”.
Jr.
H 64 (Ariston.) iiri éécv pi-zv ypdrbnrai “rtovrocj, 131:’ 0u31'0i3”, em-<11 <'.'
Q1“ Qx
.-r._
-.-4.
5%?-
. .
at “"rr6v'r0v int’ ou’>tfi1”, i-iotou
(Did.) ’Apior0tp>(og “rrévrov” 616: T013 V K051 “1'>rr’ ou.3rfi1”, rfit ¢~pu<i'
&7\7\oi at-; “rrévtog int’ 01131013”.
I 212 aiirétp era ma 1rf>p éxoin K081 we éuoapoivfinz (Ariston.) 511 €v
“,-3%:
;;_S.....
not ypoicbetou “ouirétp E-zrrei nvpocg éiv6o<; (itTl'67TTOLTO, Irouioorro at <1>?t6fi”.
(Did.) Ex/101 vpoicbouoiv “0u’>'r6ip értei Irupog éivfiog drrrérrtocto, wtouiootro
»\,,. _.
§
ta; <1>?»6E”' <3 as ’Ap’io'r0tp;(0g rouirnv [i.e. the reading of the lemma]
npoéxpivev.
53:5. .. ._
I 383 (Ariston) iirt div’ i-:1<o’tor0t<; 1r7»n(")vvru<<J5<;, tétg rr15?\tx<; oiitw st;
“div” i:i<0io'roc<;” 651 ypoi¢>s1v' ofaoérrore vétp i-zvuctfic, "Ounpoc mfihnv cbnoiv,
d0\?\& mimic.
(Did) ’Apio1'0tp>(o<; “é1<ci0t0tg”, dimer 82-; “éi<oiornv”.
I 653 K(XTOi re oufifiouz (Ariston.) iiri vpoicbetai Kai “icorroi re <§>7\éE0u”.
(aw; v v»; : :»
"21 - --
'§{_I_-.
/"
/'
/W
43*
if I
50 I.3. Didymus and his Sources
Q 7 /
on exoooetg
Didymus speaks of having consulted three kinds of witness to the text:
oti éroéoetg (which include oft oiopficiioeig), tot inrouvfiuatot, and rot
ouvvpcaiuuottot.“ We shall consider each in turn.
To us the word Ekbootg may suggest an ‘edition’, but in Didymus’
usage, at least, it does not refer only to a text that someone has consciously
edited; for that he uses the term 6t6p0u>0'ig. éxodocig is his general word
for ‘texts’, including the most ordinary and unpretentious of those current.
The distinction appears most explicitly in a scholium at P 214a, where on
icoivoti éxoéoetc are contrasted with ii] ’Ap1oro’tp)(ei0c otopficootg; this is
actually an excerpt from Nicanor, but it is fully in accord with Didymus’
language. So at A 522a‘ he cites the variant ufi Ti from (xi ’Ap1oT0ip)(ou
Kori oti éi7\7\on oxeoov rtéioon oiopfiuioetc, in implicit contrast to the vulgate
tradition which gave pfi 08. 5
The Kotvoti éxoooetg are those in general circulation, without individual
identities. Besides oti Kotvoti (e.g. A 170, P 214a as above, X 468c,
Q 214a, 344) we find the expressions fi Kowfi (E 461b, M 404, N 613b2),
on rotvotepou (M 382a, N 289a‘, X 478), on onuufioeig (E 881a‘, E 1251:‘,
235b‘, Q 7b‘), or, when Didymus wants to depreciate them, oti eikatotepott
i,,
*4 Cf. B 111b ei votp rot uuvypoiupara rtfiv iarrouvnpoirtov rrporoirroipev; B 192b‘ ictiv rofig
oiopfiuioeoi i<(x‘1 év roic imopvfipaotv oiirwg éyévponrto; F 406a1 0:3 uévov év refit; éroéosotv
oi7\7\é< Kori izv roic; ouyypoippaotv; A 3a Kort’ iévta rtfav bnopvnpoirwv év pévroi refit;
éxoooeoi Xmpig T06 V eiipapev.
:_
. it
I-ifié
‘.83 (I 3240‘, E 376a, T 95b, 189, X 301b), alias réi ¢0iu?i6ti-spot divtiypoiiboi
Z
(T 255a [v.l. eixciiétepoi], 384).”
Contrasted with the 1<0iv0i’1 izicoéoetg are the Xoiptéotepai, a category
*4
which probably includes both the ‘personal’ texts identified by a man’s
i
i name (0i’1 Koit’ 0i\/0001) and the ‘city’ texts (oii dim’) Tiiiv Tf07\f:30OV). For exam-
‘v ple, at B 12a Didymus speaks of Aristarchus’ reading as being supported
by oii 1i?iei01>i; 1<oi’i Xoipiéotepoii tibv éroéoswv. At B 53a‘ oii vrkeiouc 1<oi‘1
Xoiptéotoitoii, together with fi ’Apio"1'oi1)0'o/oug, stand in antithesis to oii
icoivoii 1<oi‘1 fi Znvooéteiog, while at B 192b‘ we have 1<o’iv toiig oiopfiiéoeoi
1<0i‘1 év t0'i<; imouvfiucioiv oiirwg évévpoirrro, “’A'rpeiiovog”, 1<0i’1 oii 1r7iei0vc
; "=:,§;
I-I-I-is
62-; tiiav Xapieotdittov oiitwg eixov, 1<0i‘i fi ’Ap1oroq>o1vs10g.16 Here it can be
Ezfji
seen that oii Xoipiéotspoii erooosig is a wider category than oii oiopfiuioeig.
=.-iéje,.z
2251;.
;.- -.< § .
.-.-.i;.
;:_,_.,.
If the recensions of Aristophanes and Aristarchus are named separately, it
:':'§1‘.5 is because of their outstanding importance, and not because they are not
;i_:_=.-i§g;;-
..,,<_ considered to be included among the Xoipléotepoil.
Where Didymus uses a phrase such as i<0i'i" évioiq Ttiiv éxfiooetov with-
:':'§='§
.-.-:2;
Elsi?
.-.-1.;
out further specification (T 124), it is likely that he is thinking especially
i i";
of the more respectable é1<56oe1g. So at F 10b he says év évioitg 'riZ>v
éxfiéoewv, rfii Te X1011 1<0i’1 rfii IVIOt(IG(X7\lUJTl1-(fit K011 tioiv ciimioiig. Some-
:-.-1 r; times he refers to ‘all’ é1<56oe1c_;, oii Triiiooii or éiitoiooit, and this evidently
-;-;=:i -I
::._*:.-
::=-1:
':'.-".1:
§_
does not mean all of the copies known to him, since at T 95b év ditroiooiig
‘.-.-:1.
:::<i:
1:
.-:;~.;
-.I..>,=
-.-<,;
is answered by i’-;v 66 T101 Tifiv eixoitotépinv, while at M 161b the reading
of oii Triiooii is opposed not only to Zenodotus (as also at N 485a‘) but to
1-I-I; §
-;-.< .~.
:-':
Didymus’ own base text, with which our medieval vulgate is in agreement.
is
:3’?-
;:j§:
=:- >-:
1 ‘All’ must therefore refer to the particular set of manuscripts that Didym-
:' j ii
;._iI2 i,- us regards as estimable. It is noteworthy that at B 196e, where A has e’i)(0v
‘5:‘$':
Eiisii
2522?? st 1<oi"1 oii pioipiéotoiroii oiitiog éivev rfic; Znvooétou, T gives oi5tiog cii
§=..,,_
§ ‘i13iE
Triiooii Tr7\i‘|v Tfig Znvoootou. Didymus cannot have used both formulations
§E:;\A-.
_._j,,.
:-.-1:‘
here. Unless he wrote 0i’1 Xoiptéotoitoii moat, a scholiast has substituted
".4
iii as
one word for the other, taking them to be synonymous.
:=:.§-:1 r5:=
Didymus’ note on F 406a’ is instructive in this connection. His own text
5551 ii
E3415
2212;
iii
E552 2'-
_
nicus, a late papyrus commentary, and the whole medieval tradition. But,
3:-ji ,;.
_..<(.
2523::
E5-'11:
arm:,.
he says, Aristarchus had dirréeuce 1<e7ie156ou, and one may well wonder
3&5
:1-';!='
21:1 >.
-.-.-1;:
where the other reading came from (Tr69ev 7T(X[)é0U), as it is not current
\‘§§.=j§:
i
:3}? E."-
g.:.i$._ __ _ _: _
15 Further examples in La Roche (1866), 87-97; Allen (1924), 271-5; see also Erbse
=.;<,_-_
-'-<>-
€= (1969-88), VI, Index iii.
.2; ,5.
*6 Cf. also B 164:-1', 1966, r 18a, 51, 362%, M 318a‘, X 2516.
I12? E
“F.
F.-
52 I.3. Didymus and his Sources
the Aristarchean éicodoeicg nor in any other of the reasonable ones at any
rate’. Again we see that there was a select set of manuscripts that Didym-
us was in the habit of consulting, at least when he was curious to seek out
a variant. He continues, Koii 013 udvov ev Toiig ékodoeoiv di?\7\di 1<ci’1 év T0'i<;
ouwpoiuuoioiv dirroilioirroivteig ofitwcg [i.e. as Aristarchus] é1<'ri0evroii. Here
again oii éicodoeig is equivalent to the select manuscripts.
In one place a Didymean note refers to the ‘old’ £:1<5é6-s1<;= I 657a, ev Tfii
istépoii nbv ’Ap1otdip;(ou “7ieiq:oivreq”, 1<oi’1 izv 1to7i?ioi'ii; ribv 01[))(0t1(.0V. If this
scholium, which is squeezed in between the text and the main body of the
scholia in A, genuinely reflects Didymus’ wording and not an excerptor’s
abbreviated expression, the dip)(0i'ioi1 presumably correspond to what are else-
where called the Xoiptéotepoii, the antithesis of the onucdoetg. They are not to
be identified with the 1T0i7i0itdi (01VT‘lYp(1(I)C£) of H 23802, which are hypotheti-
cal early texts written in the old Attic alphabet, nor with the dip)(oi’10i of Aris-
tonicus at Z 4a, which are pre-Aristarchean manuscripts that Aristarchus still
followed when issuing his imouvfiuoitoi, before he constituted his own text.
The following diagram sums up the results of the foregoing analysis.
izrodoeig
1 1
oii rrdiooii = oii X(1p1é0'T8[I)0£l oii onuiéoetc = oii xoivoii, n Koivn
*7 As generally admitted since Wolf (1795/1985), 163. Bolling (1925), 39, is cautious on the
point. The readings cited from Antimachus are set out by Allen (1924), 297--9; B. Wyss, Anti-
machi Colophonii Reliqaiae (Berlin 1936), 61-7, cf. xxix—xxxi. Cf. La Roche (1866), 22f.; Pfeif-
fer (1968), 94. —- The text ascribed to Euripides (perhaps the nephew of the great tragedian), and
cited in the ‘Ammonius’ commentary on <1) 155-6 (V 89. 17 E.) and by Eustathius on B 865,
may have been equally ancient, but it cannot be counted among Didymus’ sources. See on it
Bolling (1925), 381.; Pfeiffer (1968), 72 n. 4.
5. ii?
E::._.<\5
:_:3;.-
vpoicbei (<1) 397a2) or oi rrepi ’Av'riu0i)(ov ypoiibovoi (‘P 604b, cf. Q 753b). But
-.~:.-
.-‘iii: apart from this assumption of his that whatever was in that text represent-
5'5-I
11:}:
=.~5._
iii
:2‘:
ed conscious preference on Antimachus’ part, there is nothing to indicate
aiviii;r-
-:-it
that Antimachus ‘corrected’ the text, made atheteses, or chose between
variants. It follows that there is nothing to show that he was responsible
-.-.1 ._ for the text in any way; it may simply have been his personal copy of
E;
3'}, 5;:
:3:
Homer. Of the readings cited from it, some preserve the ancient truth;‘9
.= . S2
one (<I> 397 imovdoqniov) is a problematic replacement for a hapax no long-
FEEEE er understood; the rest are unremarkable errors, mostly banali2.ing.2°
Did Didymus have direct knowledge of Antimachus’ text, or was he
::;¢
::;i;
5;? 1;
§§1=.
using an older report of it? Aristarchus, as reported by Aristonicus and
--.;.=~
<v
as reflected in his own poetry,“ but he does not habitually refer to Anti-
'-ir.
Ii'f.=
machus’ Homeric text as he does to Zenodotus’. There is just one place,
:EigE
Eli Q]
L,
Q 71-3, where a report of an Antimachean variant appears in what looks
;=.gi¢.; like Aristonicus’ note:
gift.-
:51":
=.'<5:
di0etoi3vroi1 0Ti)(0l rpeig, iiri 1pe1380c_, Tl'6[)1é)(OU01V‘ oi’) ydip Bidi rroivroc
;§;ii
;
5-1:.‘-
ouvoiarpifiet 0i1’>ri'i31 fi ®ér1g. to 5’e “éo’ioouev” viiv divfi T013 rroipiiiuev, 0‘i0v
unoi: 7iéywi1ev' iiiriep divvoiiooivtec; oi rrepi ’Avriu0i)(0v izrroinooiv “1<7ié1|:oii
I--iii
uev diuijpioivov”.
..,.
ii
2;-iii:
1“ West (1998a), 99. Wolf (1795/ 1985), 160, wrote that ‘if some heir of Antimachus, who
@1115-
loved to carry Homer around in his mind and coins in his pockets, had perchance offered a
. manuscript written by the hand or polished by the genius of the poet of Colophon, he would
l~‘-Y 1:. -
..<,..
doubtless have found enthusiastic buyers in the Ptolemies’. Diintzer (1848), 41, numbers Anti-
E5: $5
;,_;‘.1. -‘--5. -. . . .
machus’ ‘editio’ among those ‘quae ex hominum doctorum bibliothecis venerant Alexandriam’.
-;;1..
:'=-<55-:
'9 A 298, 424, 598, E 461, N 60. In each of these cases one or more of the ‘city’ texts had
,.
f'='.‘jE:
the same reading. But as these are not agreements in error, there is no basis for the inferences
about affinities made by La Roche and Allen.
'€‘ E1
ii 2° (I) 607 (shared with Rhianus), X 336, ‘P 604, 870 f., Q 71, 753.
§='5:1=si5-.‘
55:15:
‘iii:
2’ A Id, B 2b, A 439-40/N 299b, A 754a, E 499--50021‘/500, Cf. P 134~—6€1l).
.-1
<;;.<_... _
-.
-
54 L3. Didymus and his Sources
Zenodotus
The nature and origins of Zenodotus’ text have been discussed in the pre-
ceding chapter. The question now is whether Didymus had direct knowl-
edge of it.
The case is rather different from that of Antimachus. Most of what we
know of Zenodotus’ text comes from Aristonicus, directly reflecting com-
ments of Aristarchus, who paid Zenodotus considerable attention and
took him for his principal whipping-boy. In his text he employed a spe-
cial marginal symbol, the 351Tr?ifi Trepteotiyuévn (>';), where he wished to
make remarks on Zenodotean readings.” It has sometimes been main-
tained that even Aristarchus no longer had access to Zenodotus’ original
22 Prolegomena in cod. Z, ed. Montanari (1979), 55, tfii at rrepieotivuévnt rrpoc Znvdoorov
rov 610p(-ltotrjv. But according to a fuller definition in the same source (p. 54) it was used npoi;
T0l(; yp0ic|>dic_; tdic Zqvoooteioug 1<oi’i Kpoirnrog i<0i’1 oiiirofi ’Ap1ot0’ip)(ov 1<oi’1 rdii; ‘oiopfiuiosig
oiiitofi.
51%
II.‘
.‘'~IsW-*i»n<.'.
wmI' '*~W~s1;I-L
1.3. Didymus and his Sources 55
=s»
text,23 but there is no sufficient reason to believe this. In two places the
scholia mention one Posidonius 6 dvayvtfiotng ’AptoT0'ip)-(ov as the author
of opinions on the articulation of Zenodotus’ text, which, it will be re-
::»<§
;:-s:
I:?<
‘E?/’"
.._¢
iii‘; called, was short of lectional signs.24'At Z 511 he read Zenodotus’ |i)lLl(i)60(
...;.
.--y.-
as one word rather than as {)it1<|>' ed, and at P 75 Nicanor records that he
made dxixntot dependent on 5w5i<wv, not on Géetg, and that Aristarchus
I
accepted this. Evidently Aristarchus had Zenodotus’ text read aloud to
§§Z;
-1»
2%;
1-2;
I-ii
him by Posidonius, who had to decide between alternative articulations.
._.»_
Aristarchus occasionally thought it worth noting his choices and giving
:.{r'.
:;:§;
zeg;
553;:
him credit for them.
E:i»'E
: 15;.
1;:
5 vs trpiv Aotvtxoiow oieucéot Aotyov drtoéoet. Didymus reports that the
Iii!
_.. .
iii Massaliotica and Rhianus’ text had the same, and continues Eoucev oiiv fi
.. §,_
étépot (vpot¢>r‘]) Znvooétov sivoti, as if he had no means of checking.
:14:
--<»-
=5.::<I-.
-;»1§
:-:c_-
At B 111b, a passage we have already noticed in connection with the
21%
,1
question of priority between him and Aristonicus, Didymus argues in
detail that it was a 0)(o7\u<ov dtvvdnpot, an error of the school tradition, to
assign (as Aristonicus does) the reading Ze15c; us péyotg Kpoviong to Zen-
:Wsm<ei:'5"=.'€i‘ ";-i
1.)»;
:-<...
;.-.5.
.1’;
odotus and Z6159 us péyot Kpoviong to Aristarchus. Dionysius Thrax, he
='-521'
E-125
Liii
says, was under this misapprehension, for he criticized Zenodotus for not
-2-:2
iii.
recognizing the adverbial use of péyot; and it had found its way into Twcx
$1-it
Ttiiv 1'>Tropvr| utittov. But Aristarchus himself had twice in his own writings
;;::g;
cited the hemistich in the form Zeiic us uéyotg Kpoviong, so had his pupils
iezig;
Dionysodorus and Ammonius, and so had his contemporary Callistratus.
.1»,t;
§=§=?ij:
iv
.»_»,
,..,._, ,_ .
E-5-Ii
:1-is 23 Ludwich (i884—5), I 6; Nickau (1977), 5f. Ludwich referred to N 808a Kori <3 ’Apio1"0tp—
@...@<,_‘r<:
5.‘. ii
...1;.
-.1 ..
Xog rrspi T013 otixov (the plus—verse 808a) 7\éY6l iiri év toic; Zqvoooteioig é¢>épsto. But this cor-
2:-Ll ii:
:..Z,_
responds to Aristonicus’ iiiti Znvéootog imotoiooel Didymus may have extrapolated from
i
‘Zenodotus’ to ‘Zenodotean copies’, as occasionally he may have written oti Znvoodtov, instead
if-‘-. Ef of the usual fi Znvofiotou. by careless analogy with oti 'Aplot0ip)(ov: I-I 428a‘, G 304b (T: fi A),
;j:3;.='
ii-iii
iéiiii E 89. Aristonicus (Aristarchus) has Ev 'ro'ic_; Zqvoodtou at T 26a, as does Eustathius on B 568
555i -
(p. 289. 38); but that cannot be understood as ‘Zenodotean copies’. év to'ic_; Znvofioteiotg is also
.-_
;\'.
age‘
.2 "
:-. . . . -
found at sch. Hes. Th. 5b2 (év wig Flash), cf. Pfeiffer (1968), 117 n. 5; Nickau (1977), 5 n. 16.
gr.-
24 Above, p. 38 with n. 25.
.., ;.,.
25 Cf. W. G. Pluygers, De Zenodoti carmirmm Homericomm editione (Leiden 1847), 6;
5.,i Diintzer (1848), 17~—19; La Roche (1866), 101; Nickau (1977), 4. Wolf and Lehrs had assumed
£1 2?-
:-‘ K -
i -1 g,-E5 that he had direct access to Zenodotus’ text.
;;:.. E-.,1_ .
iiiiii
...,4.
5:1 If
ii-1
=;.;§.
\.‘=:§ §_E
1 2 2 5 5 I: I 1
1-;_-4 5,:
£55? ii
~F.=.§5:
._
. 1‘
' \'.<
=»
56 1.3. Didymus and his Sources
R/oianus
The next of the personal texts chronologically is that of the Cretan Rhia-
nus, a celebrated and extremely prolific epic poet of the second half of the
third century.” We do not know where he spent his life or how his Homer
%,,i_,
.5,
1
._,,
manuscript from before 200 BC. There is no clear case of a critical inter-
55.5E%
::..¢
vention by Rhianus himself. One or two of his variants might be inter-
:'E5i2
.__,<:
.;_.,) preted as conjectures,3° but are not necessarily so. We do not hear of any
.,,.
51%
25-i i
atheteses.“ So there is no certainty that his text was a critical recension, as
::';1§
;=_51§
tilt Didymus evidently assumed,” rather than just a copy that he had owned,
."..:,.
;-.-,:
as we have argued to be the case with the Zenodotean i-ftcoootg.
2.5 2%
=....,
'==i¢
Did Didymus handle the rolls himself? If not, what was the line of
i
zizilé
:;==:;
transmission? There is a distinct tendency for Rhianus’ peculiar readings
5.5.2.,
I-fizli to be shared by Aristophanes.” This may mean that both drew on related
-ii
sources for their text, or that Aristophanes knew and sometimes followed
525%
i that of Rhianus. In the latter case we should expect him to have drawn his
I525
E;;,1,
::::
pupils’ attention to it. In fact one of his pupils, Callistratus, shares Rhia-
nus’ distinctive reading éi}\0t at K 130. On the other hand Aristarchus, so
E?
_-....;
iiiii far as we can see, ignored Rhianus altogether. Aristonicus never mentions
2
him. All the references come from Didymus.
-=1-.<-; Didymus may, then, have consulted Rhianus’ text for himself. An alter»-
5551??
--< .
i li .
=.=.=:g=-- - native might be that Aristophanes’ interest in Rhianus, while not passed
.i
‘i
on to Aristarchus, found an echo in the writings of Callistratus. We know
gait
.-:.<:'
that Didymus used these, sometimes as a witness to readings in older
..._<-..
§%£i<. manuscripts (Z 39»-49, T 327a; Z 204). Possibly he derived from Callistra-
§:iI§I§i
tus all his knowledge of Rhianus’ variants. If so, it is not surprising that
izi é
Rhianus and Aristophanes are often cited together. It may be relevant that
..,.
5:‘-it
Callistratus, like Rhianus, is cited more frequently on the Odyssey than on
$.15;
the Iliad.
tr
I--:=
Ludwich’s idea that Didymus got his knowledge of Rhianus from ‘Clear-
:;:.i,>g
.--:\~
...,;.
chus’3‘* was based on a misreading of W 81a, etinvevétov: Z-3v Tfit 'Pt0tvoi5 K081
1555
555.3?
’Apt0'r0<1>o'tvovc_; “e15n<1>z-:véwv” 6161 T013 $, sf} T661 d<1>éve1 Xpwuévwv, ting
E»
Eiiwt
5':
_“
.3-av 0 U'{'l)(O(;.
'= '- ‘%§"=i '=?_F-'3‘.%3'1-‘.
‘lfcfli
5;};-
32 He refers to it as 1'1 Kottdt 'P1otv6v at Q 85b, and now and then uses the phrase fP10tvoc_;
=:.:<.~:
:=.-1,.
?;‘;‘.=l§ Y[)0i¢61 311, 5 702, 9 299, lé 317).
5
mt’.;
.
11'. 3’ A 553, 2 10-11, T 41, r 188. ~11 81;812,e 2%, 393, t 522. The fact was noted by La Roche
(1866), 44.
I:';:1>:
14 Ludwich (1ss4-5), I 44.
‘iii:
.,_.<.. .
-.=;=,;
:=,.
\._,-1;. .. .. _
ia~'J 1,-. .'
ifs.-4;" '
a==-vY/‘.111%.
<-:.:-_-;--
ji
58 I.3. Didymus and his Sources
Sosigenes. Philemon
Two further éxodoeig are named after men of whom nothing else is
known: Sosigenes and Philemon. The references to Sosigenes are confined
to the early part of the Iliad,” so perhaps his was only a partial text. In
four of the five cases he is coupled with Aristophanes. Callistratus again
appears a possible source.
Philemon’s text is cited at B 258a as fi xarcit <I>17u*|'uovcx tov ‘(KpnTu<6v.
If he was a Cretan, we should expect tov Kpfirtx, not tov Kpntmév.
Villoison plausibly conjectured l<plTlK6V, which might imply that he be-
longed to the Pergamene school.“
The second reference to Philemon, at H 467c, is of value as showing that
Didymus’ knowledge of him did not come from Aristarchus. Aristonicus
notes that one of the reasons for Aristarchus’ orrflxfi at this verse was 5T1
ooxsi ouvitexiiofiou to ofitotoe; that is, the poet, by using oiitctoev of a hit
with a thrown spear, appears to blur the usual distinction between
ofvtoioou (used of thrusts) and {3ot7\eiv. If Aristarchus had an answer to the
difficulty, Aristonicus does not record it. Didymus more or less repeats
Aristonicus’ words: ooxei ouvicexiiofiou izvfioifie to [3ot?\e'iv icon outoioou (so
T, slightly different in A), and continues, ‘but perhaps there was some cur-
rent reading that let Homer maintain his normal diction, for Aristarchus
would not have left the matter dmcxpotuiifintov. av Toivuv tfii <I>17u‘juov0<;
oiirwcg z—’x|>épcro' “ii 62-; Hiiootoov fi’7totoev ‘i1r1tov”.’ He then quotes P 517-19
as proof that i~:>t0io0t1 can be used in connection with spear-throws. The
argument is that Aristarchus, who is constantly insisting on the validity of
his distinction between ouroioou and [30i7\eiv, surely did not allow this con-
trary instance to stand without talking his way round it, even though his
solution is not recorded. Perhaps, Didymus suggests, he answered it by
,.,,,,,.,,i._._......»----Z-—-i------
35 A 91, 124a, 4350‘, 585a‘, F 51. Heyne’s conjecture of Ztooivévnrg for Etoolcbdvng at I 453c
(not Didymus) is doubtful.
3“ Cf. Ath. 490e Kpoitnr; 5%. 6 t<pt'ru<6c_;, and Liddell—Scott s.v.; C. Wendel, RE XIX 2150
(Philemon no. 12); also ib. 2151 (no. 13) for the possibility of identifying this Philemon with
the Attic glossographer of that name.
§.5_-IQ?
2 2 5;‘;
'.3'.g,
-1?
:':.°5E
1215.2
. .5‘,
; ,.;g.
P!"
::;§-
...,.
..,\.
1.3. Didymus and his‘Sources 59
2.-,2;
555%
;5i.‘.§
preferring a variant reading known to him in the tradition, a reading that
=.-.-.s;~.
did not breach the rule; and such a reading actually exists, being attested
in fi <I>1?tfit1ovoc.
-"==>.:
5235252
-.-=".i-j.
225%. If this had been a manuscript that lay on Didymus’ desk, he would sure-
t
25.-3-Ir.-5 ly have mentioned it more frequently. The probability must be that he
Ziifi i
122*‘: ,. found it cited in one of the many ouvvpoiuuottcx that he consulted in the
....\,,.
-.-=<».;
course of his studies. 2
=.;:s"-' '
Aristophanes. Callistmms
....;s
Aristophanes’ recension was embodied in a single text with critical signs,
E_;_.;.3g_E
;:==g;:
gaze.-
referred to as 1’] ’AptoT0<b0ivou<; (A 91, 124a, 553b, 585a‘, I" 51), I’|
522%:
§:'.'.§,‘;-
=.:_:_:_;j
=;';'3_§_;E:
'§";=I$2
’Ap1oto¢0’tve10c, (A 423-4, B 192b‘), or ii Kort’ ’Ap1ot0<b0’ivn (A 17a, 142d‘,
:::_,,
f_:
5-:-1;;
:;.-X} .
::;:*@:-
2,23,. ._ --_ .. _-
- . H 436b, P 264b’).’7 There was no commentary under his name, though
sir . » ». .
there is a reference to Aristarchus’ urrouvfiuottot Kort’ ’Ap1o'roq>oivnv
..;:i.
'...,..
_'..'!;i. . . . _ _. _
. -<, . .
(B 133a), which must have been notes based on Aristophanes’ text
-'.-..,,
=.-gr. . . .
‘:1-"Q:
.=_2., -.
2:Ij;i’='
-. _ .
(before Aristarchus produced his own) or on his teaching. There were also
.1 I-I-'1-I5:
;':'5'-'i3'-
Aristophanes’ I‘7\<i3o'o0u, cited explicitly by Herodian:38 some knowledge
.=;€-;%i. 2 1-" = = - :
-=.,,-. ..
of his readings could have been gleaned from them if his text was not
.-;-;-,E-
5;.-41¢:
.-,.,,.
. .
accessible.
F
ii Was it still accessible to Didymus? His wording at A 423--4, oiitwc Git-3
eijpouev Kori év Tfit Motoooduwrucfii icon Zivcoirtxfii Kori KU‘JTpi(X1 i<0t’1
:a::::
size;
an = -.
the fact that Didymus’ central topic is the text of Aristarchus: he would
<>;-gt
..., _
\».-it , . , .
the regular 051 ’Ap10T0ip)(ov. Cf. above, n. 23, on on Zqvoodtou. On Aristophanes’ critical signs
“ti: 11
;E;a;'.'. see Nauck (1848), 15-18; on his edition, W. ]. Slater, Aristop/aarzis Byzrmtii Fmgmenta (Ber-
. lin—New York 1986), 205-10.
33 At A 567b‘; used by the exegetical scholia at K 334. Other references to his interpreta-
"-:5=1‘E;-In.=-:;1.=-;'2>-:.= -: '=e-:?
tion of a passage, perhaps derived from his oral expositions, at M 394-5, W 104a, and in the D
scholia to A 5, E 266; to his teaching on accentuation and aspiration, Herodian E 289b, O 606b,
T 30b1, .Q 84. For the Odyssey see Nauck (1848), 21—3.
i
39 E.g. F 13b, Z 148a A 137a1, K 306b, N 51a2, 245b, O 451b, W 806b; B 801a onoiwg sot’:
~‘==.;~j=: I :_- -. -
Zqvdootog Kai ’Ap1ot0cbcivq<; ypoitbouow.
4° E.g. B 164a‘ )(wp’i<; tot’) 5 eixov 051 Xapiéotottou Kori ii ’Ap1oto¢0’iv0ur; oiirtog eixsv; A
:’I’I I‘ I. -I ’.
.=';~I=.i5';1.I.'I_I.--'.'- - 137a1 fi Znvoiiétou xou ’Ap1oto¢>dtv0vg “limpet” sixov; H 436b‘ év tfil xotr’ ’Ap1ot0q>0ivn
-i
-;=,.;;:
iiiifi:
mi
IIP : 7
11.
oiirtog éyéypotrrto; cf. B 192b‘, F 51, 57a, 126a, K 349a‘, 497b. The frequent rifiétet. rrponfiétnto,
ii 1- ‘.3 ' etc. are hardly relevant, as they express either ‘Aristophanes’ critical posture or the priority of
=;'.:f' " ..
his text in relation to Aristarchus’.
=--..;.
222;;
v_:,_‘,_;_ ._
iirsf
4:» :-
:
izgs
60 I.3. Didymus and his Sources
be using the past tense of texts which he assumed were current in Aris-
tarchus’ time, irrespective of whether they were still extant in his own.
Compare the passage quoted above in connection with Philemon.
What is more significant is the fact that Didymus occasionally refers to
an intermediate source:"’ T 327a icon ’Aptorocboivng rtpon66te1 tov orixov,
(liq (1)1101 Kakhiotpatog (the grounds for the athetesis are then given, part-
ly in indirect speech, so clearly also from Callistratus); (I) 130--5a‘ ’Api0'r—
apxog 6161 tdbv unouvqudtwv ’Apioro¢>o'ivn <1>no’1 orixouc ‘ti fifiernicévou
(again with the reasoning). There are similar instances in the Odyssey:
on 424 uetouroinfifivou 56 cbotoiv into ’Ap1o'roc|>dvou<; tov orixov; C 29
K0Ot7tio'rpon"0c; st “;<o’tp1g” dvti mi’: >(0tpo’t' uetoutotfioou 256 (1)1101 tov ’Api—
otocboivn “<l>o'i'r1g” (= the vulgate); v 152 ’Ap1o'ro<1>0’ivnc st vp0i<i>e1 “ufi 66
o<1>tv”' dtvn7\6vet 5%: 51’ imouvnudtwv ’Apiorotp>(oc. Here, then, are two pal-
pable sources that provided Didymus with information about Aristopha-
nes’ text and the arguments he advanced in justification of it: Callistratus,
and Aristarchus’ commentaries. I suggested above that Callistratus was the
source in those places where Aristophanes is cited together with Rhianus.
There is little to suggest direct use of Aristophanes’ text by Didymus’
Alexandrian contemporaries. Seleucus is named as the authority for one
Aristophanean reading,” but he may have found it in an earlier scholar’s
work. An interpretation by Aristophanes cited by Aristonicus at A 4 is
certainly taken over from Aristarchus’ discussion. There are many places
where Aristonicus’ notice of an athetesis by Aristarchus is followed by the
observation that the line or lines had previously been athetized by Aristo-
phanes (and sometimes that they had been athetized or omitted by Zen-
odotus). In most cases Erbse marks these additions off and ascribes them
to Didymus.“ Whichever of the two was responsible, the information may
have come down from Aristarchus.
In the final analysis, then, it remains doubtful whether Didymus had
direct acquaintance with Aristophanes’ text. All his knowledge of it may
have come from Aristophanes’ pupils Aristarchus and Callistratus.
Callistratus’ own text is cited in a couple of places. At P 18a it is cou-
pled with Aristophanes’: ’Apiot0tp)(oc éiveu "roii éipfipov oiittog icon
fi ’Ap1o't0<|>o'tvoug K081 fi Kamiorpdtov, Kori oxeoov oiirtog i<cx’1 oti
,,i..
-1%?
at Xaptéotatou. Kori 6 ’IEitov 6v T661 rrpuirtot Trpoc rcitg éfinvfioetg éuoitoc
E§E
..>.;.
:--<:I
:'$§;.
otp ottovj oiirwc eixov, the daggered words seem to be corrupt for Kdtv Tfit
2%.;
:_.-$3:
:-2§.=. Kodduotpoitou (Hecker), or perhaps rather Kdv rfii Kurd <KcxM\i>o'rp0rrov.
212325
Callistratus’ edition is also cited in the learned ‘Ammonius’ commentary
52?;
:58;
:’:§E’:
::§Q:
(h94) on (I) 363 (V 114. 21 Erbse), where it agreed with Aristarchus’ in the
reading xvionv.
55%.; . - . -
5.".
55’-if’:
6
Aristarc/0 us
:iI'$iI - 2. -
525:-'§2 _
=-=:=-=
Didymus constantly refers to two 6K56oe1g of Aristarchus. Most often he
55% . . .
224$;
cites oti ’Ap10'r0'tpXou (when it is not just ’Api0TOt[))(0Q), or more fully 6v
Tait; ’Ap1oT0ip)(0u 6K?360e01 (B 221). In eleven places he refers to a reading
12:»: as occurring in only one of the two, 6v tfil etépou ttfiv ’Ap1oT(ip)(0u, leav-
.=2=‘-’=.-
;.:.:.,r,
'5.-.3
ing it to be understood that the other agreed with his own lemma.“
:52?-
.*.Ii‘.-':_; _-
Discrepancy between them is also expressed in other ways: 6v toiig
-:<; -,-_=_=
6m” Kori “fit 6m” (E 67a‘); fitxtiicg oti ’Ap1otdp)(0u ((3 213a, K 159a, N 60c,
2.“;
..;.~,. -.
.->34. . .
359b, 627b, E 36b, cf. B 517a); otxtfng ’Ap’iorotpX0c; (A 171a, Z 113a, A 632c,
O 450a1, (I1 213 [in h94, V 97. 31 Erbse], W 273b).’5 At T 386a Didymus
3‘-v.: av V V
=>‘»¢§1 " '1» J:
.,.. distinguishes Aristarchus’ earlier from his later reading: Trpdtepov 66
:2
=':I-.<
:'a':':'§
;=;-5;-. . . .
:1-.15;
='=-2.>., -. . ---.- - -. -.---
.:z=:¢-.
_. vpoiditov uetévpoupev iiotepov ..., while in a scholium on 6 727 he refers
-I-.<<‘.- '
;=>.~;' - . . .. . I.
-...:~
.'- -223.2’. I2 31 2 '1: 2 to fi Xotpleotépot Ttliv ’Ap1oT0ip)(0v, indicating that in the Odyssey, at least,
.5.i5£§.?;'
:,== ,,
-, .5 ; ;- g he regarded one of the two texts as superior to the other.
;_\;§;2I: 2-. 2; ;
-=..,_..
This is all consistent so far as it goes. At H 130a‘ the phrase 6v Tofig
...,,
-2:.-2, e‘
:==:g>;:
‘2s~=‘
$155.-=z-
.. .
'
6Er|t0top6v0ug ’AptoT0ip)(0u, which seems to imply a larger number of
.=:-:2,..
4 61<66oc1g, is obscure and perhaps corrupt.“ In several places Didymus cites
?§i="%§:-
?i2':I2
§"5§'l‘i ‘-
-\,.,,_..___._..__
.¢-><
-.<,.
1 > »I~I: : 44 B 13121’, 579, P 416b, A 28221, E 60b1, O 405:1, I 65721, 6819.1, TI 430b, 613, E 1820, 579.
‘*5 The same meaning is probably to be assumed in most other passages where 61)(u3<; appears
:a%:‘€:
122%‘? £5
...§,,
>.-:24». : 2- 2 2- 2- 2 -.- without Aristarchus’ name: A 142d‘, H 186, (-') 311a, I 7c, 2150‘, 368,681a1, K 505b, A 230, 325b,
i . . O 645a, TI 252a, P 631, (D 122e, X 2a, 103, 108212, 109c, '197c. 5i)(03<; is also used when a variant
*1 5 5 2 - "- "-5 :
reading of Aristarchus was recorded from another source such as the iuropvrjpatu (M 277a‘)
-=;‘:'=.3: ‘E.
:r.§.\ ;.
..
or Ammonius (Z 76b‘), and even occasionally when the variant was not Aristarchean, as at O
gt-23;;
..\,.;._--. - -. . .. ..
i‘:3;:1;-.' I =_'
311$ E3 .' I 1:1-.'.-II."-:
;: -.'.
39415, tr 61(2), 414a2(?), o eo4b(.>).
2,‘;-\~: 5.
4° Cf. Lehrs (1882), 23, ‘Si corrigendum est 6v tdic; éfinttxopévotc eosdem dixit commenta-
rios, quos hie rixpzliwpévovg [sc. B 111b, tév rji<p1.[3u>t16vwv 131ropvnp0'ttu5v]: sin recte scriptum
\\
2>>”\-3
‘E-Iéiithf
6v tofic éfintotopévaig intelligi vult editionis Aristarcheae exemplaria a diligentioribus librariis
--\= ,.'.‘:I.»I::
descripta et bene correcta.’ Lehrs’s emendation is accepted by Pfeiffer (1968), 217, but not by
<a=aa-iE5
'-’§-Fi :1 '
aYa>;,1é:
Erbse (1959), 293, who thinks that the so-called 61<66oe1<; were the same as the urropvtjpotta;
K.
I 5 = 52 2 I. = E 5 -. 1
on this untenable view see below.
=’=5>§$i
:15-' i.\\j.E
_{>c:_:::_:_':-'. :__
j"_'.-.2=:§::;:-.'.-_'.2;-.
.
62 I.3. Didymus and his Sources
r’] ’Ap101'0'tp)(etoc or 1’] ’Aptot0ip)(0u, but here the noun to be supplied may
be vpottbfi or 616p6toot<;, or it may be that he temporarily has only one of
the texts in his sights.” With 616p6too1c; the singular is regular, as there is
a distinction between 616p9too21c and 6i<60o1<;. Aristarchus’ Bttipfiwotg was
the sum of his critical activity on the text, while the 6i<o6oe1<; were the
manuscripts that embodied it. Didymus’ work was entitled irrt-;p’t Tfig
’Aptot0tp>(eiou otopflofioctog, yet in it he was constantly speaking of the two
6K56oetg.48 The relationship between the words comes out clearly in the
titles of Ammonius’ essay or essays irepi tot’: pf] vevovévou xhciovag
6i<56oe1c_; Tfig ’Apiottx.p;(eiou Biopfiofaoewg and itepi tfir; em-zxfiofieionc
Biopfitiiuetoc.
But these titles of Ammonius raise a problem, whether or not they are
alternative designations of a single work (as Wolf and others have plausi-
bly supposed).""’ It appears that in the generation after Aristarchus’ death
there was already a dispute over how many 2’-zxfidoeig of his Bidpfitootc
there were. Villoison, followed by Wolf, supposed that the second or re-
vised 6i<5ootg was not issued by Aristarchus himself but put together by
his successors from his marginalia and commentaries and from their own
records of his oral teaching.”
Lehrs accepted that Aristarchus had produced two 6i<56oe1c;, taking this
to be implied in Trap’: tfic enexoofieiong Btopfiofaoetoc. I-Ie concluded that
Tr7\eiov0t<; in the alternative title must be understood to mean ‘more than
the two’. Erbse contested this on the grounds that where one title was
cited, the reader could not be expected to interpret it in the light of the
other, cited in another place, and that the natural antithesis of xheioveg is
not ‘two’ but ‘one’.5’ But when Ammonius wrote his book, his studious
_,,_,.,.,,,_,i¢»------i---w-w
‘*7 A 203a, 273a‘, B 53a2, etc.; other passages listed by Erbse (1969—88), VI 23. For the
supplement ypotofi cf. O 24bi oipsivtov 81-. I’] ’Ap1o2rtip)<ou vpotofi; Herodian, A 572a
1<0<’1 éxexpoirqoev r'| ’Ap10to’tp)(ou. xotitot Ptévov our Exouoot; id., M 158 o’tv0t?\6vu5g pi-;v
dvavtvtéom-:1 6 Optiii, éxcxpoitnoe 51-: I’| ’Ap1ot0ip)(ov; Epimer. 1 19 (quoted by Erbse on
A 203b). Cf. Erbse (1959), 295.
"*8 Cf. B 865 fi xtxrét ’Apiototp)(0v Biépflwoig; Nicanor at P 214a, on i<otvot’1 61<56oe1<; 6)(ouoi
F] st ’Ap1o'ro'tp)(e1oc fiiépflwou; 6161 Exceptionally Didymus has at B 192b xtiv rot'i<;
Enopfitéoeot :<0t’1 6v rein; érropvijpuotv oiitwg évévpomto, apparently with reference to Aristar-
chus in both cases, cf. Erbse (1959), 286 n. 2, 288.
‘*9 Wolf (1795/1985), 194 n. 19; cf. Monro (1901), 441 n. 88. '
5° Villoison (1788), xxvii; Wolf (1795/1985), 195. Cf. Apthorp (1980), 132; van Thiel (1992),
1 n. 1, 29; (1997), 25f., 31, 34; Montanari (1998), 18f.
51 Lehrs (1882), 24; cf. Ludwich (1884-5), I 16; Erbse (1959), 276f., followed by Pfeiffer
(1968), 216. This was already the view of Villoison and Wolf. Erbse quotes Pl. Rep. 540d
ii rtixcioog ii sic, to which one may respond with Soph. Tr. 943f. 6501" e’i rig 660 j ii 1<0’ut’1
tr7\eiou<; fipépotg Aovilctou.
=:
=_-$5
-' X-‘I
5'§%1:
-$5
5%?
= $5 1.3. Didymus and his Sources 63
r
553/2:
z=.*.=;
15$:
=:;,.
._,<.
readers knew well enough whether the issue was ‘more than one’ or ‘more
2.55,:
:I5§E
I-‘-"$2
E5525
than two’, and we may fancy that we do too, as the accepted View in the
255%;
:’§§';§
school after Ammonius was that there were two. We cannot know for sure
*2
2
whether the second edition was Aristarchus’ own handiwork or a posthu-
mous reconstruction, but there is no particular reason to prefer the latter
s.
possibility. _
2‘
Lehrs’s view, then, may be accepted. Apparently the idea had got about
ii
that there existed several divergent texts representing Aristarchus’ recen-
sion. Perhaps one of the Pergamenes had criticized the Aristarcheans by
an
' -
saying ‘Who knows what Aristarchus actually wanted, when there are
'.5_IEJ.:_:
='i' I=2.i=I I 3I various different texts that claim to be his?’ Ammonius, Aristarchus’
=E5':§:
;
acknowledged successor, would have written to explain that there was just
25%:
one original €1<6ooig and one revision. This established the existence of the
. .' ; I 3
two él-ioooeig for the later Aristarchean school, and it was these two that
Didymus had in view.
"I;-' . ._
'§ .. . But did he have them literally in view? Were they there in the library
:_::_.;;§:. _ . I _ _ . .
522%‘:
for him to consult, and did he? A positive answer is suggested by his refer-
ences to ‘finding’ readings in one or both of Aristarchus’ texts: B 131a‘ év
:’e’:§§§$z
-_ rfii érépou rtliv ’Apioro’ip)(0v aiipopev; B 517a rd. roiaiitct 6i)(t'E>c; év r0(i<;
.;:..;,;
’Ap101"0ip)(ov ebpioropev; A 3a év uévtot refit; er-coéoeoi Xwpig T013 V
. ._ 2
=:=: :~i>>\-I. . _- _- :_
‘-12%-'
?I-.-;<.I
--- eiipotusv. At E 808 the situation is less clear: the bT scholium says of the
3-;;r,-fig: .. -.'- '.
-'-'-=;:-Y‘;-'_ '. .' ' '
; '.
line 015 1<orB67tou 622 e1'>pé6n év toig ’Ap1ordnp><0v (T: rrozpdr ’Aptor0ip)(w1 b),
but in A it is Toiirov tov otixov ofsx ei'>pfi06ou 1-C(X967\0U cbcxoiv év refit;
’Ap1oT0ip)(ov, as if Didymus was here reliant on someone else’s report.
There are a couple of other places where he expresses some hesitation
about what Aristarchus’ reading was: K 124a‘ |.1oi7\’ étréypero: <5 ’IEitov “uéy’
é1réype'ro”' 1<:0E1 mirrors vi ’Ap1oroip)(e1og oiirwc eixev; N 2b‘ Znvéoorog K051
’Ap1oroq:-divng “rrepi tfi1oi”' ufirror’ 0i5v oixtfig. In the first case Didymus
has found a quotation of the line in a work of a predecessor in the school,
Demetrius Ixion, and wonders if it should be taken as authority for Aris--
‘.‘.€£§§<
:'i;.’.=_\:2§I .'. - _I--
2-.- -.3 : -.
...\,
==Ys?E$I .' ' '
tarchus. In the second case he is perhaps speculating that Aristarchus began
-‘-’$§§I3"§ -. " - ' ' by following Zenodotus’ and Aristophanes’ reading and later changed his
? _iI -23%" - - "- 1=
.....‘iii.
_.:’=?’_§.:‘,5a53i. -. .. mind. But in both cases we have to ask why he does not simply refer to
.-.~'...5..
is troubling:
-=::;f$-- 5_'.' -.
Ei=.'=>':§&‘-
:'_:.;;-;4{.'.
:
.-,~.-_¢_.=,s»,-_ -. _ . -. .
Y"
;
2?}./, -
‘Q E IE ';
...»-
64 L3. Didymus and his Sources
Aristarchus’ own writings and in those of his pupils, and submitting that
these ouwpoiuuozra deserve greater credit than the imouvfiuottot. But why
does he not report what the ékodoetg give?”
(ii) At A 345—6a he notes that the lines are not athetized in the
intouvfipata, but criticized by certain members of the school (oi
fiuétepol). Why does he not mention the presence or absence of obeloi in
the i:1<?560eu; ?5“
(iii) At T 365-8 he reports Dionysius of Sidon as alleging that Aristar-
chus at first athetized the four lines, but afterwards ‘removed the obeloi’.
But, he says, Ammonius in his work Trepi Tficg §Jre1<6o6eionc_, 51op60$ot-Lwc;
says nothing of the kind. The inferences are that Ammonius’ book con-
tained some sort of register of the changes that Aristarchus made between
his first edition and the étrékooolg, that Didymus was wont to consult it,
and that he had no further means of checking such a matter.“
(iv) At <1) 130—5a‘, a passage already cited in connection with Aristo-
phanes, he records that Aristarchus in the utrouvfiuata reported a six~line
athetesis by Aristophanes and the reasons for it. Didymus then adds,
‘perhaps Aristarchus agreed with the athetesis, as he makes no answer to
Aristophanes’. Again, the incodoetcg should have thrown light on Aristar-
chus’ position.“
What are we to make of all this? Erbse has proposed the radical solution
that the two so-called z’::1<66oe1g were not texts of the Iliad after all, but
identical with two sets of imouvfiuara, one the set left by Aristarchus, the
other the book of Ammonius that registered his teacher’s changes of mind
and so constituted an érréuooolg of his work.” This is just about possible
in terms of the varied usage of the word §I<5oo1c_; that Erbse illustrates, and
it would answer some of the questions raised above. But there are grave
difficulties in it. At A 3a Didymus writes: Kort’ iivlot Ttbv ufrouvnuoitwv
“évw1vo)(6ei” <1>€;peT0u' o'i at (boron Znvooéteiov eivou tfiv vpoc¢-riv' év
uévrol ro(ic_; ésoooeot Xwpig T06 V eiipocuev. If the imouvfipara mentioned
here are those of Aristarchus-Ammonius, they cannot on Erbse’s hypo-
thesis be contrasted with the €:K660s1c_:,; if they are not, then the (Aristar-
chean) é|<i56oe1g, which according to Erbse were also inrouvfiuara, ought
to be labelled with Aristarchus’ name, otherwise there is no clear antithe-
Iii’:
z=:.
xiii
524:
law.»
» 1.3. Didymus and his Sources 65
ai
./<,-'.
I§i=:
-Ali;
ii:
-<..
._{,._
sis. Again at I-I 452a‘ on ’Ap1o1'0'tp)(0u (sc. éroooetg) are contrasted with
.§_\_
.,
.,,
1&2
the Aristarchean utrouvfiuata; Ammonius, however, had reported that
Qt:
5'8:
i
Aristarchus first marked them with dots and subsequently took them out
ii;
'26::
§’;‘.
altogether. Here is a reference to an Aristarchean text with marginal signs
.
.,_.
that corresponded to neither of the versions of Aristarchus’ recension
$8’:
=';T»;55
distinguished by Ammonius.59 Finally, let us recall that Aristonicus wrote
riiiii
._.,,.;
W .
3..
‘<8-:
a lengthy work devoted to explaining the critical signs with which Aris»~
3,.
:*:I .
tarchus had furnished the text. Where were these signs to be found, if the
{Q}: : :
§_§i'e.=. 11, ; Aristarchean €1<i50o1g existed only in commentary form? Erbse seems to
suppose that it was Aristonicus himself who deployed the signs.6° But Ari-
.4
we
».~,. , ., ; 1
fin
:2-
stonicus sometimes cites older explanations of them by Apollodorus and
Dionysius Thrax, and their presence in the margins of texts is also
--ia-
-1.. -
presupposed in a pre-Aristonicean papyrus commentary on B 763, To
onueiov iitll rrpog to oeiirepov atpérepov ciurfivrnoev onueiofvtar 5}: 51-1
-a
-ii’. 516: rravrog [; 767 to onusiov Trpog Tov <1>6[5ov, 511 i<t7\.6‘
.6’
.
Aristonicus, to be sure, assumes a single Aristarchean text and nowhere
distinguishes between two recensions. Perhaps there was no need to, so far
as the marginal signs were concerned. They had probably been transcribed
into many copies, any of which he could use as the basis for his work.
He did not need to seek out the original manuscript or manuscripts of
Aristarchus.
Didymus, however, should have pored over them constantly, if they
.-_~:~.,.
§'§’<FI I :-
:>' have two texts at his elbow that he believed to be authentic carriers of Ari~
gig .; -I
starchus’ recension, we are left with the problem of his apparent failure to
consult them on certain occasions when they should have answered his
1-:~§.<:. . .. questions.
K I
66 L3. Didymus and his Sources
Lehrs gave a partial answer. He observed that although care would have
been taken in copying Aristarchus’ text, errors and discrepancies were
bound to creep in, and this justified Didymus in undertaking his critical
task: ‘fuit igitur aliquot saeculis post perutile quae tum ut Aristarcheae
ferebantur lectiones ad fidorum monumentorum regulam exigere’. More-
over, even after producing his second edition, Aristarchus had continued
for years studying and teaching Homer and issuing commentaries. ‘Ergo
hoc etiam perutile, lectionibus editionum constitutis, variante lectione ex
utraque congesta, addere ex commentariis et ex traditione (ea vero disci-
pulorum scriptis vel etiam mernoria continebatur) lectiones paulatim ab
eodem adscitas. Tum demum recte de Aristarcheo textu Homerico consta-
bat.’62
For Lehrs, then, Didymus had the means to establish the texts of
the two éicodoetg, but realized that for a full account of Aristarchus’
516p6wo1c;, the sum total of his critical work, they had to be supplement-
ed with material from the urrouvfipata and the school tradition.
Ludwich has a modified form of this scenario, in which Didymus has
two manuscripts but does not altogether trust them:
Darauf [that is, to the question why Didymus sometimes fails to use his manuscripts
to settle doubts] ist zu antworten, dass die von ihm benutzten beiden Manuscripte
ihm augenscheinlich nicht jede wiinschenswerte Garantie fiir unbedingte Zuverlassig-
keit boten. Dass es Autographa Aristarch’s waren, daran hat wohl niemals Iemand im
Ernst gedacht; ware Didymos im Besitze solcher gewesen, so hiitte er wahrscheinlich
iiberhaupt nicht den Gedanken gefasst, ein Buch Trap‘: rfic; ’Ap1ororp>(si0v 6iop9u3oewc_;
zu schreiben. Schon der Umstand, dass ihm eine solche Arbeit zeitgemiifi erschien,
spricht deutlich fur eine bereits damals sehr merkliche Unsicherheit der Aristar-
chischen Uberlieferung.
Didymus was dependent on copies, not originals, and they seemed to him
less trustworthy than the imouvfiuorrot and ouw/poiupaTa.“3
This seems a very reasonable solution. We shall suppose that Didymus
had two manuscripts which had good documentary credentials as witnes-
ses to Aristarchus’ text, but which were not claimed to be the master’s
autographs and accordingly could not be relied on absolutely.“ Wherever
.~i
'-Es:
iiiEEE
i-it I.3. Didymus and his Sources 67
.512
.22?
1-:32
the uitouvfiuatot or the ouyypoiuuorra offered divergent or additional
information, they had to be taken seriously and perhaps given precedence
-. ==s
= age
5
E
in3
IEEI-3
-IzI=i ly acquit him of a certain amount of carelessness and inconsistency — not
_: surprising in view of what we know of his methods.“ It may be, for exam-
IE1:
ple, that he sometimes wrote oti ’Ap1o1'oip)(ou on the basis of one
manuscript without taking the trouble to check the other. It was, after all,
it-2
physically difficult to keep two rolls open on one’s desk, and to keep one’s
Ii"-i i place in them, while writing in a third.
Aristarchus’ is the last of the personal texts of which we hear. The au-
ii" thority of his recension was so great as to preclude, or consign to oblivion,
any later attempt to compete with it.“
iii
at:_._._i ,5
I;
The local texts (ai rcocrdz 7T6)u'.-31¢)
:2:5
5:.
I-E <5
"Z-2 E5
We now move on to what are commonly called the ‘city editions’. Six of
it
‘I=I'-ii}
them are cited in the Iliad. In order of frequency of citation they are: fi
§;?:-i i
=_.=’:’_§=:gi.:,i§§:.
Mtxoo0r7\u.o"ru<fi (27 citations), fi Xia (15), r'| ’Apvo?u1<fi (7), fi Zivwtrixfi (4),
a
r 1‘; Kuirrpiot (3), and fi Kpntucfi (2)67 Only towards the end of the poem do
2
i‘
2
t
we start to encounter collective references to on duo Ttiiv 7t67\ewv (T 386b‘
,._.,
- 3 <1
=:§.§':;
-<\-
~ P. Oxy. 4452 fr. 2), <1) 4540, 576a, X 51b, ‘P 206c), on tr "rtliv vtékewv
(<11 351), on 6161 ttiiv trékewv (Y 308), on raw 1r6>\r-ztov (X 294a), on KO(T0t
E;:g Q:
11?
-Z-i i?
.=_:-.“ it
it
Tf07\81Q (<I> 11a, 86d, 535a‘, X 93c, Q 82c), or oti 1ro7\1'r1K0ti (‘P 77a‘, Q 30a).
In some of these cases a reading is simply ascribed to on duff) T(.i§V Tf67\8(.0V
;§-’i5
*.‘.1.'F;"
flip???
without distinction, in others to ‘some’ of them (§v T101 Tdiv Tro7uT11<u3v
i§§§'§:
;,;.;§.
and the like). Although there are still a number of references to individu-
al members of the set through T-Q, this use of collective expressions must
reflect an increasingly slapdash approach, whether on Didymus’ own part
5i=.Ԥi: or that of a later excerptor, as he worked through the last few rhapsodies.
an
2:21:-
E515:
...,», These texts are sometimes called éxodoetg (B 865 [Eust.], F 10b, M 283
[Eust.]), never olopi-)u5oe1g.68 The readings reported from them are of the
-::.\§:
535% E
'\-==
E
'li'='§EE
:.-.= £-
1-:-zit
"5 Cf. S. West (1970), with particular reference to his work on Demosthenes.
552?’-5 i’;
1
'58 Erbse (1959), 289, correcting Wolf (1795/1985), 158.
15
E
F
<:-=e;- 1
“Yjf.
;i.§§§§E "-31. - ..
tee-.§... _
:e¢.=-:i.-
=~.»- :.-::
@5151=5:='==-.2'=::-Iz.--"-.;
:
68 I.3. Didymus and his Sources
character we might expect from early Hellenistic copies.“ There are occa-
sional plus-verses (T 351a-01, 0t 424a), omissions (P 134-6, Z 39-49,
<1) 290-2, or 97-102), and formular variants (M 281, T 76, Y 62, X 51). There
are a few readings superior to those of our vulgate, though they are most-
ly shared with other early copies such as those of Antimachus, Sosigenes,
Aristophanes, and Aristarchus.”
There are many inferior variants, and here the agreements with Kort’
tiivopot texts are of greater interest and significance. They especially affect
the Massaliotica and Chia, which tend to share conjunctive errors with
Zenodotus, Rhianus, and Aristophanes." As both came from Ionian
towns, it is tempting to associate them with the separate Ionian textual tra-
dition which we have postulated for Zenodotus. We have noted Aristo-
phanes’ links with both Zenodotus and Rhianus. The Chia is found
paired with the Cypria at (I) 492, and the Cypria with the Cretica (and
Theagenes) at A 381.
Why were these manuscripts designated by the names of cities or re-
gions? Villoison saw them as local official copies, kept for public use.” The
traditions about the Panathenaic recitations of Homer have led many to
suppose that there must have been an official state copy at Athens, analo-
gous to the official texts of the three great tragedians after the time of
Lycurgus; but there is no real evidence even for this official Athenian
Homer, let alone for anything of the sort in other towns. Nor is it easily
conceivable that there was just one official text for the whole of Crete or
Cyprus.”
Wolf reasoned that all these exemplars must have been gathered in the
Alexandrian library, and that they were simply identified by provenance,
as it was not known who had written them.” He adduced the well-known
passage of Galen where it is related that ‘Ptolemy’ (Euergetes?) was so
-wwuw-i
6° There is nothing to exclude a fourth-century date, but for the Massaliotica and Chia, at
least, anything earlier than that is very unlikely in view of the spellings opevg (F 10), 7\uJtsuv~
rot (M 283), 6lp)(86VT ((11 282); cf. Bolling (1925), 40. In any case we should beware of treating
these manuscripts, which came from widely separate parts of the Greek world, as a unified
group.
"’ A 424, 435, 598, N 60, Q 109.
7‘ A 97, O 44, H 59, P 134-6, Z 502, T 76, 86, T 188. In the Odyssey the Cypria is associa-
ted with Rhianus and Zenodotus at u 135 (P. Oxy. 3710 ii 7f.).
72 Villoison (1788), xxvi.
73 Cf. Allen (1924), 291; Pfeiffer (1968), 110. Ludwich (1884—5), I 8—10, surveys earlier views.
74 Wolf (1795/1985), 160, cf. (1830), 17. Most scholars have accepted his view. We need not
dwell on the bizarre notion of Bolling (1925), 41, that these were editions prepared at Alex-
andria for export to the places named.
i
1
.=.
1?;=1;
.5;
HE
I.3. Didymus and his Sources 69
s
zealous in the acquisition of old books that he commanded all who sailed
.-M
HY:
:1i=
1?:
iii;
in to Alexandria to surrender their books for copying; they received the
"Bi?
new copies in exchange for the originals, which were retained for the
library and labelled Ttfiv ex 7t7\oiwv together with the owner’s name.” It is
a fine story, and something of the kind may have happened now and then
when a ship was found to be carrying some rarities. It may be doubted,
however, whether Ptolemy’s officers would have bothered with texts of
Homer, of which there was no shortage, or detained the mariners while
books of the Iliad were transcribed anew. The local texts may have been
acquired more casually over time. The fact that certain of them came from
such remote places as Marseilles and Sinope is hardly sufficient basis for
Ti
';l\-»-i sea-¢.'~I.“'.I':"':'
:;,_._
the suggestion of ‘a calculated desire to get texts from peripheral areas’.7"
=_;fi._
5.-,,;
5355
.,.,.
That would have been an astonishing anticipation of a modern methodo-
logical insight.
i i?
at
Wolf and Dtintzer, as we saw at the beginning of the last chapter, as-
4:5
=4;
.15:
‘a
in:
sumed that all these copies were already present in the library in Zenodo-
lr
tus’ time and that he collated them for his edition.” It has remained a
i
widely held conviction among scholars that they were at any rate collated
"E45?
E35:
EE§;'.
; by Aristarchus, and that Didymus’ reports of them derive from Aristar-
chus.78 As I indicated in the previous chapter, I believe this to be a serious
55;;-:
555%?
.4
misconception. Here are the reasons. ~
‘=‘.3; I
I. At Z 39-49, after noting that Aristarchus’ athetesis of the eleven lines
nt-
had been anticipated by Zenodotus, Didymus adds: Koddtiotpottog 0135i: év
._\ \.
‘_-I 5;‘
5.‘ =5 Tfit ’Apyo7\i1<fi1 dmolv 0u3'ro13c; <j>épeo6ou. The authority for their omission
-E-1
E_I;
in the Argolica, then, was not Aristarchus but his contemporary Callistra-
4;".
5325
tus, whose writings are elsewhere cited by Didymus, Herodian, and Por-
phyry. There is no basis for the assumption that the statement about Cal-
‘Y:-.-3 ;.-ii
listratus is a quotation from Aristarchus; it is very rare for him to cite
551;;
F5. ii;.-.
;-. other scholars by name, except when he refers to the texts of Antimachus,
.5
iiili:
:32.
Zenodotus, or Aristophanes. On the contrary, there is some likelihood that
i-3-S
.. ,_
:1;
;. ,.
ea
Callistratus was commenting on Aristarchus’ athetesis and that his state-
?;i§i_
ri- ment about the Argolica came in his work Trpog trite; d6eTfio£-Jig which
Didymus cites at A 423—4.
25.5§
14- . .
:11 t.=
IE?i.
=;<. £- -i-----------------------------
.. ..
‘Jufi:
3.
1:14: 75 Galen in Hipp. Epid. III 2. 4 (p. 79. 8ff. Wenkebach); Wolf (1795/1985), 160f.; cf. Diint-
ii-
zer (1848), 40. The passage is quoted in Greek and discussed at length by Allen (1924), 292 f.;
-; 5.
..-:--
cf. W. D. Smith, The Hippocratic Tradition (Ithaca—London 1979), 199-202.
a=.u=.-
7° Haslam (1997), 70.
~t;§ ‘.
,_. E.
’;
5 _: -_'3': ; : :
77 Wolf (1795/1985), 158-61, cf. (1830), 17; Diintzer (1848), 40f., ‘has editiones, quas iam
‘=i:.
Zenodoti aetate Alexandriae fuisse dubitari nequit’.
ii
78 So e.g. La Roche (1866), 21; Ludwich (1884-5), I 44, 195f.; Erbse (1959), 280; Pfeiffer
ii-
::; 1,-
(1968), 94; ]anko (1992), 26; Nagy (Z000).
,..
=-:< >: .
2; -'= z;:.'--"J-1-.
,:-_.
>:1 r
».-...
/..-,
.
.-
14;:r,_j.-.-_:_:_I_-:;:_-_
'<';'_~ 3‘.
¢,,
._ _.
70 L3. Didymus and his Sources
The case is even clearer at A 381 érreii u0i7\0t 0i ¢'i?\0c_; fisv, where Didy-
mus notes: )1é7\su1<6g ¢>n0w év Tfil Kvrrpion K081 K[)T]T1Kfil “érrs‘i poi v15 0'1
¢00>; iiev”' rot‘: (£~)s0tyévn<; 5%. oiitwq :rrp0¢»éps'r0u. Here it is Didymus’ con-
temporary Seleucus, not Aristarchus, who knows the reading of the
Cypria and Cretica. Similarly in the ‘Ammonius’ commentary on (I) 290
we are told that Seleucus év T051 y’ Kottéz ttfav ’Ap10tcip)(00 onpsiwv defend-
ed the verse against Aristarchus’ athetesis, but that év T031 5’ ‘ro'3v
Aropfitoructfav the same scholar himself athetized it together with the two
following lines, and reported that all three were absent from the Cretica.
These are the only two Homeric passages where the Cretica is cited, and
both times Seleucus is specified as the source. The Cypria is cited three
times in the Hind, and at least one of the citations came again from Seleu-
cus. It seems likely, therefore, that he was the sole source for those two
manuscripts, at least. There is no likelihood that he got his knowledge of
them from writings of Aristarchus that were not available to Didymus.
In the seven citations of the Argolica Callistratus is only once identified
as the source, but it is a reasonable hypothesis that the rest also came from
him. If he consulted it in one place, he presumably consulted it in others.
Those who believe that all or most of the 1r07\1"r11<0ti were collated by
Aristarchus offer as their strongest piece of evidence the scholium on
A 423--4:
Aéiig ’Ap10t0<;5)(00 at T06 0t'1'fi<; ’IM<i60<; i>1r0uvfi'pa'r0<;' <1 ‘pet’
dctuiuovotg’ en’ duéuovag, 0: ton rrpbg 601030009, tiyotfioég. -1“
-2
Q’
O ’ 'on
2 :
01’
O1’
‘Kurd
6061101’ dtvfi T013 00 606101‘ oiitwg yup viiv 'Opnpog 'ré6eu<sv. i-ivroi
5‘e rrolofioi ‘nerd: oofita’, iirrtoc; fit 0u3'r0'ic; 0u’>t66ev to uetoi érri. Xptfivtoti
615 K031 Tl'7\6iO\/SC; ciimor toiiv 7TOlT]T(l§V Tfil rzorrdi dvti 'rfi<; érri. Zo¢01<7tfig'
‘éytb Kort’ 0u’>'r6v, dag opting, éEép)(0uou’.” oifittog Bi: ciipouev Kai év Tfil
Mozoooduwrucfit icon Xwtorrucfii 1<a‘1 Kvrrpiou Kai ’AvT1t10t><sito1 iccxi
’Ap10t0<bcxvz-zicnt. Kamiotparoc 62-; év T031 1'Ip<‘><; tétc cifietfiosig éuoiwg, K011
0 Z1603v10<; mi 0 ’IEiwv év T031 iéicrtoi Hpog rétg éfinvrioeig. Kori To ‘G001 st
<éiu0t> rrdtvtsg ‘érrovtou’ 5101 T013 oi.
Didymus begins with a verbatim quotation from Aristarchus’ commentary
on the first rhapsody. Aristarchus glosses the phrases pet’ dpliuovotg and
Kurd ooiitot, and notes that ‘some people’ read t1s'r0t ooiitot with the same
meaning. To justify 1<0c'r(;i as a genuine poetic usage he quotes a fragment
of Sophocles (898 Radt). Now, is the next sentence, oiirwg at siipouev I<1'7\.,
still part of the quotation from Aristarchus, or is it supplementary infor-
mation appended by Didymus? The following sentence, with its citations
of Callistratus, Dionysius of Sidon, and Demetrius Ixion, is obviously
Didymus, so that the quotation from Aristarchus must end no later than
as
2*;-:
_-it
it?
i
1.3. Didymus and his Sources 71
'12
‘.5’
<6;
it
i?
i
’Ap10To<1>otveit01. Many scholars have taken it to end there, so that it is
:_§li§ Aristarchus who says he has found the reading he favours in the five
iiiii:
um on ‘P 870-1, except that here the latter precedes the former. First he
egg‘ 3.
5;: $.-
51? §:
quotes the readings of the Massaliotica and Antimachus, then he continues
.-I: §;E:
Q-.
t
X_ 0 uévtor ’Apiot0tp)(0c; 610: tdiv imouvnuciruov 1<t7\. In T it is the other way
._-.i ,._
E;-I?
-..,<.
round, but the manuscript reports are clearly marked off from Aristarchus’
-1:8;-.
.55-'1 .
i
._._ _
contribution: énsiyopevog dtrréoirotoe KT7\. oiirtoc; ’Apio'r0tp>(o<;. fi 5‘s
2:5-‘E-.
Maoocduwttkfi oiitur ’AvTiu0t)(0c; 06 Neither version supports
Erbse’s opinion that ‘was er [: Didymus] zu W 870/ 1 mitteilt (Fassungen
:;-:, §:
l
i
i der massaliotischen Ausgabe und der Edition des Antimachos, vgl.
: \.=. g-
Y-E1 . - Scholion T), fand er augenscheinlich in Aristarchs Kommentarenisl
ii
.1
;;-\;.
Didymus’ statement at A 423-4, oiitwg 0i-3 eiipouev Kori év Tfii M0t00ot7\1—
=;'-. ,.;
2:5:-fa
:1-1-<.. $.-_.
)1.)1.
tomcfit icon Ztvtorrntfii icon KUTfpiOt1 K081 ’Avr1uoo<sit01 itoii ’ApioT0<|>on/sitot,
.~
2:
2'53 £1 suggests that he had personally consulted at least some of these exemplars.
iii: %:
'=-.\..,.
:=i'I< ii
:a:;;,=.
iiiii I allowed earlier that he may not have had access to Aristophanes’, but his
:=_:_J;.
=23‘-‘ ,.3:
5..
333:3 E:
‘:';';'1 5'
c’-'-5 >.'
;E;j:;
i -
:;--'et-
-'vI=.*§< 79 S0 e.g. La Roche (1866), 100; Ludwich (1884——5), I 195 f. (on the grounds that Didymus
!='.-: r.
aw- did not have those manuscripts, and that the construction is not continued after ’Ap10'ro<1)0tvsit0t);
i:a':':§I
1 ».1:
Erbse (1959), 280 and in his edition; Nagy (2000).
.,,.. 3° T. Bergk, ZAW 1846, 500 = Kl. p/ail. Sc/ar. II (Halle 1886), 438 n. 27.
8' Erbse (1959), 282.
\‘;’:::-El $1
Ii ‘F
;-at . . ..
E>P=".‘£E
EPFIQI
55:141.
.-iv.
\.'=:‘.;f;.
1».
'IA
-,-=>_.,.
§_..,,_
//0..
,0
is
72 1.3. Didymus and his Sources
siipouev should be valid with respect to the items that head the list. The
Massaliotica in particular has a strong claim to be among his direct sour-
ces in view of the frequency with which he cites it, 27 times in the Iliad
and thrice more in the Odyssey, and his habit of naming it first among the
1r07uru<oti. We have the impression that he had it within easy reach and
turned to it more readily than to any of the others. By the same criterion
we may judge it likely that he had direct knowledge of the Chia, which he
cites fifteen times.
To sum up, the indications are that Didymus himself culled readings
intermittently from the Massaliotica, the Chia, and possibly the Sinopica,
but that he derived from Callistratus his information about the Argolica
and from Seleucus his information about the Cypria and Cretica. None of
the six manuscripts was necessarily much older than the scholar who first
cited them, if the Alexandrian librarians continued the practice of acquir-
ing books from overseas and labelling them according to provenance. The
character of their variants, and particularly of the Massaliotica with its
plus~lines after T 351, might suggest a pre-Aristarchean date, except that
Aristarchus’ standardizing influence may not have been effective in distant
lands as early as it appears to have been in Egypt.82
i_
82 The use of the pluperfect at F 10b, év évioug rtfiv éicfiéoewv, rfii re Xiou K00 rfii
Maooaaimtiafii Kori new 6E)0\oug. in aiuipoug éyéyponrto, need not imply that they were pre-
Aristarchean, as inferred by Wolf (1795/1985), 162 n. 43; see Ludwich (1884-5), I 45 n. 58.
éyévpomto is used also in reference to Aristarchus: B 133a (his earlier inrouvfiuottot), 192b‘,
l" 415b' (év Tfil érépou), K 397-9a, h94 ad (D 126 (V 86. 24 E.).
33 Cf. La Roche (1866), 84f.
-1-31%
3;
55:;
we
1.3. Didymus and his Sources 73
..g
I-‘E2 Er
under his pillow, the one known as fi (it7T0 T06 v0ip0n|<0g.8‘l It evidently did
i.=i;5.=i
not find its way into the Ptolemies’ library and played no part in the scho-
iri ilfil lars’ lucubrations. The other is the one referred to in the so-called Anec-
. _,
dotum Romanum of Osannzss '
ii
ii
t013to' “M013o0tt_;. deifiw Kai ’A1r67t7\wv0t i<>tu'r6t0Eov”, ting K00 Nmoivwp
.-_--.3;
i uéuvnrou K00 Kpoimg év "mic; Atopfitorucoig.
.-355$
2 2-1-E25
15:;-,9;
;:.-.=.-is The incipit shows that this manuscript had a ‘Homeric hymn’ prefixed to
--.<<
;‘.‘='I-55?
the main narrative.” The citation of ‘Nicanor and Crates’ probably means
.;-...);
‘Nicanor, who mentions Crates as his source’. Crates, however, could not,
:';_¥;3-i .
:'='=§El;5 for chronological reasons, have described the book as Apellicon’s. Apelli-
con was a rich collector from Teos who acquired from the descendants of
Neleus of Scepsis the dilapidated library of Theophrastus, which included
".I,fif,4,ei l some manuscripts of Aristotle. On Apellicon’s death in 84 BC Sulla
brought his library from Athens to Rome; there it was rummaged over by
,-.-.-.-,-.=t- . - . .
Y
the grammarian Tyrannion, whose work on accentuation is often cited by
:E:'.I':ii' ' 3. '
Herodian. If Apellicon’s name is rightly recognized in our text, we may
a speculate that Tyrannion found among Apellicon’s books an Iliad with the
; "; ';1:§E' “
peculiar incipit that he had read of in Crates’ Diort/ootika -- very likely the
5'2-I-i-{qt
' '
same volume - and that Nicanor got his information from Tyrannion.
i
55:5...
._-_._.,r
.- .- = . : ‘ 5
: .;.;I;I;.H§§ " It may be mentioned in passing that the scholia to the Odyssey cite a
::;-_-3;
couple of other oddly designated z’-:1<560e1g of that epic: Ii i<ui<?\u<r’| (Tr 195,
'1
p 25), and fi em ]‘Movoeitov (E 204; Mouoeiov Buttmann). The latter was
Z-Z-L321}; L
cited by Callistratus. The former was perhaps part of a Gesamtaasgabe of
E?-i¥.?.?i ;
the Epic Cycle.
. i
tdt tirrouvfiuorrot
'12.;
We now come to the next category of works used by Didymus for his
Eé%§&; li- _. -.
;=:;...i
reconstruction of the Aristarchean 0i6p6u>0t<;, the ivrrouvfiuocrot. The term
..,=.’,' ,' . .
iE:§E_f 5
:::_:::\r
is usually rendered as ‘commentaries’, and this is more or less acceptable
é
inasmuch as it signifies a work of systematic exegesis that follows the order
,r
of the text, picking out from it words and phrases for comment. We have
1
1
1
8"’ Above, p. 25 n. 73.
S5 Le. the prolegomena that appear before the D scholia in cod. Z, most recently edited by
i Montanari (1979), 47-56.
E
3° So Schimberg: rim’ e7m<(I>voc; cod.: dd)’ ‘E7\- Osann: 'Arte)\?\1|<tf>v0<; Ribbeck. ’A1rsI»\Pu1<<i5v-
was 2 »
*'»'F:i2\'I .
Tog (Strab. 13. 1. 54, 14. 1. 30) is more likely to be the correct form of the genitive than —i<tBv0c;
:;-=;:.r
r:===-st.
zziizi -- --
(Plut. Salla 26. 1).
;e;=_-.-;
37 Cf. West (1966), 150; (1978), 136f.
-=.= -' .
l
li
r
.'
">;§;?E
i
‘:\"' |
74 I.3. Didymus and his Sources
For our present purpose we may allow the sense to fluctuate between
‘commentary’ and ‘set of notes’.
So far as Homer is concerned, the genus first appears with Aristarchus.
Didymus several times names him as the author of 1‘>Tr0uvrju0tT0t and
quotes sentences from them: Aéiig ’Ap10T0ip)<o0 éx T013 oz’ Tfig ’I?u0i00<;
i>rr0uvfiuotTo<;, A 423-4, cf. B 420a‘, 435a‘, F 406a‘; 6v TlVl T<I>v fixptfiwuévwv
farrouvnuoirwv ypdcbei T0u3T0t l<OLTO£ 7\é‘<§1v, B 111b; ’Ap10T0ip)(ou Aéietg
ex Tdav 13TrouvnuoiTwv, B 125a; ev Toic 1<cxT’ ’Ap10T0q>0ivnv brrouvriuotoiv
’Ap1oT0'cp)(0u, B 133a; ’ApioT0tp>(og év i3iTouv1‘§uaTi qmoiv oi5T0oc, Y 471a‘;
’ApioT0tp)(o<; 010: Ttijv iurouvnutitwv qmoi, <I> 130-5a‘, cf. ‘P 870-1a.
Aristarchus may be understood as the subject in other places where we
find just év Toig imouvfiuaot (Z 4a [Aristonicus], H 365a) or 510 Ttfiv
i>Trouvnu0iT0ov (A 391b). Conversely, where his opinions are cited without
a particular work being specified, we may suppose that the reference is to
the i3TrouvfiucxT0t: <1>fiotv 0 ’ApioT0tp)(o<; ESTT (I 159, cf. I 401b, N 808a,
E 371)’, 208, O 469-70, II 105-69., T 81—~—2, <19 3633).
In two of these places it will be seen that a distinction is implied be-
tween different Aristarchean 0Tl'OpVI‘i11(XT(X. At B 133a the reference is to
Aristarchus’ i>TrouvrjuaT0t Kort’ ’Ap10To¢>0ivr|v, presumably a set of notes
based on Aristophanes’ text and perhaps in some part echoing his teaching.
At B 111b it is to T0t fixptfiwuéva 0TfO1lVI‘|p(XT(X, notes prepared more
painstakingly, evidently the seczmdae came.
Didymus does name one other author of t‘>TtouvfiuaT(x, if only in
G) (296b, 441a): an otherwise unknown Diogenes (RE no. 52). That should
warn us not to assume too readily that all references to 1‘>Tt0uvrju0r.T0t
must be to those of Aristarchus. At B 111b, besides quoting from the
fixpifiwuéva i>Trouvfiu0tT0c, Didymus speaks of ‘some of the ivrtouvfiuomtx’
which included the gloss uévot tin/Til T013 ueydkwg, mistakenly presup-
posing uévot as Aristarchus’ reading. At B 221 Ev T101 Ttfw TfOO\(J!.t(i)V
notes that appeared to be old? Didymus several times uses the vague
expression 6v T101 or 1<0tT’ 6vi0t rev i>Tr0uvnu0iTwv.39 Are these Aristar-
rem
'l'0t ovwpo'iuu0tT0t
It remains to review the numerous other writings on Homer and Home-
"~‘ em1u‘-,mv r>*:
ric problems that Didymus refers to. The earliest of all who wrote about
Homer, the allegorist Theagenes of Rhegium, is cited for a variant at A 381
WM
together with Seleucus’ report of the same reading in the Cyprian and Cre-
tan texts; it looks as if the reference to Theagenes also comes from Seleu-
cus, who was interested in Stoic-type allegory and etymology and may
have known of Theagenes through a Pergamene source such as Crates.
Sophistic and Peripatetic discussions are occasionally alluded to. The
lines T 269-72, which Aristarchus athetized, iTp0n0eT013vT0 Trap’ éviou; T03v
00<i>10TtT)v, év eviou; 5i-3 01508 ia<i>6p0vT0 (schT: Aristonicus or Didymus?)
Readings of Chamaeleon (c.350-after 281) are quoted at M 231a, T 62c,
~F>=m\% and ‘P 94, just as if there were an i5i<0001g under his name, but they were
probably taken from his work in at least five books irepi 'I7\ici0oc;, which
is cited in the Geneva scholia on (I) 390 and in sch. Ap. Rhod. 2. 904-10a.9l
3" A 3a, E 857b, I-I 6a, 95a‘, 452a‘, etc.; Erbse (1959), 278 f.
9° At II 467a’ oi imouvnurxtlototi apparently = Didymus; Erbse’s marginal ‘Did’ was
perhaps intended for a‘, cf. Erbse (1959), 286. Aristonicus is cited in the commentary on 0 in
av-1.
P. Oxy. 3710.
9' Cf. Lehrs (1882), 26 n. 9. For Chamaeleon’s Homeric studies see F. Wehrli, P/aaimas von
Eresos. Cbamaileon, Praxiphanes (Die Schule des Aristoteles, ix, Basel 21969), 52f. (frr. 14-22)
and 75-8.
aka-
76 I.3. Didymus and his Sources
i.j.
92 See Lehrs (1882), 37f. Cf. also the explanation of his onpeiot in the Z prolegomena, p. 54
Montanari: ii pi-zv oiiv 611r?tfi darrspiorncrog rrocpottifieron rrpog tonic; v?\woooypo’ic|>0ur; ii
iarspooéfiwg émfisfiapévoug T6: T06 irolnrofi K051 pi] 1<c17\tBg' ii rrpbg roar; éiaafi eipnuévotc; Aéfis-;1c;'
ii rrpbg T6: évotvrioc Kori potxépevot, Kain iétapoi O)(l’i}.l(XTOt Tr0imroM0t 1-ca‘: Cqtripatot.
93 He does make one other reference to a glossary, to explain the adjective stincbevéoav which
Rhianus and Aristophanes read at ‘P 81 (above, p. 57). The glossographer is named in A as Cle-
archus, but probably Clitarchus is meant; on him see W. Kroll, RE XI 654 f. (Kleitarchos no.
4).
94 ’A1ro)\)\o$v1oc_; A: 'Arr07t7\65topog T, no doubt under the influence of I 147a2.
95 For edition and commentary on his non—historical fragments see Montanari as cited
below, n. 101.
i
r
f‘
;.:;:..
;=-=L@=:
iiiki:
:-£5
1%;
I.3. Didymus and his Sources 77
5.;
§=ié‘:_I
Athenocles lived not later than Aristarchus, whom he anticipated in
iQ
‘IE-i i?
.:.:.g _.. atheteses at C 144 and E 503, and perhaps not much earlier, as Ammonius
5:5:
no commentaries of his own, and we cannot say in which other work he might
:..W_ have treated such a question, the likeliest hypothesis is that his views were
g
mediated by Aristarchus, perhaps in the imouvfiudrot Kat’ ’Ap1oT0<1)civnv.
.,,,_.
ax:-.
&-. dual and the plural, Aristarchus several times castigates Zenodotus for
.1???L
2555:;
i ignoring it, but in one place he names Eratosthenes and Crates as offen-
;::s!_
:52‘. .' .': ;-
ders in this regard: oi 6é7\0vte<; ouvxeiofiou rd 6u'1'1<6t Trap’ 'Oufiptot,
’Ep0tT0o6évn<; Koii Kptitng, Q 282. He scolds Eratosthenes at K 364b for
another inaccuracy concerning a dual, namely his construing 61c61<t:T0v as
an imperfect instead of as a historic present.98 But it was Aristarchus’ con-
M,-5
§:_: g.
..
temporary Crates who was the more active of the two in Homeric studies,
..r§
i
\'?=i$.:'
JE
i
enough to earn him the sobriquets 6 'Ounpu<6c; and 6 1<p1'ru<6g. He wrote
a 5tdp6t001c_; ’I7\10i5oc; Kai ’O5v0oei0tg in nine books, also cited as
Alopfitotucoi, and another work in at least two books cited as 'Our|pu<0’t.99
51%: -2 -_ __ .2
Aristarchus probably criticized him more often than now appears, as
according to the account of the Aristarchean onueiot in the Z proleg0me—
na his readings as well as Zenodotus’ were signalled by the §tTr7\fi
.
vtepteortvpévndoo His exegetical work finds reflection in the bT and D
at ' '-
I
-21;: ..
“\\.j;- .' -. -- -.
an,
_ E
".1-12::-;;II'.:3_3-_'-.'-- j
if_'§.:=§'3§'j-:1-1:
10° Quoted above, n. 22.
.-K
\'=:-r._
:,:;..
_._,_, §. .. _ _ . _ .
h E
\»:.\,:;__
/
,\}§>>_,§;\§
78 I.3. Didymus and his Sources
i__.
‘°' Sada 71' 3035. The fragments of Agathocles, Hellanicus, and Ptolemy are edited by
F. Montanari (together with Lesbonax ed. D. L. Blank and Comanus ed. A. R. Dyck) in vol. 7
of the Sammlung griechischer and Zateinisc/yer Gmmmatfleer (Berlin--New York 1988). On Pto~
lemy see also La Roche (1866), 75 f.; A. Dihle, RE XXIII 1862 (Ptolemaios no. 78).
‘°2 Ludwich (1884-5), I 48 n. 63, ‘Ich denke mir, dass es ein Eihnliches Werk war, wie das
des Didyrnos iiber Aristarch’.
‘°3 On him cf. R. Schmidt, Commentatio de Callistmto Aristop/oarzeo in Nauck (1848),
307-38; La Roche (1866), 67f.; A. Gudeman, RE X 1738-48; I-I.-L. Barth, Die Fmgmente aus
den Sc/mften des Gmmmati/eers Kallistmtos 214- Homers Hias and Odyssee (Diss. Bonn 1984).
-15%:
'_?:§E
'='-$2
1153;:
case of Athenocles (above), the juxtaposition may mean that Chaeris was
Didymus’ source for Diodorus. A Diodorus is also cited in the papyrus
commentary on u, P. Oxy. 3710.
as
Of Aristarchus too several different writings are cited besides the éxodoetg
and uftouvfiuata. Didymus refers to his Trpog Ktouocvév (A 97-9, B 798a,
Q 110b*), rrpog dltkitav (A 524a, B 111b), nepi ’I7u0’r6o<; Kori ’O6uooeiotc_',
(I 349-~50), and Trpoc, to Eévtovog ncxpoioofiov (M 435a‘); Aristonicus and
Nicanor cite also his treatise rrcpi T013 votuoroifiuov (M 258a, O 449—51a,
55$. . -
>3 as-9).1°6
These works were naturally treasured and studied by Aristarchus’ Alex-
andrian successors, who may properly be described as constituting a
‘school’. Didymus himself uses the term, referring to the Aristarcheans as
0'1 duo tfig oxoafig (K 225, P 481a, 681c‘), oi dot’ ou3'ro13 (B lllb), oi Kort)’
fiutig (B 865), or oi fiuétepol (A 345»-6a).1°7 He identifies Ammonius as 6
ottxoefitiuevog Tfiv o)(o)\fiv (K 397—9a). We have referred more than once
to his discussion of a o)(07\u<ov dtvvonuot at B 111b, where Dionysius
Thrax and Aristonicus are among those in error. A reference at A 455a to
vpotuuattotai, whose reading is opposed to that of Aristarchus, is taken
by Erbse as a further equivalent of oi duo Tfig o)(o?\fic_;.
-‘ii-55 I
z;;':\E.
=.-:; 5. -. - . -.
3;-1E5-.' -
Ammonius, as Aristarchus’ direct successor, is one of the most impor-
i egg 5 tant figures. He must have known everything there was to be known about
=;~;-; ;
-;.%-;<§f- .. .. . .-
25,:-4“;
the master’s teachings, and Didymus cites him as an authority on them
;i:=.€--§.:ir%r:"- .--' - -. -.-
_-:-:-'~t-. -I :;: ;.'-
_,.:.jE_. -_ ._ ._
-t':fg' :.' 3 : .
._._.M-
riifilii . - - ...
ii‘-15
:.f_'l: -
_.?;;?-'-.' I
5. _
:
‘°4 Cf. Aristonicus at <1) 126—7a and 169b/c.
E;>.:,~:
;a:.\E. *°5 See L. Cohn, RE V 708 f. (Diodoros no. 51), who proposes dating him in the first half of
=;_.,.;_
§=
the first century BC. ‘Er starnmte wie Artemidoros <5 ’Ap1oTo(§>oivetog aus Tarsos (Strab. XIV
ea,
675), wo vielleicht eine grammatische Schule bestand, die sich nach Aristophanes von Byzanz
:-::<r-'-r::.- -- benannte (Ahrens Bucol. gr. II p. XL).’
'°‘* cf. Pfeiffer (1968), 213.
‘°7 Cf. also Ap. Dysc. Ado. 145. 6 0'1 (into tfic; ’Api0ro'tp)(ou o7(o7\fic;; Herodian at A 396b‘
“iii? E
;5=:.i;_
'Apiot0tp)(oc; mi oi om’ ouitofi. and A 2c 0'1 dot’ ’Ap1oto'tp)(ou.
i
J
it.,._.
-v::.ii'
,-,._., K.3 -'.
=,==;;,
/‘;=;;;;; 3 3 - _; : ; . ; ; -. ;..
80 I.3. Didymus and his Sources
.»i~
")3 Cf. [Longin.] 7T. iitpoug 13. 3. Ammonius is cited also at O 162b. For his extra-Homeric
scholarship cf. L. Cohn, RE I 1865 f. (Ammonios no. 16).
“*9 La Roche (1866), 73-5; Pfeiffer (1968), 252-66.
“° Cf. C. Wendel, RE XVI 2477. Neoteles is also cited at G) 325a‘ (Nicanor) and II 175c‘;
Aretades at y 341.
1“ B 111a/b, O 221b, M 3019., O 86.’-1, 71221, II 810a, P 24a, 12551, 218i)‘, T 4921.
“Z On Dionysius Thrax cf. La Roche (1866), 70f.; Pfeiffer (1968), 266-72.
“3 Didymus speaks of him. as ‘using’ the Aristarchean reading without comment. On him
cf. L. Cohn, RE V 1005 (Dionysodoros no. 18).
::,_.>2.
:3:
5.
155:
=
.»~.
‘:9-:
I.3. Didymus and his Sources 81
Z-1%?
5. 1?}:
-5..
known only from Sada 6 1173, lived a century after Didymus, and con-
=...§_
-‘.;»:;
troversy against Callistratus was unlikely to be written at so late a date. A
.1,
K
more significant figure is Dionysius of Sidon, who is frequently adduced,
:."k;
153.":
.;-,6; not only by Didymus but also by Aristonicus (at M 36c), Apollonius
iii?
;-/~.
..,..
Ea; Sophista, Apollonius Dyscolus, and Herodian. He often disagreed with
="§:'5
Aristarchus over readings. Like his Thracian namesake, he gave what was
={iiit
considered useful testimony on Aristarchus’ atheteses and changes of mind
..,...
.,.-.
;.-§:;
(T 365-s, K 329)!“
WE:
Iii;
E:-ire 1.
Aristarchus’ interpretations. His readings are mentioned by Didymus also
i
E at B 127a, 192lZ)i, ® 1O3b1, K 41, 124211, 548, and E 316.
i
:-.i.
. .,..
More loyal pupils of Aristarchus were Parmeniscus and Chaeris.
E3? 2';
_.¢_.
Chaeris’ Homeric work consisted in, or at any rate included, a book of
Atopfiwtucoi (sch r| 80).“7 He was used by Tryphon and Herodian as well
>Z¥*~'=t @~I-'1§IT 'I"i£_ as by Didymus, who refers to him at B 865, Z 4b, 71a‘. In the last-named
.;-3;. . . -.: -. ._
..,r.
3521 2‘
21- E‘.
passage he appears to have commented on an observation by Athenocles.
V‘
,
I,
He is also cited in the papyrus commentary P. Oxy. 4452 (fr. 1. 24) at T
5.4;;
.- ,..
:25-l»;:
-. . I. .
357. Parmeniscus wrote against Crates (év Ttiit 01' Trpog Kpoirntot, G) 513a‘),
;-N:
raj,-
and Didymus cites from this work a statement of what Aristarchus read
$2._j ‘E2:5. --.
in that line. He refers to Parmeniscus again for a similar purpose at I 197b.
qr.
\:_-vii. I.
-'1/--.
._?_ Other mentions in the scholia, however, are not Didymean.“8
5.-1,.
We come next to two more Ptolemies: Ptolemy the son of Oroandes,
.147. - . . .
*t1§;;',f.'.»»»
“=55; : ;
5'3!- ._
EI.(gI _'
mf-
EEi¢jEI:;:;:-.'-'
<5g:;:- ; I
r,-
=:::§'E' : _- .' : '1
:
-.
"4 He is also cited at A 423-4, 607a, B 192b‘, F 35d, E 746a, H 5c, E 40b, P 155a, <1) 218
mi Q? I
(h94, V 98.1 E.), \P 587b. On him cf. La Roche (1866), 71; L. Cohn, RE V 983i. (Dionysios no.
135).
.v-1;,-,-. .
115 Sada 5 430; La Roche (1866), 77f.; T. Staesche, De Demetrio Ixione gmmmatico (Diss.
=.;, .
exr. ... Halle 1883); L. Cohn, RE IV 2845-7 (Demetrios no. 101).
\j-fr-I :
, :
-.
‘I6 Cf. the similarly titled works of Callistratus (above) and Zenodotus of Mallos (below),
and Pius’ duro7toyiou rrpoc; tort; dfietrjoetg.
5:1?
:-.x\
“7 La Roche (1866), 81f.; L. Cohn, RE III 2031.
1&
.~
"8 E 638b (Nicanor), A 424d (exegetic), E 249b (Herodian), E 100 (D), 514—15a (Nicanor).
See further M. Breithaupt, De Prtrmenisco gmmmatico (Leipzig—Berlin 1915); C. Wendel, RE
'i xv111(2) 1570-2.
Z
,.~
5/’
82 1.3. Didymus and his Sources
"9 K 292b (Herodian); E 229(?) (exeg.); Cb 1? (h94), 163 (/0), 356b (exeg.), <1) 492b? (G). On
Pindarion cf. La Roche (1866), 72; A. Dihle, RE Supp. IX 1306 (Ptolemaios no. 79a); Montanari
(1995), 41-ss.
'2° As suspected by A. Dihle, RE XXIII 1863 (Ptolemaios no. 79). Cf. also La Roche (1866),
72; M. Boege, De Ptolemaeo Ascalonita (Diss. Halle 1882).
*2‘ Ap. Dysc. Synt. 227. 18, quoted in Erbse’s apparatus ad loc.
‘Z2 H 272 (}Ié7teut<or; 6%; dfietei), 807b, Q 476b; h13O on T 350; h94 on <11 155, 195, 290.
‘Z3 Cf. Theodoridis (1976), 4f.
E:=l=:
I
2.51%
I-FE:
:1-.5-i
.,.;.
F
1.3. Didymus and his Sources 83
IE
11:15.
the reading but because he justified it morphologically, perhaps in his
'2:'-:=.-.
'-Ii‘;
"Ii; work rrsp‘r'rt6v eic; jfi Anydvtwv pnudtwv.
:1.=-i
-.;-_.;.i .
=='->25
‘IF £1
1-1..
the second is once mentioned by Herodian for a variant division of
-5:: El
..__,,_,_.
3:11;-
E;---<.-
;_-~;.
.---3,.
words?“
2.
:_=_:_=§;-_
Two more who figure only in the exegetical scholia are Pius and
i i?
:;.;{;
Autochthon. Their date is unknown, and either may well have lived after
;=. ,,
1:"-ta
F;
Didymus. Pius wrote a work defending various passages against Aristar-
g;
,1E,
chus’ atheteses.“
The well-fed Geneva scholia on <1) cite readings of Nicias (446c) and
.;.;..€;
:-_<;.
Arideikes (4460 and 474a). The latter is mentioned also in Et. Gard. 222.
55:4 = 13 for an etymology, but is otherwise unknown. Nicias is often cited by
tar
;-H. Herodian. He may be identical with the Coan grammarian Curtius Nicias,
fig‘. _ ._.
5??!
a contemporary of Cicero.‘26
;';';=, g: .. . .. .
. .,
5_=.=_'i _:
,<.
Didymus at work
Mp. ._
::'.f
r
2321;.
represent sources to which he may have had direct access; at any rate we
22211, gi
1 *24 Zenodotus: N 730, O 262, ‘P 79b; also in the papyrus commentary on 0, P. Oxy. 3710.
r
Cf. K. Nickau, RE Xa. 45-7 (Zenodotos no. 4). Hermias: H 207a.
‘*5 Sch. M 175-81b (dmoaoyoéuevog stpoc; ttitr; dflstfiosu; ’Ap1ot0'ip)(ou). The other passages
where he is cited are E 638d, Z 234bi, A 100b, 227c, <3) 55-9b, 147, 293. Cf. D. Strout-R. French,
'I'f__i
1:_:_:_.33
RE XX 1891f. (Pius no. 2). Autochthon is cited at A 132a/133c, N 407c, E 31-2, O 627b, T 270,
-;-5;;
ii
X 3. The mention of him at I 132a2 is evidently erroneous, cf. 132a1 and 274.
;;';&>:€
.<. 12° For Nicias the Homerist cf. La Roche (1866), 109f.; C. Wendel, RE XVII 337 (Nikias
555$‘
"l
':'::(i;._ .. ..
. -
no. 29). For Curtius Nicias see F. Miinzer, RE IV 1868f.; the testimonia are collected in H.
at
:
Funaioli, Gram-maticae Romanae Fmgmenta (Stuttgart 1969), 382f.
==:-gr
.-<.<..
.~.=A.a;f1<._@
...<t- -. ..
}5:I;I.-;-:-.-.-:-:
84 1.3. Didymus and his Sources
rion three each; and two each of Callistratus, Ptolemy Epithetes, and
Dionysius Thrax. In total he seems to have had within reach up to six
é1<56oe1g of the Xorpiéotepou class, apart from the non-individualized
onuofifieug; the commentaries of Aristarchus, perhaps in more than one
school recension; and getting on for thirty other scholarly works, some of
them in many volumes.
Theagenes
‘ Antimachus
I
Phiiitas
Zen0tlt:i{
. fihod. Argolica
Eratosthenes Rhianus
Sosi 9° nes
Ar. Byz. Philemon
Iloiuionxor; Massafioiica
, Cypria Chia
5 Crellca ?S1'nop1ca Alh e nod e s
Crates Ptoi. Eptth. ARiSTAHCt—lJ§ Callistratus
Of course he did not have all of these on his table at once. Even for a
modern scholar, whose books have the compact codex format and for the
most part robust bindings, it is difficult to work with more than three or
four of them open simultaneously. In Didymus’ time each title was repre-
sented by a papyrus roll or a series of rolls, which took up desk space
when opened, could not be leafed through like a modern book, and could
not so conveniently be piled one on top of another.
How then are we to picture Didymus at work? As he was working
through the Iliad from A to Q, he must have had at least one text open
before him. We have seen (pp. 51f., 55) that he worked with a base text
which did not agree with thatbf Aristarchus and more resembled the later
vulgate.'27 Since his objective was to reconstruct Aristarchus’ 616p6wo1g,
he must surely also have had open before him at least one of the two exem-
plars identified as containing it. It may have had variants (Aristarchean or
other) already entered in its margins. In discussing the problem of the two
,i%.
‘Z7 Add that at A 527 he refers to ‘the current’ reading, F] ¢>ep0pévn, which differs from both
of Aristarchus’ texts but corresponds to what most of the medieval manuscripts have.
I.3. Didymus and his Sources 85
/
,.. i
4. The Papyri
The purpose of this chapter is not to assess the contribution of the papyri‘
to our knowledge of the tradition, but to perform the more basic public ser-
vice of providing a more or less up-to-date list of those that have been
published hitherto and of the unpublished ones cited in my edition. I say
‘more or less’ up to date, because even if none has escaped my notice at the
moment of writing, more may have been published by the time this book
appears. Homeric papyri are far and away the most numerous of all Greek
literary papyri, and they will continue to accumulate. There is no end to
them. When Arthur Ludwich produced his great edition of the Iliad almost
a century ago, he was able to cite 33 papyri. In the Oxford Classical Text of
D. B. Monro and T. W. Allen (third edition, 1920) 103 are enumerated, a
total which Allen in his 1931 edition raised to 122. By 1948 Paul Collart
could list 372;2 in the second edition of Pack’s handbook3 there are 464; by
the summer of 1990, according to Dana Sutton’s survey,‘l the number had
reached 703; and for my own edition I have been able to make use of 1543.
In drawing up his catalogue Collart took Allen’s earlier list as his start-
ing-point: it supplies his first 103 items. Collart’s numeration has itself
been adopted and extended by subsequent researchers. H. Mette in 1955
and 1966 registered more recently published papyri as numbers 373—513;5
then in two further surveys, instead of going on past 513, he made a series
of interpolations in the existing list, which accounts for the awkward pro-
liferation of numbers such as 501.1, 501a, 501a1, 501b." Sutton continued
the list from 514 to 665. Thus we can say that a standard numeration has
evolved.7 This is what I have followed, extending it further to more than
double its previous compass.
In critical apparatuses papyri have usually appeared as Q1356 or Q3247 or
whatever, and Sutton lists them as p1, p2, etc. With the plethora now at
fli
‘ I follow the convention of using the term ‘papyri’ to include ancient manuscripts written
on other materials than papyrus: parchment rolls and codices, wooden tablets, ostraca, etc.
2 In P. Mazon, Introduction ii l’Iliade (Paris 1948), 39-57.
3 R. A. Pack, The Greek and Latin Literary Texts from Greco-Romrm Egypt (Ann Arbor
1965).
4 D. F. Sutton, Homer in the Papyri (Macintosh database on two diskettes, Scholar’s Press
1992). An updated online version is now accessible at http://eee.uci.edu/~papyri/
5 Rev. Pb. 329 (1955), 193-205; Lustru-m 11 (1966), 33-9.
“ Lustmm 15 (1970), 99-103; 19 (1976), 5-9. Between 450 and 513 his list actually contains
115 items. Other irregularities in the numbering, such as gaps in the sequence, have resulted
from joins made between pieces that had been allocated different numbers or from reclassifica-
tion of items under a different heading.
7 Only van Thiel has failed to avail himself of it. He cites papyri by their Pack numbers.
I.4. The Papyri 87
hand - anything up to ten papyri may have to be cited for a given reading
- this has become too cluttersome. In my apparatus I have therefore omitt-
ed the Q3 and used the bare numerals as sigla, just as is done for the mass
of manuscripts in modern editions of the Septuagint and New Testament.
Although numerals also fulfil other roles in the apparatus, I think no ambi-
guity arises, as the syntax makes it obvious what is a manuscript siglum
and what is not.
The following list, in so far as it is coextensive with Sutton’s - that is, as
far as 665 — is adapted from his, with countless errors and confusions cor-
rected and additional bibliographical references added where appropriate.
Nos. 666-704a comprise further papyri published up to 1998. Nos.
705-1544 are unpublished papyri from Oxyrhynchus in the Papyrology
Room at the Ashmolean Museum, Oxford. The world will share my grati-
tude to the Egypt Exploration Society for allowing me to use them for my
edition. In accordance with the Society’s wishes, I did not give any details of
these papyri in the edition. I now have the Society’s permission to divulge
the inventory numbers and summary details of those which I had occasion
to cite in my apparatus.8 The remainder, which account for the gaps in the
numbering of the list below, comprise nineteen fragments that are recorded
as existing but could not be located, and 182 that I examined but did not cite
because, so far as I could observe, they contain (apart from insignificant
errors) only'text that is agreed by all sources. As no. 1545 I have added a
mural inscription from Rome which, though published in 1938, had not
previously been listed, and as nos. 1546-68 a further group of papyri recent-
ly edited by M. Manfredi and colleagues in Florence.9
8 The datings given for these papyri are those assigned when they were first identified. Where
no dating is offered, it is because none has so far been assigned.
9 Manfredi (2000). The volume, evidently prepared in ignorance of my edition, was present-
ed to participants at the congress ‘Omero 3000 anni dopo’ held at Genoa in ]uly 2000. As this
was too late even for the Addenda to my second volume, I give here the readings of the new
papyri, together with a couple of other modifications to my apparatus:
A 33 eeaioley 1547 34 qmeoav 1547 37 11811 1547 40 n 1547 42 reilq‘ @3119 1547 4s veogiv 1549
244 eteiooicg 1550 510 TllO(1)O_ll 1552 B 12 c2101 1553 15 Tpwelooi 6:5 1_<[n6e 1553 849 Auulijugvlog
1554 F 162 lilCev 1555 165 oixéiiwv 1555 168 p61C(_)V§-JQ 1555 170 yap 1555 174 ylvtorouc; 1555
354 Eleivofioldov 1556 355 n 1556 417 oknon 1557 456 new 1557 458 optic; ‘6 1557 E 364 et 366
om. 1558 477 lévt_;_i|,1ev 1559 481 1-’ 1559 486 tripeooillvll 1559 487-9 hab. 1559 4s9 iiuqv 1559
Z 16 ]y_e[t0'rl 1560 inter 16 et 20 unum v. tantum habuit 1560, fort. ornissis 17 Buuov — 19 c’>iuc|xo
21 I_II'|§QlU(_}lV 1560 28 non teulxe r;;[o]i_17_1[0i] Q(l7l' topoov 1560 187 6 cap Q21 1561 H 447 ev'ii|1eh 1563
452 [to] vel lot] 1563 453 7l'O7\lGOO(}.l8V 1563 455 eeurec 1563 458 ooov T 1563 466 fie 1563 466
e7\eoE_i{e 1563 467 7T(XpE‘,O'l'Q(00(V 1563 468 vrlpoi-:11 Kev 1563 471 )§_1?_\_1oi 1563 I 49 e1]?\q7\ou6uev 1564
52 uetelclxpvecv 1564 53 av 1564 54 errhev 1564 55 A)(O(lllOl 1564 56 pufiwv 1564 413 eorou 1565
423 qnelivltp 1565 K 15 rro71>\01]g ex 331 18 éniil 331 292 pro ‘Ptol’ lege ‘Pindario’ 556 y‘e0[e7\tov
331 570 8Vl?Tpl5|1VI] 331 572 0(1T0lVlCOV'I‘lO 331 H 150 tour; 6 E1 1566 247 motto 1567 T 166 ilovti
1568 \I' 688 otpnvfig 6%. Orio 146.13 (monuit P. A. Hansen).
1.4. The Papyri
CATALOGUE OF PAPYRI
Ambros. gr. F 205 inf. (1019) (parchment codex, v-vi): A 1-13, 68-77, 178-90,
283-9, 320-32, 345-69, 436-52, 509-25, 606-11, B 117-25, 311-23, 387-99, 827-40,
A 103-16, E 1-25, 407-19, 422-32, 634-57, 765-81, 848-66, Z 207--25, 282--301, 407,
418-32, I-I 445-58, (:9 25-6, 233-53, 510-21, I 1-25, K 168-75, 204-18, 557-68, A
552-71, 597-611, M 1-25, 139-50, 226-39, 425-37, N 393-408, E 414414, o 377-98,
571-s5, n 209-25, 75s-72, 783-4, P 4-25, as-105, <1» 202»-13, 393-409, x 4-25, 91
509-25, 765-78, Q 244-53, 329-42, with illustrations. Ed. A. Mai, Iliadis fragmen-
ta antiquissima cum picturis item scholia antique: ad Odysseam edente Angelo Maia
Ambrosiani Collegii doctore (Mediolani 1818); photographically reproduced by_M.
Ceriani and A. Ratti, Homeri [liadis pictae fragmenta Amiarosiama pbototypice
editae (Mediolani 1905); llias Amlarosiarm. Cod. F.205.P. inf Bibliot/aecae Ambro-
sianae Mediolanensis (Fontes Amlarosiani, xxviii) (Bern-Oltun 1953; limited edi-
tion). Many accents were added in the early Middle Ages.
P. I-Iawara (Bodl. Gr. class a.1(P); Packz 616; Hawara, ii-iii): A 506-7, B 1-6,
45-9, 111-15, 155-7, 200-5, 223-8, 245-53, 289-92, 323-39, 345-54, 368-82,
391-404, 411-22, 432-46, 454-70, 472-91, 493-511, 515-36, 538-60, 562-98,
601-21, 624-731, 735--877 with scholia (= pap. I Erbse). Readings reported by A.
H. Sayce in Sir Flinders Petrie, Hawara, Bia/omit, and Arsinoe (London 1889),
24-8; partially (B 854-64) reedited by Seider (1968-70), ii. 77f.; B 666-709 are
covered by the plate in Erbse (1969-88) I, B 755-75 by that in Turner (1971), pl.
13. A new edition is expected from R. P. Salomons. Collated by me.
British Library Pap. 126 (P. Lit. Lond. 5; Packz 634; Ma’abdeh, papyrus codex,
iii): B 101-494, 1' 1-461, A 1-40. Readings reported by Kenyon (1891), 81-92; B
458-93 reproduced in Turner (1971), pl. 14. Collated by me.
British Library Pap. 136 (P. Lit. Lond. 11; Pack’ 697; iii): F 317-37, 341-76, A
1-28, 55-70, 74-80, 84-96, 111-50, 159-92, 198-201, 208-45, 256-94, 303-544, E
1 (reclamans). Readings reported by Kenyon (1891), 93-7. Collated by me.
P. Genav. inv. 90 (Packz 890; second half of iii BC): A 788-834, 837-M 9. Ed. ].
Nicole, Rev. P/9. 18 (1884), 101 ff.; S. West (1967), 107-17.
British Library Pap. 127b (P. Lit. Lond. 12; Packz 762; iii-iv): E 721-5, 731-43,
815-21, 845-50. Collation in Kenyon (1891), 98 f. Allen wrongly combined this
papyrus with the items designated 6b (see at 87) and 6c (see at 86), as the single
papyrus 6.
P. Grenf. 2. 2 + P. Hib. 21 + P. Heid. inv. 1261 (Packz 819; Hibeh, iii BC): G)
17--22, 24»-32, as-50, 53473, 18044, 187-206a (2), 2155-221, 249--53, 2554-us.
s. West (1967), 74490.
P. Petr. 1. 3(4) (British Library Pap. 486d = P. Lit. Lond. 21; Pack’ 879; Guréb,
111 ac); A 502-~17, 518437. s. West (1957), 103-7.
British Library Add. MS. 17210 (parchment codex palirnpsest, vi): M 273--471, N
133-265, 333-9s, 465-530, 553-72s, 797-837, e 1-20, 156-419, o 155-223,
356-421, 491-557, II 199--264, 331-97, 664-731, 798-862, E 93-358, 426-92, T
136-268, 335-424, T 1-172, 306-503, (I3 1-397, 465-611, X 1-113, 181-378, ‘P
57-323, 457--589, 656-788, 856-~97, Q 1-20, 285-483. W. Cureton, Fragments Of
the Iliad of Homer from a Syriac Palimpsest (London 1851); M. ]. Apthorp, ZPE
110 (1996), 103 ff.
L4. The Papyri 89
P. Bibl. Brit. inv. 732 (P. Lit. Lond. 22; Packz 899; i): N 2»-12, 28-34, as--56, 73-87,
149-425, 430-36 456-653, 657-74, 740-7, 769-75, E 120-277, Z78-93, 332-54,
358-476, 480-522. A. S. Hunt, joumal ofP/oilology 26 (1899), 25--59.
British Library Pap. 107 (P. Lit. Lond. 25; Packz 953; i-—ii): Z 1-218, 311-617. Ed.
E. M. Thompson and G. F. Warner, Catalogue of Ancient Manuscripts in the Bri-
tish Museum (London 1881), I 1-6.
P. Grenf. 2. 4 +- P. Hib. 1. 22 + P. Heid. 1262-66 (Packz 979; Hibeh, iii BC): CI)
302-12, 370-414, 421-30, 607-11, X 27-38, 48-55, 73?, 77?, 81-4, 96-113, 125-43,
151-5, 197-8, 226-7, 247-56, 25919-62, 291, 316-160, 326-8, 340--3, 3924-401,
419-26, 440-8, 458-60, 462-5, 513-‘P 1, 85-96, 115-41, 153-92, 195-229, 236-52,
265-8, 276-81. G. A. Gerhard, Ptolemdische Homerfmgmente (Berlin 1911),
20 ff.; S. West (1967), 136-91.
British Library Pap. 128 (P. Lit. Lond. 27; Paekz 998; i BC): 1P 1-79, 402-633,
638-814, 823-97, Q 1--83, 100-58, 164-243,‘ 248-74, 276-82, 337-41, 344-51,
382-7, 402-79, 490-520, 536-48, 559-77, 596-611, 631-657, 672-728, 737-44,
754-59. Kenyon (1891), 100“-8; id., joum. P/oil. 21 (1893), 296-343; pap. XIII
Erbse. Photos of parts in Lameere (1960).
British Library Pap. 114 (P. Lit. Lond. 28; Packz 1013; Elephantine?, ii): Q
127-804. C. G. L(ewis), P/oilological Museum 1 (Cambridge 1832), 177-87;
recollated by Leaf for his edition and by me.
P. Oxy. 20 (British Library Pap. 742, P. Lit. Lond. 7; Pack‘? 668; ii): B 730-6,
745-54, 770-809, 815-28.
P. Oxy. 223 (BOCll. Gr. class. a.8(P); Packz 733; iii): E 1-278, 284-303, 329-51,
353-74, 397-406, 420--1, 425--42, 544--48, 701-5.
P. Chic. 7 (inv. 92; Packz 818; Karanis, ii): G) 1-29, 35-68. E. ]. Goodspeed, A]/1
2 (1898), 347-56.
P. Fay. 4 (British Library Pap. 816 = P. Lit. Lond. 17; Packz 830; Bacchias, ii): (5)
332-6, 362-9.
P. Fay. 5 (Packz 606; Theadelphia, ii): A 404-47. W. Criinert, APF 2 (1903), 340 f.
P. Flor. 2. 109 + P. Grenf. 1. 2 (Bodl. Gr. class. d 20(P); Pack‘! 821 + 822; Fayfim,
i—-ii): G) 62-75, 96--116. The Florence papyrus was formerly listed as no. 55; the
two were joined and reedited by C. Gallazzi, ZPE 71 (1988), 60-6. Allen’s 8320
is no. 218.
P. Oxy. 445 (British Library Pap. 1190 = P. Lit. Lond. 14; Packz 778; ii-iii): Z
128-7, 134-7, 148, 173--99, 445-529 with scholia (= pap. IV Erbse).
P. Oxy. 446 (Packz 901; ii-iii): N 58»-99.
P. Oxy. 447 (Packz 1001; ii): ‘I’ 81-91.
P. Oxy. 544 (P. Cornell Univ. VIII; Packz 715; iii): A 182—98. Reedited as P. Cor-
nell Univ. VIII by Priest (1975), 64-6
P. Oxy. 550 (British Library Pap. 1191 = P. Lit. Lond. 250; Packz 880; ii): A
505-16, 521-47, 555-67, 572-602.
P. Oxy. 551 (Paclrz 916; ii): E 227--53, 256-83. Image accessible at
http://Wwwprinceton.edu/papyrus/inclex.html
P. Oxy. 558 (Packz 989; iii): X 115--34, 143-60.
P. Oxy. 560 (PaCl{2 1010; iii): ‘I1 775-85, 834-47.
P. Oxy. ass (Packz 630; i BC-i AD): B 5o~s.
90 I.4. The Papyri
P. Morgan (Packz 870; papyrus codex, iv): A 86-96, 121-848, M 1-TI 262, 281-9,
294-9, 316-25, 329-427, 434»-66, 471-82, 486-99. ‘The Morgan Homer’, des~
cribed by U. von Wilamowitz-Moellendorff and G. Plaumann, SPAW 1912,
1198-1219; cf. M. ]. Apthorp, ZPE 81 (1990), 2-4. Collated from a microfilm
kindly procured by Dirk Obbink.
P. Oxy. 538 (Packz 595; papyrus codex, ii): A 273-297, 318-42.
P. Oxy. 545 (Pacltz 731; ii—iii): A 478-90. Reedited as P. Cornell Univ. IX by
Priest (1975), 66-8.
P. Oxy. 546 (Packz 808; ii): H 237-44, 264-73. Reedited as P. Cornell Univ. X by
Priest (1975), 78-81.
P. Oxy. 548 (Packz 841; papyrus codex, iii): I 235--301. Reedited as P. Cornell
Univ. XII by Priest (1975), 86-93.
P. Oxy. 549 (Packz 869; iii): A 39--52. Reedited as P. Cornell Univ. XIII by Priest
(1975), 9648.
P. Oxy. 553 (Packz 964; papyrus codex, iii): T 97-117, 132-51.
P. Oxy. 687 (Packz 691; late i BC): F 1-81, 184-9, 207-16.
P. Oxy. 750 (Packz 631; iii): B 57-73.
P. Oxy. 752 (Packz 713; iii): A 87-96.
P. Oxy. 754 (Packz 732; i): A 532-39.
P. Oxy. 756 (P. Yale 1. 5 = inv. 67; Packz 744; papyrus codex, iii): E 324-3412,
379-90.
P. Oxy. 758 (P. Yale 1. 7 = inv. 69; Packz 757; ii-iii): E 583-96.
P. Oxy. 760 (Packz 761; i): E 715-18, 720-9.
P. Oxy. 765 (Harvard Semitic Museum inv. 4373; Packz 845; ii-iii): I 320-33. Re-
edited by B. ]. I-Iennessey, BASP 7 (1970), 72-5.
"3
Oxy. 768 (Packz 889; iii): A 736-64.
7')
Oxy. 770 (Packz 909; ii): N 372-7, 405-13.
7)
Oxy. 772 (PaCl{2 947; ii-iii): P 353-73.
P. Tebt. 427 (Packz 698; Tebtunis, ii-iii): F 338-63, 386-97.
BKT 5. 1. 4 (inv. 7807; Packz 842; iii-iv): I 277-88, 299-312. Ed. W. Miiller, APP
41 (1995), 101.
BKT 5. 1. 4 (inv. 8440; Packz 735; Eayfim, i): E 69-81, 84-93, 99-103. Ed.
G. Poethke, APF 40 (1994), 4f.
BKT 5. 1. 5 (inv. 9968; Packz 924; papyrus codex, iii-iv): O 383-90, 421-30. Re-
edited by Seider (1968-70), ii. 79f.; G. Poethke, APF 40 (1994), 8f.
P. Aberd. 109 (Packz 975; Fayfim?, iii): (D 1-26. Ed. E. O. Winstedt, CQ 1 (1907),
260.
P. Aberd. 111 (Packz 993; ii): X 226-33, 265-72, 300-5. Partially ed. by
E. O. Winstedt, CQ 1 (1907), 261.
P. Ryi. 1. 51 (PaCl{2 -1016; i BC)! Q 336-43, 366-74, 377-401.
P. Par. 3 + P. Vat. (Packz 900; E1ephantine?, i BC-i AD): N 26-47, 107-11, 143-73.
The Vatican papyrus, formerly listed as no. 220, was edited by G. Lumbroso,
Rendic. R. Acc. Lincei Ser. 5. 2 (1893), 831, reprinted by O. Marucchi, ll Museo
Egizio Vaticano (Rome 1899), 272.
P. Par. 3 (bis) + British Library Pap. 127a (P. Lit. Lond. 24; Packz 952 + 959;
Thebes, ii): E 1-23, 28-33, 76-»-92, 95-136, 141--5, 152-79, 227-31, 273-6, 278-89,
92 L4. The Papyri
116 BKT 5. 1. 3 (P. Berol. inv. 6869 + 7492-5 + 21158) + P. Aberd. 134 (Packz 572
"TE: -1- 2787; i-ii): A 54-64, 71-104, 114-123, 131-64, 338-43, 412-32, 456-65,
.;_
3:1;:? 494-534, 537-90, 602-9. Pieces formerly listed as nos. 116, 117, 129, 133, 135,
5=1E§ joined and reedited by H. Maehler, W. Miiller, and G. Poethlte, APF 24/25
(1976), 6-12; Ioannidou (1996).
iii
1..
[117: see no. 116.]
11s P. Oxy. 536 (P1618 579; ii--iii): A 127--47.
22111
==_1(.f.
119 British Library Pap. 272 (P. Lit. Lond. 2; Packi 580; ii-iii): A 129-50.
:-1:;
--1
ll
E:
51:3 41EI? 120 P. Oxy. 749 (PaC1{2 582; ii): A 160--76.
25;»
121 Mon. Epip/9.4m. 2. 613 (Packz 586; limestone inscription from Thebes, school
:--1 F.‘
:-1 >:
exercise, vi-vii): A 201 (and passim).
1.1
55:II 7.1111.5
-.-:11.
122 P. Oxy. 537 (Packz 589; iii): A 215-20, 250-66.
123 P.S.I. 745 (Pacltz 590; Oxyrhynchus, ii): A 215-~52, 276-312, 337-41, 345-6, 348,
1 360--5.
1?
124 P.S.I. 1083 (Pacltz 593; Oxyrhynchus, iii): A 223-45.
125 ~ P. Fay. 141 (Packz 596; Karanis, i-ii): A 273-362.
iiiii
=.=3z=
5;.-,5 126 Bodl. Gr. class. e.58(P) (Packz 598; i-ii): A 298-333. F. G. Kenyon, The Palaeo-
...,];
gmp/ay of Greek Papyri (Oxford 1899), 139.
<
1:
£5515 127 P. Tebt. 425 (Packz 600; Tebtunis, late ii): A 311-27.
':'EI'
1:111
k
128 BKT 5. 1. 5 (inv. 10574; Packz 607; papyrus codex, iv): A 406-19. Ed. G. Poeth~
=2 15
; _. ,
t
'=;:)p,>-
16, APF 40 (1994), 2.
§:'=f1E
2:211
E55;_1;'.
[129: see no. 116.]
I:-‘F 1
5:5; :-
r. 130 BKT 5. 1. 4 (inv. 9584; PaCl(2 609; i): A 449-61. Ed. W. Muller, APP 39 (1993), 6.
;5
;='.=. E‘
'E'i:I]]?: 131 British Library GR 1906.10-20.2 (Packz 611; school exercise on wooden tablet,
.'.';_,E; v): A 46s-73.1361. c. Raffael1a,C/oron. d’Eg. 6s (1993), 14511.
s.=;j1
E351
zE==‘ t
132 P. Ryl. 1. 44 (Pack: 612; i BC): A 471-so, 495-506.
,5
)1
1
1
1
1
94 L4. The Papyri
Iii?
;s-_-
:$:
51;‘
I.4. The Papyri 95
zg.
i?
i.45
.§.;.
173 P. Aberd. 106 (Packz 714; Fayfirn?, i): A 109-~20, 163-83, 188-211. Partially edi-
..3>..»_-._=
16d by E. 0. Winstedt, CQ 1 (1907), 2581.
:5:
174 P. Oxy. 1386 (P6616 718; iii): A 257-72.
1
175 P. Ryl. 1. 46 (Packz 721; Oxyrhynchus, i): A 357-64.
1 176 P. Oxy. 947 (P6618 725; iii): A 443-52.
177 P. Iand. 1. 1 (inv. 93; Packz 727; i BC): A 454-66, 483-7. Reedited by P. A. Kuhl-
mann, Die G-iessener literarisc/Jen Papyri und die CaracaZla~Eriasse (Giessen
5
1994), 30-3.
178 P. Tebt. 428 (Packz 734; Tebtunis, ii-iii): E 52-5. Image accessible at
5;:
: s=
5 1':
http://sunsite.berkeley.edu/APIS
. gi
179 P. Oxy. 755 (Packz 737; iii): E 130-73.
180 P. Oxy. 1387 (Packz 739; ii): E 206-24.
it
P. Berol. inv. 11636 (Packz 742; school exercise on wooden tablet, iii-iv): E
_. E.
ii; 181
._.;-.
265-317 with a glossary to 287--317 (= h53). G. Plaumann, /flmtiic/oe Beric/ate aus
I-1:;
den /eo'nz'glz'c/Jen Kunstsammlungen 1913, 220; L. M. Raffaelli, APF 36 (1990),
11:5
§:
5-12.
:1:
182 P. Amh. 2. 22 (Packz 751; i-ii): E 481-95.
1'51 1;
=-.11:
.'-'1 I1.
1
183 P. Rain. inv. 12516 + 267322.11) + 26733-4 + 26736 -1- 26761 + 35721 -1- 39833
;.<=:
41 %; (Packz 755; Fayfim or Hermupolis, ii-iii): E 541-6, 554-62, 565-7, 570-80,
=1'?j.'.. . -
.51
., . 597-610, 617-20, 701-4, 729-38, 766-76, 807-53, 861-70, 890-6. 12516-26761
‘r
were formerly collected under no. 295; 35721 is edited by P. Sijpesteijn, ZPE
40 (1980), 91. <31. Gerstinger (1926), 881.
55.1:
; =§;'
184 P. Clxy. 757 (P. Yale 1. 6 = inv. 68; Packz 756; i): E 578-86.
5.‘-T11? I
.
185 P.S.I. 748 + 749 (Packz 763 + 777; parchment codex, iv?): E 724-35, 744-55, Z
99-102, 119-22. P.S.I. 749 was formerly no. 186; the two were joined by C. Gal—
lazzi, ZPE 71 (1988), 56-60.
5.1 if
1 [186; see no. 185.]
=2‘. 1-
§_:j 1-"
187 P. Oxy. 1388 (Packz 779; i BC): Z 133-50, 156-60.
188 P. Oxy. 761 (Packz 780; school exercise, late i BC): Z 141-9.
:___,: _.
1‘¥1§:-- 189 P. Osl. 2. 7 (Packz 783; Fayfimi’, iii): Z 196-224, 236-76. First edited by G. Rud-
5551‘
1
96 I.4. The Papyri
I E 8:5
'.-;:a-'
-.--.4;
P.S.I. 752 (Packz 946; parchment codex, iv-v): P 334-46, 368-81.
P. Oxy. 1817 (Packz 948; papyrus codex, v-vi; perhaps the sarne manuscript as
254): P 379-84, 418-24, Z 412-14, 455--6, 564-81, 603-17. Reedited by Lameere
(1960), 175-90.
P. Oxy. 685 (Bodl. Gr. class. f.75(P); Packz 950; ii): P 725-32 with scholia (= pap.
_.,.
-;:.4.-
XI Erbse).
:=';;.<';
'.‘='.§:
_:::-9;
P.S.I. 14 (Packz 954; parchment codex, Oxyrhynchus, iv-v): Z 76-99, 112-35.
.;=-.-y.
=:;"e:_
:::_4:
:-I-*5;
iii-1.?
P. Ryl. 1. 50 (Packz 957; papyrus codex, iii): E 395-401, 428-34.
E111 ' P. Mich. 2 + 2755a + 3160 (Packz 565 + 958; ii-iii): E 23--45, 174-81, 209-16,
'=.._..;,._
219-31, 237-42, 274-81, 286-99, 306-43, 354-617. Partially edited by G. Win-
fig. .. . ter, TAPA 53 (1922), 128ff., and C. W. Shelmerdine, BASP 12 (1975), 19ff.;
joined and reedited by Priest (1975), 123-64. Parts were formerly listed as nos.
445 and 649. No. 648 is perhaps part of the same manuscript.
P. Rein. 1. 1 (P. Sorb. inv. 2010; Packz 963; iBC-i AD): T 41-51.
P. Oxy. 554 (P6618 966; ii): T 251-59.
“t
5‘EEEI-_§§:5; - .1 _- -_ : j
P. Oxy. 555 (Packz 969; iii): T 417-21.
;=j=---=1:
"=_1s-.-. -. = - - - -
E§:_-4,:
;-;;;;16
.-.-.-.<;_
P Fay. 160 (Packz 970; Bacchias, i-ii): Y 36-110.
:,,,-E;-K _ -
7)
. . _ .
Oxy. 556 (Packz 972; ii-iii): T 241-50.
"')
Oxy. 951 (Packz 974; papyrus codex, iv): T 425-37, 470-82.
'3
=f=fz:;E_§§.'- --.-. - - . Fay. 6 (P. Cair. GH 10764; Packz 976; early i): (I) 26-41.
-i1;‘ .1;.g
P. Aberd. 110 (Packz 978; Fayfim?, iii): (D 58-65. First edited by E. O. Winstedt,
1; ; ; » ,' :
CQ 1 (1907), 2601.
P. Oxy. 557 (Packz 981; iii): (I7 372-82.
P.S.I. 753 (Packz 983; Oxyrhynchus, iii): (D 511-27.
&_&::='1»".=
Eéi:-:-:-Ié":-i=-15;
BKT 5. 1. 3 (inv. 6794; Pack’ 984; parchment codex, iv-v): (D 547-76, 580-609,
;-_1,. X 390-421, 423-54.
1;,-:.'_'._. P. Amh. 2. 159 (Packz 985; parchment codex, iv): (D 608-11, X 30-7.
:-I-I=I-$2
-.55:._1.-__.-1‘,;
=;. ; ___i
:_<E;
P.S.I. 139 (Packz 986; Oxyrhynchus, ii-iii): X 1-17, 22-38.
P. Oxy. 559 (P6618 987; ii): X 1-18, 40-57.
.‘,
:_,_:::‘,_ ._ .
. .
.. .
P. Oxy. 1818 (Packz 988; papyrus codex, Oxyrhynchus, v-vi; perhaps the same
Iz 5 J I manuscript as 234): X 109-36, 153-77, 190-202, 216-43, 255-78, 283-314,
;_§_§_.' .,1_. .. 336-57, 376-97, ‘P 345-70, 383-406. Reedited by Lameere (1960), 148-74.
P. Yale 1. 12 (inv. 8, = P. Fay. 211; Packz 995; Euherneria, i-ii): X 254-90, 328,
350-4, 358-65.
1 British Library Pap. 1545f (P. Lit. Lond. 26; Pack‘? 996; ii-iii): X 449-74.
P. Univ. Lips. inv. P.253 (Packz 999; iii): ‘I1 22-49, 79-100, 427-~47. F. Blass,
1 Beric/ate der Sac/asisc/Jen Gesellschaft 69 (1904), 211.
.- .- .¢
::,;;._§.__ _
P.S.I. 140 (Packz 1000; papyrus codex, Oxyrhynchus, iii): ‘P 63-88, 93-7, 126-47,
=::_=.=,-,9. 152-6.
?§-§-§-§=‘§ f9,
,,, . P.S.I. 141 (P6612 1007; ()Xy1'l1y1'lCl1U.S, ii-iii): 81 485-91, 499-509.
P. Oxy. 952 (P. Yale 1. 14 (inv. A7); Packz 1012; iii): Q 74-90.
P. Oxy. 561 (Pack2 1015; iii-iv): Q 282, 286, 318-31.
é.%i%-is§§'.':': 2 3-I
Eirc;
6:5:-1:
.4
....»-
98 I.4. The Papyri
[284 = w2.]
285 P. Rein. 2. 66 (P. Sorbonne inv. 2144; Paclcz 684; ii-iii): F 33--43.
287 P. Rain. inv. 29285 (Packz 629; v-vi): B 41-7, 86-92.
288 P. Rain. inv. 19794 + 19791; Packz 8.53; school exercise, i-ii): K 41-57, 76--100,
103-12. H. Oellacher, Griec/aische Zitemrisc/9e Papyri H no. 1. 19791 was for-
merly listed as no. 367.
289 P. Rain. inv. 26728 (Packz 603; Soknopaiou NesosP, iii): A 343-60, 378-95. Cf.
Gerstinger (1926), 87.
290 P. Rain. inv. 30048 (Packz 613; Fayfim?, ii): A 479-84. Cf. Gerstinger (1926), 88.
291 P. Rain. inv. 26737 (Packz 635; Fayfim?, iii): B 104-112. Cf. Gerstinger (1926), 88.
292 P. Rain. inv. 26767 (Packz 669; Soknopaiou Nesos?, ii): B 738-41, 751, 753-4,
759, 765-93. Cf. Gerstinger (1926), 88.
293 P. Rain. inv. 26749 (Packz 672; Soknopaiou Nesos P, iii): B 754--64, 797-801. Cf.
Gerstinger (1926), 88.
294 P. Rain. inv. 26744 (Pack: 728; parchment codex, Fayfim?, vi): A 455-74,
476-505, 507-14. Cf. Gerstinger (1926), 88.
[295: see no. 183.]
296 P. Rain. inv. 26730 + 26745 (Packz 806; ii-iii): H 223-37, 266-348. Cf. Gerstin-
ger (1926), 89f.
297 P. Rain. inv. 26741 (Packz 827; ii-iii): (5) 216-32. Cf. Gerstinger (1926), 90.
298 P. Rain. inv. 19768 (Packz 832; Karanis, i): (5) 436-61. Cf. Gerstinger (1926), 90.
299 P. Rain. inv. 26729 (Packz 837; Soknopaiou Nes0s?, i-ii): I 152-61. Cf. Gerstin-
ger (1926), 90.
300 P. Rain. inv. 26753 (Packz 852; i BC): K 1-26. Cf. Gerstinger (1926), 90.
301 P. Rain. inv. 26751 (Pack: 877; parchment codex, Fayf1m?, iv-v): A 449-55,
479-85. Cf. Gerstinger (1926), 90f.
302 P. Rain. inv. 26763 (Pack2 886; Soknopaiou Nesos?, i): A 708-21. Cf. Gerstinger
(1926), 91.
303 P. Rain. inv. 26739 + 29359 (Packz 896; i): M 248-67. Ed. G. Bastianini, ZPE 51
(1983), 13-18; cf. Gerstinger (1926), 91.
304 P. Rain. inv. 26732 (Paclcz 898; Soknopaiou Nesos?, iii): M 300-13. Cf. Gerstin-
ger (1926), 91.
305 P. Rain. inv. 26752 (Packz 914; Hermupolis, ii-iii): N 762-74. Cf. Gerstinger
(1926), 91.
[3061 see no. 232.]
307 P. Rain. inv. 30049 (Paclcz 965; iii): T 121-33. Cf. Gerstinger (1926), 91.
308 P. Univ. Washington inv. 162 (Paclcz 911; papyrus codex, iv-v): N 512-27,
545--60. Ed. Z. M. Packman, BASP 10 (1973), 47-53.
309 P. Cair. inv. 56225 (Packz 556; school exercise, Oxyrhynchus, i-ii): A 1 (written
repeatedly). Ed. W. G. Waddell, Et. Pap. 1 (1932), 17.
310 P. Cair. inv. 49654 (Packz 949; Oxyrhynchus, iii): P 649-71, 681-3. C. C. Edgar,
Arm. Sew. 26 (1926), 203; W. G. Wacldell, Er. Pap. 1 (1932), 11.
311 P. Cair. inv. 47268 (Packz 1003; Oxyrhynchus, ii-iii): ‘P 135--91. C. C. Edgar,
Arm. Sew. 26 (1926), 203; W. o. Waddell, Er. Pap. 1 (1932), 11;.
312 P. Cair. inv. 45617 (Packz 1017; i-ii): Q 373-96. C. C. Edgar, Ann. Sew. 26
(1926), 203; W. G. Waddell, Er. Pap. 1 (1932), 12f.
100 I.4. The Papyri
2
351 P. Harr. 120 (inv. 176d; Pack’ 892; ii): M 1-3.
i 352 P. I-Iarr. 121 (inv. 176b; Pack‘? 893; ii): M 126-34. Perhaps from the same roll as
i
no. 351.
353 P. Harr. 122 (inv. 40a; Pack‘! 931; ii): H 60--84. C. Ferro, Amzlecta Pepyrologica
4 (1992), 37-9.
354 P. Ryl. 3. 539 (Packz 575; late ii BC): A 92-118, 244-61. Reedited by S. West
(1967), 28432. .
,.
1
355 P. Rein. 2. 70 (P. Sorbonne inv. 2104; Packz 888; papyrus codex, vi): A 734-53,
1'5
11
772-90.
356 P. Rain. inv. 26740 (Pack2 791; school exercise, Soknopaiou Nesos, ii): Z
1
1 373-410. Ed. P. ]. Sijpesteijn, C/6661666 d’Egypte 49 (1974), 309-13.
1
1
[357: see no. 104.]
35s P. Ryl. 3. 541 (P6018 652,1); B 400-72.
359 P. Ryl. 3. 542 + P. Mich. inv. 44 (Packz 750; iii): E 473-95. The Michigan piece
edited by K. McNamee, ZPE 46 (1982), 123f.; joined by O. Bouquiaux-Simon,
ZPE 51 (1983), 591.
360 P. Ryl. 3. 543 + P. Mich. inv. 919 (Packz 881; ii-iii): A 516-66, 573--98. The
Michigan piece first edited by N. E. Priest, ZPE 46 (1982), 81-5, and listed as
no. 617; the two joined and partially reedited by O. Bouquiaux—Simon, ZPE 74
(1988), 297--300.
361 P. Ryl. 3. 544 (Packz 1014; early ii): Q 169-76.
362 P. Leeds inv. 4 (Packz 618) + P. Brux. inv. E.7160 (Packz 617): A 522-46. The
1
Leeds piece first edited by W. G. Arnott, Proceedings of the Twelfth Interna-
tional Congress of Papyrology, ed. D. H. Samuel (Toronto 1970), 20f., and listed
5';-. . . as no. 457; the two joined and reedited by Lameere (1960), 55-64.
363 P. H6666. 3. 196 (inv. 776 + 778) + P. B166. inv. E.7344 (P6618 717) -1- 13.7190 +
P. Princ. 3. 110 (inv. AM 11227117.; Packz 723; i): A 199-219, 238-57, 260-74,
1.-
371-84. The first Brussels piece was reedited by Lameere (1960), 65-92, the
second (formerly listed as no. 569) was published by]. Bingen, Chronique
1
1
d’E’g3/pte 40 (1965), 3511. The Princeton piece was formerly listed as no. 397.
364 P. Brux. inv. E7161 (PaCl{2 897; ii-iii): M 265-92. Ed. Larneere (1960), 98-104.
1 [365 = h27; 366 = W23; 367 see no. 288; 368 see no. 232.]
E 369 P. Rain. 3. 3 (inv. 26469; Packz 971; papyrus codex, Fayiim, iv): T 205-15, 234-43.
E 370 P. Rain. 3. 4 (inv. 30485; Packz 933; papyrus codex, iv): II 331-49, 372-90.
1 371 P. Fouad 1 (P.I.F.A.O. inv. 31 + P. Soc. Pap. Alex. inv. 82; Packz 710, 711; school
exercise, ii): A 61-73. Edited (without reference to P. Fouad) by]. Schwartz,
BIFAO 54 (1954), 65. The Schwartz publication was listed as no. 395 and the P.
Fouad one as no. 567.
;, 1 ._,I
372 P. Milan. Vogl. 1. 8 bis (Packz 644; Tebtunis, ii): B 277-80, E 218-21, (3) 274-82,
372-4, with four other unidentified fragments. L. Friichtel, P/2W 61 (1941), 383;
E
A. Malnati, ZPE 121 (1998), 1831.
373 P. Hamb. 2. 155 (inv. 383; Packz 560; 1)! A I-16, 37-51, 55-67.
374 P. Mich. inv. 13 (Packz 565; ii-iii): A 32-57. Ed. N. E. Priest, ZPE 46 (1982) 51f.
11
1 376 P. Princ. 3. 108 (inv. GD 7740; P210112 588; ii): A 209-39.
1
1 377 P. Strasb. inv. Gr. 31+ 32 (Pack:-' 591; ii): A 215-442. Ed. ]. Schwartz, BIFAO 54
1
(1954), 45-62.
25:525-521. -. .' ; :
551. ._
;;5;1:.'::;.':;.'_-.
._ ..
-.-.»
.:--:;-
102 1.4. The Papyri
378 P. Princ. 3. 109 (inv. GD 7656; Pacltz 592; papyrus codex, v): A 216-37, 574-97.
379 P. M1611. inv. 2810 (P6616 599; K616616, 11); A 308-75, 397-520, 561-78. Ed. N.
E. Priest, ZPE 46 (1982), 58-69.
380 P. Strasb. inv. Gr. 83 (Packz 601) + P. Cair. SR 3675; iii): A 339-64, 374-5,
377-83, 392, 395-7, 544-611. The Strasbourg papyrus was edited by ]. Schwartz,
BIFAO 54 (1954), 63-5, the Cairo piece by W. Van Rengen, Studla Varia Brn-
xellensia 2 (1990), 207-15; the two joined by C. Gallazzi, ZPE 91 (1992), 77-9.
381 P. Soc. Pap. Alex. inv. 230 (Packz 602; iii): A 342-90. Ed. ]. Schwartz, BIFAO
46 (1947), 30-32. A
382 P. Soc. Pap. Alex. inv. 244 (Packz 605; iii): A 401-6, 437-42, 450-1. Ed.
]. Schwartz, BIFAO 46 (1947), 32 f.
383 P. Soc. Pap. Alex. inv. 265 (Packz 610; ii-iii): A 455-84. Ed. ]. Schwartz, BIF/1O
46 (1947), 33f.
384 P. Erl. 3 (inv. 117; Packz 621) + P. Brux. inv. E 8135 (iii): A 531-4, 565-611.
]oined and reedited by W. Van Rengen, Chronlqne d’Egypte 41 (1966), 335-9.
The Brussels piece was formerly listed as no. 457a.
385 P. Hamb. 2. 156 (inv. 384; Pack‘? 627; ii): B 12-26.
387 P. Hib. 195 (P6618 647;1-111661611); B 305-19, 338-60, 373-85.
388 P.I.F.A.O.iI1V. 326 (Packz 666; i): B 717-22. Ed. Schwartz, BIFAO 46 (1947), 71.
389 P. Berol. inv. 13424 (Packz 681; ii): F 23-36. Ed. W. Miiller, APF 39 (1993), 7f.
390 P. Berol. inv. 17049 + 17039 (Packz 686 + 693; iii): P 59-82, 239-51. Ed. W. M111-
ler, APF 41 (1995), 4f. 17039 was formerly listed as no. 392.
391 P. B6161. 17054 (P6618 688,111-1113c); r 125-35. Ed. s.W661(1967),611.
[392: see no. 390; 392a = no. 49, listed separately by Mette in error.]
393 P. Hamb. 2. 157 (inv. 385; Pack‘? 702; i-ii): P 384-8, 390-410, 440-61.
394 P. Mich. inv. 15 (ii-iii): P 408-19. Ed. C. W. Shelmerdine, BASP 12 (1975), 21 f.
[395; see no. 371; 397: see no. 363.]
398 P. Cair. Crawford 61 (P. Gradenwitz 133; Packz 726; Roman period): A 444-50.
399 P. Soc. Pap. Alex. inv. 245 (Packz 729; ii-iii): A 475-508, G) 362-99. Ed. _[.
5611-66116, BIFAO 46 (1947), 34-6.
400 P. Bodm. 1 (inv. 1 + 2; Pack: 736; Panopolis?, iii--iv): E 99-102, 107, 111-19,
124-32, 151-81, 191-215, 228-50, 259-80, 292-314, 323-45, 354-76, 385-105,
414-34, 444-64, 471-740, 750-810, 815-32, 838-62, 870-92, 903-9, Z 22-30,
61-7, 179-83, 190, 212-29, 233-430, 438-68, 470-507, 510-29. Ed. V. Martin,
Papyrus Bodmer I: Iliade, Chants 5 et 6 (Cologny—Geneva, 1954). The portion
containing Z was listed separately by Metre as no. 401.
[4012 See I10. 400.]
402 P.I.F.A.O. inv. 94v (Paclrz 738; early iii): E 142-62. Ed. ]. Schwartz, BIFAO 46
(1947), 361.
403 P. Soc. Pap. Alex. inv. 275 (Packz 743; iii?): E 289-300. Ed. Schwartz, BIFAO
54 (1954), 661.
404 P. Michaelides 3 (Packz 745; ii-iii): E 365-97.
405 P. Soc. Pap. Alex. inv. 273 (Pacltz 748; iii): E 450-504. Ed. ]. Schwartz, BIFAO
46 (1947), 37-41.
406 P. Berol. inv. 11645 (Packz 752; i): E 486-95. Ed. W. Miiller, Miscellanea Papy-
rologica, Papyrologica Florentine VII (ed. R. Pintaudi, Firenze 1980), 281.
4'
11:
11. 407 P. Berol. inv. 17009 (Pack2 764; parchment codex, iv): E 743-8, 786-93, and an
.11.
:11"
21:3
.,_=1§ unplaced scrap.
1%
E15.1:
:1:
:1'.1;
408 P. Soc. Pap. Alex. inv. 242 (Packz 765; late ii): E 762-859. Ed. Schwartz,
51 ? BIFAO 46 (1947), 41-48. -
.1.
.11.
:2 409 P. Princ. 3. 111 (inv. GD 7533; Packz 770; papyrus codex, iii): Z 1-15, 25--39.
:1
.12
1E
:2 410 P. Hib. 193 (Packz 774; Hibeh, iii BC): Z 4-7. Reedited by S. West (1967), 71-3.
411 P. Berol. inv. 11676 (Packz 775; ii): Z 75-86, 117-27. Ed. W. Miiller, Miscellanea
1»
1.-
111.
412 P. Mich. inv. 14 (i): Z 211-21. Ed. N. E. Priest, ZPE 46 (1982), 76f.
1;
413 P. Soc. Pap. Alex. inv. 213 (Packz 785; ii): Z 226-309. Ed. ]. Schwartz, BIFAO
.153.
IE
46 (1947), 48-53.
[414; see no. 480.]
415 P. Soc. Pap. Alex. inv. 216 (Packz 787; iii-iv): Z 316-44, 354-72. Ed. ]. Schwartz,
BIFAO 46 (1947), 53-6.
1
E
416 P. Bon. 1. 2 (inv. 5b; Packz 793; i-ii): Z 437-60. Ed. O. Montevecchi and
G. B. Pighi, Aegyptus 27 (1947), 172.
417 P. Berol. inv. 11685 (Packz 794; ii): Z 455-76. Ed. W. Miiller, Miscellanea Papy-
rologica, Papyrologica Florentine V11 (ed. R. Pintaudi, Firenze 1980), 283 f.
418 P. Haun. 1. 1 (inv. 5n; Packz 797; FayfimP, ii): H 4-15.
.-1.
.->-
::f. . 419 P.I.F.A.O. inv. 312 (Packz 801; Oxyrhynchus, iv?): H 65-86. Ed. ]. Schwartz,
1:11 BIFAO 46 (1947), 56f.
420 P. Hamb. 2. 158 (inv. 83; Packz 804; i): H 141-58.
421 P. Soc. Pap. Alex. inv. 212 (Packz 810; iii): H 283-98, 313--43. Ed. Schwartz,
BIFAO 46 (1947), 57-60.
422 P. Soc. Pap. Alex. inv. 263 (Packz 824; ii-iii): (9 150-62. Ed. ]. Schwartz, BIFAO
46 (1947), 60.
423 P. Hamb. 2. 159 (inv. 696; Packz 835; i): I 35-41.
E
424 P.S.I. 1274 (Pacltz 855; Oxyrhynchus, ii): K 84-102.
1 425 P. Berol. inv. 11911 + 17038 + 17048 (Pacltz 857 + 863) + 21155 (Hermupolis,
early iii): K 91, 110-15, 123-44, 150-60, 231-4, 240-77, 279-316, 320--512,
519-530, 535-48, 557-68. 17038 was listed by Mette as no. 430. Papyri joined
and reedited by H. Maehler, W. Miiller, and G. Poethke, APF 24/25 (1976),
13-26; Ioannidou (1996).
426 P. M1611. inv. 19 4 20615 (i-ii): K 192-213, 483-7. Ed. A. B. B1666, ZPE 59 (1985),
1 29f.
1
427 P. Cair. Crawford 59 (P. Gradenwitz 131; Packz 859; Roman period): K 205-20.
1
428 P. Berol. inv. 17152 (Packz 860; ii-iii): K 215-26.
429 P. Soc. Pap. Alex. inv. 224 (Packz 862; iii?): K 258-81. Ed. ]. Schwartz, BIFAO
1 46 (1947), 62f.
[430= see no. 425.]
431 P. Berol. inv. 7507 (Packz 871; Fayfirn, i): A 101-6. Ed. G. Poethke, APP 40
(1994), 5.
E 432 P. Hamb. 2. 153 (Packz 875; iii BC): A 265-89, 678-90, M 127-31, 1895-98. Re-
E
edited by S. West (1967), 91-~103.
433 P. Berol. inv. 11522 (Packz 887; ii): A 712-31. Ed. W. Miiller, APF 41 (1995),
14.
1
1
1
1
104 1.4. The Papyri
454 P. Strasb. inv. 1242 (P5518 891; ii-iii): A 815-26. Ed. ]. Schwartz, BIFAO 54
(1954), 57.
435 P.S.I. 1298 + Nuooi papiri Zettemri Fiorentini 8 (Packz 904; papyrus codex,
Anti1100p01iS, V-Vi): N 232-44, 260-326, 548-66, 577-—-93, 725-44, 754--831, E
107--61, 0 336-43, 364-71, H 171-97, 198-227, 234-57, 268-93, Y 106-58, X
143-64, 168-90, 406-28, 432-52, ‘P 495-510, 521-36.
436 P. Strasb. inv. 2480 (Packz 910; i): N 496-509. Ed. ]. Schwartz, BIF/1O 54
(1954), es.
457 P. M61-1. 1. 3 (P5518 915; papyrus codex, iii): 2 108-26, 162-77.
438 P. Berol. inv. 11910 + 21156 (Pacltz 919; ii-iii): E 235--60, 263-~7, 274-97,
299-304, 307-39, 342-75, 379--410, 437-47. ]0ined and edited by H. Maehler,
W. Miiller, and G. Poethke, APF 24/25 (1976), 26-32; Ioannidou (1996).
439 P. Berol. inv. 17005 (Packz 923; papyrus codex, vi): O 376-~96, 405-24. Ed.
W.Mii1ler, APP 41 (1995), 151.
439a P. Ibscher 3 (P. Berol. inv. 18123; Packz 928; i BC): O 658-73.
440 P. Berol. inv. 13922 (Packz 930; ii): H 1-37. Ed. W. Miiller, APP 41 (1995), 171.
441 P. Berol. inv. 16348 (Packz 936; ii-iii): H 507-18, 530—48. Ed. G. Poethke, APP
4o (1994), 91.
442 P. Berol. inv. 16970 (Packz 939; iv-v): H 695-702, 727-33. Ed. W. Miiller, APP
41 (1995), 181.
443 P. Erl. 4 (inv. 55; P9.C1(2 940; ii-iii): 1-1 758-69.
444 P. Mich. inv. 12 (i-ii): P 541450. Ed. N. E. Priest, ZPE 46 (1982), 931.
[445: see no. 239.]
446 P. Berol. inv. 11524 (Packz 956; i): Z 219-28.
447 P. Hamb. 2. 160 (inv. 654; Packz 967; i-ii): T 285-300. Reedited by A. Wou-
ters, Ancient Society 2 (1971), 62-5.
448 P. Gabra (Packz 968; Tunat-e1—Gabal, iii): T 365-~72. Ed. ]. Schwartz, BIFAO
54 (1954), 681.
449 P. B6101. inv. 16985 (Packz 980; i BC): (I1 359-401, 436*-47, 455--66, 474-90,
494-512, 514-31, 533-51, 557-608, X 1-18, 104-11, 265--83, 420--5, 439-55,
458--515, ‘P 1-2 (reclamantes). Described and partially illustrated by H. Poeth-
ke, Troja and T/amkien (Berlin 1980), 50-2 with Abb. 101/2.
450 P.I.F.A.O. inv. 112 (Packz 982; ii): (I) 414--24. Ed. ]. Schwartz, BIFAO 46
(1947), 551.
451 P. Michaelides 2 (Packz 997; i): ‘P 1--25, 37-59, 63-8.
452 P. Berol. inv. 11761 (Packz 1005; ii): ‘P 400-11, 444-55. Ed. G. Poethke, APP
44 (1998), 11.
453 P.S.I. 1275 (Pacltz 1011; Oxyrhynchus, ii): W 877-97.
453a P. Yale 2. 90 (inv. 1546; papyrus codex, ii): A 1-94. First edited by G. M. Paras-
soglou, C/Monique d’Eg3/pie 46 (1971), 313-17.
454 P. Mil. Vogl. 2. 29 (inv. 152; Packz 562; Tebtunis, iii): A 8426, 31-40.
455 P. Mil. Vogl. 2. 30 (inv. 216; Packz 564; Tebtunis, i): A 28--38, 58-68. First edi-
ted by A. Gianformaggio, Acme 9 (1956), 75 f.
4558 P. K6111 1. 21 (inv. 1030 + 46) + P. Mich. inv. 6653 (1): A 129-46, 151-90,
192-211. The Michigan portion was edited by R. Merkelbach, ZPE 14 (1974),
891., and reedited by N. E. Priest, ZPE 33 (1979), 35--7 and 46 (1982), 541.
‘1
1
475 P. Genav. inv. 202 (Packz 749; ii-»iii): E 461-510, 516-58. Ed. P. Schubert, Mus.
H.411». 4s (1991), 1-7.
106 1.4. The Papyri
475a P. Yale 2. 91 (inv. 1650 + 1651 + 1652; school exercise?, early i): E 625-36.
[476: see no. 280.]
477 Bodl. Gr. class. 1.42(P) (Packz 768; i-ii?): E 855-79.
478 P. Rain. inv. 39833 (Packz 769a; ii): E 891-6. Ed. H. Hunger, Wierz. St. 76
(1953), 150.
479 P. Mil. Vogl. 2. 34 (inv. 416; Paclcz 771; Madinet Madi, i): Z 1-21.
479.1 P. Oxy. 3154 (1); Z 2s-44, 55-s5.
479a P. Amst. 1. 2 (inv. 79; first half 01 ii): Z 171-83.
480 P. Berol. inv. 21102v + 17153 (papyrus codex, Hermupolis, ii): Z 217-232,
247-53, 264-82. Ed. H. Maehler, Mus. Halo. 24 (1967), 611.; id. (1975), 3681.;
Ioannidou (1996).
480a P. Sorb. inv. 2302 (Ptolemaic period): Z 280-9. Ed. B. Boyaval, BIFAO 65
(1967), 57-65. (Omitted by oversight 1rom the 1ist in my edition, p. xlix, but
cited in the apparatus.)
481 P. Strasb. inv. Gr. 2675 (Packz 789; papyrus codex, iv): Z 282-300, 326-44, 1
350-63,401—5,530-49,578-91,620-34,663-75,705-13,Ki1-4,33-44,70-82,
117-23, 157-66, 200-9, 240-54, 285-97, 327-41, 378-88, 418-31, 469-77,
509-22, 558-66, A.21-9, 113-14, 147-55, 191-8, 237-40,.274-82, 312-23,
358-66, 401-9, 445-52, 485-93, 522 O1‘ 523, 531-8, 576-81, 617-24, 660-5,
698-707, 741-7, 779-~91, 827-34, M 15--27, 62-9, 100--10, 142-52, 182-93,
225-36,356-69,401-12,445—54,19 18-23,56-67,107-12,140-52,180-93,
223-35,264-77,304-19,347-62,388-404,429-43,469-84,511-24,552-69,
594-510, 535-53, 583-97, 725-41, 770-s4, s13-27, E 2s-34. Ed. 1. Schwartz,
BIFAO 51 (1952), 151-58.
482 TESI.1376(PaCk2798;PdQI1150-66.
483 P. Genav. inv. 85 (Packz 800; ii): H 63-124. Ed. C. Wehrli, Mélanges Esther
Bréguet (Geneva 1975), 1-5.
483a P. Ant. 157 (parchment codex, vi-vii): H 117--27, 135-55, 162-9.
484 P. Mil. Vogl. 3. 112 (inv. 419; Pack‘? 809; Madinet Madi, ii): H 255-9.
485 P. Mil. 2. 14 (Packz 814; i-ii): H 427-41.
486 P. Mil. Vogl. 2. 36 (inv. 414; Packi 815; Madinet Madi, i BC): H 482, ('3 1.
486a P. Ant. 158 (parchment codex, iii): G) 203-10, 241-7, 495-555, I 1-3, 149-224,
369-90,407-30,440-50,455-90,494-513,592-606,630-49,A.375-91,409-20,
11613-98.
486b P. Fay. ed. R. A. C5155, ZPE 5 (1970), 253-4 (ii-iii): 1') 158-91.
487 Bodl. Gr. class. 1.24(P) (Packz 826; iii): G‘) 198-213.
488 P. Coll. Daris 2 (inv. 8; Packz 827a; iii): E) 262-8, 292-314. Ed. S. Daris, Aegyp-—
tus 42 (1952), 11911.
489 P. Heid. Siegmann 202 (inv. 1969; Packz 828; Batu-el-Harit, iii): G) 264--300.
489a P. Fay. 5d. R. A. Q5155, ZPE 5 (1970), 254-5 (11--111)= o 424443.
490 P.S0fb.1.3 finv.2246-P2266;H-fiQ:()452-73,478-83,503-8.
490a P. Ant. 159 (papyrus codex, late iii): I 60-80, 89-97, 105-26, 138-47.
491 P. Med. inv. 71.81 (ii): 1 248-62, 285-92. Ed. S. Daris, Aegyptus 52 (1971), 681.
491a P. Ant. 160 (papyrus codex, iii-iv): I 222-344, 354-9, 367-9, 464-95, 501-34,
538, 543-653, 657-62, 664-73, 676-700. C1. M. Apthorp, ZPE 57 (1984),
52.
I.4. The Papyri 107
491b P. Ant. 161 (papyrus codex, v): I 470-2, 509-27, 549-65, 616-25, 657-64, K
48-64,82-102,54 152-6,191-6.
492 P. Rain. inv. 39835 (Packz 849a; ii): I 483-99. Ed. H. Hunger, Wien. St. 76
(1963) 1601. 4
493 P.S.I. 1377 (Packz 851; ii): 1 682-709.
494 P. Mil. Vogl. 3. 113 (inv. 432; Packz 856; iii-ii BC): K 106-21.
494a P. Ant. 162 (papyrus codex, ii-iii): K 178-84, 228-34.
494a1 P. Yale 1. 9 (inv. 1062; Abutig, i): I 272-91.
494a2 P. Alex. (inv. unstated), ed. H. Riad and]. Schwartz, C/aronique d’Egypte 43
(1968), 114-21 (school exercise, wooden tablet, iv-v): A 10-23, 35-47, 49-60,
71-81 with glossary to 31-46.
494a3 P. Alex. inv. 546 (Fayfim?, i-ii): A 347-63, 802-20. Ed. P. Montanari in
A. Carlini, Przpiri letterari greci (Pisa 1978), no. 11.
494b P. Berol. inv. 21109 (I-Iermupolisi’, ii): A 360-77, 427-56, 639-60, 673-703,
716-19. Ed. H. Maehler, Mus. Helo. 24 (1967), 621.; id. (1975), 3771.; loan-
nidou (1996).
494b1 P. K6111 2. 75 (inv. 5605; ii-iii): A 515-38.
495 P. Strasb. inv. Gr. 1600a—b + 1624 + 1632b + 1654 a-b + 1660b (Packz 885;
papyrus codex, iv): A 652-83, 689-720. Ed. Schwartz, BIFAO 61 (1962),
169-71.
496 P.Sofld 1A-finv.2265;BAagdohgifi.BCD:h1228-38,246-65.
496a :P.hdon.(3r.inv.88 G BCD:hA 296-304,335-72.Ifi1IF.1dcnuanarL.AIkenae~
um 52 (1974), 25-32.
497 P. Oxy. 2540 (i-ii): N 474-s4.
[497a Mette = no. 308.]
498 Bodl. Gr. class. d.45(P) (Packz 913; ii): N 584-640.
499 P. Oxy. 2541 (iii): E 274--302.
500 P.()Xy.2542 fi—H):C)158-63,172—82,212-38.
SOOJ P.(PXY.3155(H-H01(2316,318,373-406,420-37.
500a1 P. Kijiln 2.76 (inv. 4783; i BC): O 625-30, 650-7.
500a3 P. Mon. Gr. inv. 610, 4 (parchment codex, v): O 664-75, 689-700. A. Carlini,
Pdpiri letterari greci (Pisa 1978), no. 37.
500b P.()Xy.2748(fi):T1129—60.
501 P.L4fl.\Rnfl.3.114-finv.1091;Q:TI452-73,491-500.
501J P. Mil. Vogl. 6.258 (Tebtunis, early i): X 173-7.
501a P. Ant. 163 (papyrus codex, ii--iii): H 697-703, 735-40.
501a1 P. Med. inv. C..N.R. 68.3 (ii--iii): P 127-64. Ed. S. Daris, Aegyptus 47 (1967),
191-4.
501b P. Ant. 164 (papyrus codex, iv): P 220-37, 267-86.
5010 P. Sorb. inv. 2. 303 (Ptolemaic): P 566-7819. Ed. B. Boyaval, BIFAO 65 (1967),
57-69.
502 P. Heid. Siegmann 203 (inv. 1971 R; Packz 951; Oxyrhynchus, iii): P 729-35.
502a P. Ant. 165 (papyrus codex, iv): Z 135-44, 177-185, ‘P 624-32, 669-75.
503 P.S.I. 1458 (Packz 955; Oxyrhynchus, i): Z 187(?), 213-23.
504 P. Oxy. 2530 (ii): T 26-30. Identified and reedited by me, CR 16 (1966),
274.
108 1.4. The Papyri
504a P. Dubl. Trin. Coll. Sel. Box inv. 128 (ii): T 291-315. Ed. A. Wouters, Ancient
Society 2 (1971), 56-62.
505 P. Rain. inv. 39701 (Packz 970a; i-ii): T 86-100. Ed. H. Hunger, Wz'en.__..St. 76
(1953), 151.
506 P. Mil. Vogl. 3. 115 (inv. 412 + 413; Packz 973; Madinet Madi, i): T 377-87,
418-51.
507 P.S.I. 1378 (Packz 977; Euhemeria, ii-iii): (I1 56-73.
507a P. Ant. 166 (parchment codex, iv): (I1 488-502, 530-44.
508 P. Rain. inv. 31936 (Packz 987a; early iii): X 27-38. Ed. I-I. Hunger, Wien. St.
76 (1963), 161.
509 P. Mil. Vogl. 2. 35 (inv. 223; Packz 990; i): X 131-62.
510 P. Mil. Vogl. 3. 116 (inv. 417; Pack‘? 992; Madinet Madi, i-ii): X 142-58.
510a P. Ant. 167 (papyrus codex, iii-iv): X 202-14, 251-63.
511 P. Mil. Vogl. 3. 117 + 118 (inv. 425 + 428; Packz 1002 + 1004; i BC); ‘P 91--107,
393-8, 405-10. Identified as belonging to the same roll by C. Gallazzi, ZPE 71
(1988), 66-70; the second fragment had been listed by Mette as no. 512.
513 P.S.I. 1379 + 1189 (Packz 1008; Oxyrhynchus, ii): ‘P 524-55, Q 648-81. 1189
was formerly listed as no. 332.
514 P. Oxy. I-1515. 2 (11); A 1422.
515 P. K515 2. 5s (inv. 47s4;11); A 5--29.
516 O. Petrie 403 + 472 (Packz 561; school exercise, Byzantine): A 1-10, 21-28.
516a O. Petrie 404 (Pack: 561; Byzantine): A 10-~15, 28-32.
517 P. Hamb. 3. 193 (inv. 779; 1); A 22_44.
518 P. K6111 1. 20 (inv. 49; 11); A 27-33.
519 P. Mich. inv. 1576(9), (iii): A 37-47. Ed. N. E. Priest, ZPE 46 (1982), 531.
520 P. Bryn Mawr 1 (inv. 6; i): A 40-4. Ed. T. Renner, BASP 22 (1985), 2721.
521 P. Narm. inv. 66.60a (Madinet Madi, ii-iii): A 42-9. Ed. C. Gallazzi, Ann. Sew.
59 (1953), 1841.
522 O. Petrie 402 (Packz 561; school exercise?, Byzantine): A 58-60.
523 P. K515 2. 59 (inv. 1547), (11); A 55--77.
524 P. Kiiln 2. 70 (inv. 5138; school text?, i BC): A 108-31, 137--53.
525 O. Petrie 406 + 471 (UC 32220, 32221; Pack‘? 561; school exercise P, Byzantine):
A 1154-21, 124-7.
526 P. Rain. inv. 29372 (Packz 2875; ii): A 149-57. Identified and reedited by
M. Gronewald, ZPE 47 (1982), 104.
527 P. K515 5. 207 (inv. 4715; ii): A 1504-53.
528 P. Narm. inv. 66.88a (Madinet Madi, ii): A 196-210, 232-45. Ed. C. Gallazzi,
Ann. Ser"0. 69 (1983), 185-8. -
529 P. Mich. inv. 3430 (papyrus codex, iii-iv): A 212-~67. Ed. N. E. Priest, ZPE 46
(1982), 55-s.
530 P. Berol. inv. 21135 (Fayfim?, iii): A 246-58. Ed. H. Maehler (1975), 366; loan-
nidou (1996).
531 P. K6111 3. 134 (inv. 5902; ii): A 251-66.
532 P. Mich. inv. 4457a (ii): A 340-6. Ed. N. E. Priest, ZPE 46 (1982), 69.
533 P. Hamb. 3. 194 (inv. 780; ii): A 360-72.
534 P. Yale 1. 4 (inv. 489; ii): A 361-93.
I.4. The Papyri 109
4 [599 = w45.]
1
600 P. K5111 1. 30 (inv. 2605; ii): 1 252-6, 283-91.
601 P. Berol. inv. 21221 (Eayfim?, ii): I 312--36. Ed. Maehler (1975), 3741.; 1oan—
4 nidou (1996). .
i 602 P. Mich. inv. 6052 (i-ii): 1 579-99. Ed. N. E. Priest, ZPE 46 (1982), 801.
E
603 P. K6111 2. 74 (inv. 5962; ii--iii): 1 682-96.
6036 P. Cair. inv. ]E 45616 (1): K 21-8. Ed. A. 1\/1616611, ZPE 121 (1998), 184.
604 P. K616 1. 31 (inv. 642; ii): K 32--40.
._,>
605 P. Yale 2. 92 (inv. 1601a; iii): K 33-42.
606 P. Berol. inv. 21130 (papyrus codex, Hermupolis, vi-vii): K 102-26, 130-54.
1 Ed. Maehler (1975), 375; Ioannidou (1996).
5_::-7;
607 P. Yale 1. 10 (inv. 552; i BC-i AD): K 311--19.
E
2
608 P. K6111 1. 32 (inv. 2682; ii): K 319-26.
609 P. Mich. inv. 6972 (ii BC): K 421-34, 445-60. Ed. A. T. Edwards, ZPE 56
1
(1984), 11415.
610 P. Yale 2. 93 (inv. 1602; ii): K 439--61.
1
611 P. Berol. inv. 21164 (parchment codex, iv): K 488-502, 513-27. Ed. Maehler
1 (1975), 3751.; Ioannidou (1996).
612 P. K616 4. 182 (inv. 3293; 11); K 575-9.
] . .
613 P. Berol. inv. 21159 (E1ephantine?, ii-iii): A 130-6. Ed. Maehler (1975), 3761.;
:;;-].
...,
I121; .
614 P. K6111 1. 33 (inv. 2694; ii-iii): A 212-56.
615 I’. Berol. inv. 18177; i--ii): A 237-54. Ed. W. Miiller, Forsc/mngen und Bericb-
te der staatlic/yen Museen zu Berlin 8 (1967), 1011.
1 616 P. Wessely Prag. inv. Gr. iv.175 (ii-iii): A 461-7. Ed. R. Pintaudi and L. Vid-
1 man, Phil. 130 (1986), 138-40.
[617; see no. 360.]
iii;
618 Nuovi papiri letterari Fiorentini 7 (ii): A 613-8.
:::>
:..1-
E,-.... 619 P. Heid. inv. G 4011 (i BC--i AD): A 819-32. Ed. Seider (1968-70), II 701.; D.
Hagedorn, ZPE 108 (1995), 189--92.
620 P. Mich. inv. 5574 (iii): M 114--46. Ed. N. E. Priest, ZPE 46 (1982), 85-7.
621 P. Berol. inv. 21181 + 17004 (papyrus codex, Hermupolis, vi): M 179--213,
217-50. Ed. Maehler (1975), 379-81; G. Poethke, APP 40 (1994), 5-8; 10311-
nidou (1996).
622 P. Berol. inv. 21185 (Soknopaiou Nesos, i): M 459-71. Ed. Maehler (1975), 381;
Ioannidou (1996).
623 P. M1614. inv. 1210 (iii): N 290-301. Ed. N. E. P6661, ZPE 46 (1982), 87.
3-§ §;'1 624 P. Narm. inv. 68.49c (Madinet Madi, ii): N 409-14. Ed. C. Gallazzi, Arm. Sew.
252151
=:5:-L
.1151
;_.:.:.__.i
69 (1983), 1881.
='.=.-3::
=5I--.5-_-..2%5 625 P. Laur. 3. 274 (papyrus codex, vi-vii): N 425-42, 460--9. Ed. R. Pintaudi, ZPE
§_§'§} §
E5:-‘:1 42 (1981), 39-41.
§§§j§§i
§_
-:.-:¢ 626 P. Berol. inv. 21151 (papyrus codex, Hermupolis, v): N 532-7, 544-8. Ed.
Maehler (1975), 381; Ioannidou (1996).
:5{::3
5.5.4
»::»:<
627 P. Berol. inv. 21191 (papyrus codex, Hermupolis, v-vi): N 783-92, 820-30. Ed.
2
Maehler (1975), 382; Ioannidou (1996).
628 P. K6111 34 (inv. 2588; i): E 311-26 with glosses (= pap. VI1Ia Erbse).
='s'§:-1
.2;2';_lz
112 1.4. The Papyri
629 P. Mich. inv. 4990 (ii-iii): E 314-20. Ed. N. E. Priest, ZPE 46 (1982), 871.
630 P. Berol. inv. 21216 (i BC): O 5-31. Ed. Maehler (1975), 382; Ioannidou
(1996).
631 P. Mich. inv. 1575 (i-ii): O 157-81. Ed. N. E. Priest, ZPE 46 (1982), 88-91.
632 P. Oxy. 3323 (ii-iii): O 162-97, 11 55-65, 101-50.
633 P. Genav. inv. 194 (vi-vii): O 318-27. Ed. C. Wehrli, Mus. H6119. 37 (1980), 2121.
634 P. Hamb. 3. 198 (inv. 757; Arsinoite nome?, ii): O 399-400, 405-6, 431--56.
635 P. Mich. inv. 6654 (papyrus codex, iii-iv): O 596-609, 631-46. Ed. N. E. Priest,
ZPE 46 (1982), 91-3.
636 P. Berol. inv. 21194 + 11913 (papyrus codex, Hermupolis, vi-vii): O 631-63,
688-95, P 268-337, 475-93, 512-27, ‘P 323-6, 357-60. The fist part edited by
Maehler (1975), 383-7; joined and reedited by H. Maehler, W. Miiller, and G.
Poethke, APF 24/25 (1976), 32-8; Ioannidou (1996).
637 P. Yale 2. 94 (inv. 6899. + b; ii-iii): 11 97-113.
638 P. K5111 1. 35 (inv. 529; i--ii)! 11 265-74.
[639; renumbered as 370.]
640 P. Yale 1. 11 (inv. 1082; AbutigP, late i): H 422-38.
641 P. Harr. 2. 176 (inv. 273c; iii): H 520-39.
642 P. Mich. inv. 3753 (ii): 11 532-8. Ed. R. Heisler, ZPE 56 (1984), 10.
643 P. Berol. inv. 21195 (papyrus codex, Hermupolis, iv-v): H 611-22, 644-56. Ed.
Maehler (1975), 387; Ioannidou (1996).
644 P. Oxy. 3440 (ii-iii): 11 612-54.
645 P. K6ln 1. 36 (inv. 56; cartonnage from papyrus codex?, v-vi): P 166-88, 209-34,
X 16-18, 62-9.
646 P. Yale 2. 95 (inv. 532; early iii): P 575-90.
647 P. Oxy. 3663 (111); >3 33--50, 55-8, 73, 98-123, 182-93, 206-27, 261-77, 293-309,
325--42, 355, 375-89, 392-408.
648 P. Mich. inv. 17 + 18 (ii-iii): )3 223-41, 251-75. D. C. Elliott, B/1519 14
(1977), 106-8. C1. On n0. 239.
[649 is part of no. 239; 650 renumbered as 369.]
651 P.S.I. 1459 (1)1 353-65, 419-~36.
652 P. Berol. inv. 21152 (papyrus codex, Hermupolis, vii‘): (I) 307-17, 338-49. Ed.
Maehler (1975), 3871.; Ioannidou (1996).
[653 was wrongly listed as a papyrus oi the Iliad. It contains )( 127-64.]
654 P. Yale 1. 13 (inv. 518; i BC--i AD): X 402-22.
655 P. Berol. inv. 21204 (papyrus codex, Hermupolis, v-vi): ‘P 485--9, 525-8. Ed.
Maehler (1975), 3881.; Ioannidou (1996).
656 P. Kfiiln 1. 37 + 7.301 (inv. 55 + 179; i): Q 1--13, 16-23, 43-6, 49-57, 70-92.
657 P. Cair. inv. 25 (S.R. 3049; papyrus codex, Philadelphia, iv): Q 1-34. Ed.
M. Wahba, Mvfipn Fecopyiou ’A. l”Ietpo1"r015?\0u (ed. A. Biscardi et al., Athens
1984), II 4411.
658 P. A6461. inv. 191 (111); Q 94-9, 102-10. Ed. P. ]. Sijpesteijn, ZPE 30 (1978),
231 1.
659 P. K6111 1. 38 (inv. 1885; ii): Q 303-11.
660 P. Y616 2. 96 (inv. 1542,11); o 318-84.
661 P. Oxy. Hels. 4 (i BC-i AD): Q 514-43.
1.4. The Papyri 113
662 P. Strasb. inv. Gr. 2344.1--5 (Packz 432; iii—-ii BC): T 325-9. First edited as a
fragment of Euripides’ Alexandros by W. Crijnert, NGG 1922:1, 1-17; iden~
tified and reedited by M. Huys, ZPE 79 (1989), 261-5.
662a P. Princ. inv. AM 11224D (ii-—-iii): ‘I? 464—~8. Ed. B. H. Kraut, BASP 24 (1987),
35.
663 P. Oxy. 3825 (early ii): A 61——-86, 98-120, 204?, 229-52.
664 P. Oxy. 3826, (parchment codex, iv-v): A 517-22, E 1-4, 31-45, 62-75.
665 P. Oxy. 3827 (ii): A 337—61 and in the margin 316a or 346a.
666 P. Ryl. 3. 495 (ii): E 445-50. Identified by M. Gronewald, ZPE 86 (1991), 1.
667 P. Aberd. 145 (Packz 2795; i—~ii): B 482-7. Identified by M. Gronewald, ZPE 87
(1991), 9.
668 P. Aberd. 146A (Paclsz 2796; i): T 197-202. Identified by M. Gronewald, ZPE
s7 (1991), 91.
[669 was wrongly listed as a papyrus of the Iliad. It contains )( 175-208, 214-19,
220-49, 255-60.]
670 Papyrus in private collection, edited by P. Schubert, Mas. Helo. 47 (1990), 34
(ii): B 126-30.
671 P. Vat. G1-. 64 (ii-i BC): H 32-4, 37-»42, 44-5, so-9, es“-81. Ed. R. Pintaudi,
Stndi Classici e Orientali 39 (1990), 165-73; G. Bastianini and C. Gallazzi, ib.
40 (1990), 433-6; M. W. Haslam, BASP 28 (1991), 33--4.
672 Papyrus in private collection, edited by D. Montserrat, BlCS 38 (1991-3), 55-7
(iii BC): P 637--44, 679-—-85.
673 P. Duke inv. 4 R (i BC): X 111-49. Images of this and the following Duke
papyri are accessible at http://0dyssey.lib.dul<e.edu:80/papyrus/
674 P. Duke inv. 58 (i BC): H 346-9.
675 P. Duke inv. 59 (i BC): G) 3-~12.
676 P. Duke inv. 970 (1); A 127-39.
677 P. Duke inv. 971 V (iii): A 528-40.
678 P. Berol. inv. 21333 (Hermupolis, v): IT 3145. W. Brashear, Proceedings of the
XX International Congress of Papyrology (Copenhagen 1994), 284.
679 P. Berol. inv. 21339 (iv): W 670--6, 710—-14. W. Brashear, Proceedings of the XX
International Congress of Papyrology (Copenhagen 1994), 284f.
680 P. Berol. inv. 1760901); B 629-37. Ed. W. Miiller, APP 39 (1993), 7.
681 P. Hal. inv. 33 (ii): B 851-61. Ed. W. Luppe,/1PF 37 (1991), 6.
682 P. Berol. inv. 11747 (ii-iii): H 1——1 7, 37——68. W. Miiller, APP 41 (1995), 7—9.
683 P. Berol. inv. 17002 (i BC--i AD): A 116-~40, 144~61. Ed. W. Miiller, APF 41
(1995), 121.
684 P.S.I. inv. 1914 (i BC-i AD): E 907-Z 2 with colophon to E (Z 1-2 are recla—
rnantes). Ed. Anon., Dai Papiri della Societa ltaliana. Omaggio al XX Con~
gresso Internaz. dz‘ Papirologia (Firenze 1992), 5-10.
ass P. K6111 7. 298 (inv. 3602v; ii): B 41-3.
ass P. K6111 7. 299 (inv. 99; iii): E 283-5.
687 P. K6111 7. 300 (inv. 4367; ii-—iii): K 7--28.
688 P. Koln 7. 301 (inv. 1791"; late i): Q 12-17.
689 P. Mich. inv. 3694 (iv): B 284-302. Ed. Schwendner (1988), 2f.
690 P. Mich. inv. 6239 (ii): B 488—532. Ed. Schwendner (1988), 4-8.
114 I.4. The Papyri
691 P. Mich. inv. 1218 (ii): B 745—54. Ed. Schwendner (1988), 8f.
692 P. Mich. inv. 6225 (iii): B 738-69, 773-89. SCl‘1W€ndI1€1‘ (1988), 9-13.
693 P. Mich. (no number; ii-iii): F 408—17. Ed. Schwendner (1988), 13.
694 P. Mich. inv. 5576a (ii): F 401--11. Ed. Schwendner (1988), 14.
695 P. Berol. inv. 21229 (Hermupolis, v--vi): K 324—-35, 362—74. Ed. P. Sarischouli,
'E7\7\r|v11<o’1 45 (1995), 12-15; Ioannidou (1996).
696 P. Berol. inv. 21278 (Hermupolis, iii-iv): E 719—-24, 765-70. Ed. P. Sarischou-
li, 'EMnv11<0'1 45 (1995), 10—12; Ioannidou (1996).
697 P. Berol. inv. 21279v (ii): Y 351-65. Ed. P. Sarischouli, 'E?\2\nvu<o’1 45 (1995),
7--10; Ioannidou (1996).
698 P. Mich. Koenen (P. Mic/J. XVIII), 769 (inv. 4768, ed. G. W. Most; i-ii): H
III-5»
698a ch. inv. 5694d (papyrus codex, Karanis): H 457-78, 490—506. Unpublished.
699 eid. 4. 289 (i): A 1-6 preceded by prose mentioning Telemachus. Ed.
ontanari, P. Held. IV (Heidelberg 1986), 10—13 = id. (1995), 127—30.
700 rol. inv. 17211 (iii): B 561-78. Ed. G. PO€tl'll§e, APP 44 (1998), 1—-2.
701 19117597“ °"?§L“erol.
§:~* inv. 11912: H 223-40, 251-65, 272-95. Unpublished.
702 P. Berol. inv. 31080 (i): (9 288-300, 327-31. Ed. G. Poethke, APF 44 (1998), 71.
703 P. Berol. inv. 16007 (i BC--i AD): P 51——71, 76-98. Ed. G. Poethlce, APF 44
(1998), 8-10.
704 P. Berol. inv. 16980 (V): ‘P 158—64, 194-201. Ed. G. Poethlce, APP 44 (1998), 10.
704a Wooden tablet edited by K. A. Worp, Mnem. 51 (1998), 207-10 (iv): M 294-7.
705 P. Ashm inv 20 3B 31/B(6-~7)a (11); A 1-31.
706 P. Ashm inv 5 1B 56/H(b) (ii-iii): A 1--23, 141--54, 170-so.
707 P. Ashm inv 30 4B 37/F(5-7)a: A 1--23.
708 P. Ashm inv 21 3B 27/B(3-4)b (ii-iii): A 1-22.
709 P. Ashm inv 114/95(1) (1-41); A 1-19.
710 P. Ashm. inv 101/139(d)= A 1-15.
711 P. Ashm inv 17 2B 61/H(a) (ii-—iii): A 1-10, followed by six lines that do not
seem to be A 11ff.
712 P. Ashrn inv 27 3B 40/F(4)a (ii): A 1-s.
713 P. Ashm inv 119/64(6) (iv): A 1-3, E 132-9.
714 P. Ashm inv 5 1B 57/B(cc) (ii-iii): A 1-2.
715 P. Ashm inv 7o/6(6) (11-111); A 2-9.
717 P. Ashm inv 106/8(6); A 6-10.
719 P. Ashm inv 35 4B 64/K(1-3)a (i): A 1s_2s.7
722 P. Ashm. inv 33 4B 85/E(12—14)a (141); A 2s-40.
723 P. Ashm. inv 27 3B 39/G(1—3)d (ii-iii): A 33-48.
724 P. Ashm inv 105/101(a) (1111); A 34-42.
725 P. Ashm inv 85/18(a): A 37-46.
726 P. Ashm inv 97/41(a): A 41-56.
727 P. Ashm inv 40 5B 116/G(6—9)b (iii): A 59--65 (= 93) + 96-8.
729 P. Ashm. inv. 22 3B 20/A(1-2)a (11); A 683-76, 97-108.
730 P. Ashm inv 66 6B 28/C(1—~2)e (iv-v): A 73-6.
732 P. Ashrn inv 48 5B 31/K(1)a: A 90-1, 96-100, 120—-5, 128-52 (and an uniden—
tified fragment).
1.4. The Papyri 115
")
Ashm inv 10 1B 160/1(8) (11); A 91-112.
P Ashm inv 118/44(6); A 94-8, 101--8.
T) Ashm inv 9 1B 177/C(a) (ii-iii): A 95—100.
T) Ashm inv 27 313 43/F(1)6 (ii): A 102-8, 127~38.
T) Ashm inv 33 48 79/B(2-5)6 (ii--iii): A 114-63.
P Ashm inv 8 113 199/D(1--2)a (ii-iii): A 120-8, 140-6, 160-78.
o Ashm inv 37 4B 111/M(1--3)a: A 122-34, 135--67.
-1.
P. Ashm. inv 100/160(8) (ii-iii): T 349--66, 387-99, 409-16, 426-36, 447-51 (and
O
some scraps).
7)
Ashm inv 104/36(6); 1" 355-64.
7)
Ashm inv 105/178(4) (11-111); r 370-87.
O
7)
Ashm. inv 51 4B 19/B(4-7)a (iii): F 371--84.
7)
Ashm. inv 51 48 19/c(1)b (11-111); 1" 387-404, 409-424.
P Ashm inv 5 18 38/F(g) (111); 1" 396-406.
P Ashm inv 28 4B 61/G(1-2)a: F 404-34, 446-61.
P Ashm. inv 7 18 3/1(1); r 413-34.
P Ashm inv 106/60(8); r 418-30.
P Ashm. inv 16 2B 54/B(a) (ii-iii): F 418-25.
P Ashm inv 44 5B 62/K(1--2)a: P 424-39, 445-9.
7)
Ashm inv 105/97(5); F 429-11.
7)
Ashm. inv 20 38 30/0(6) (111); 1" 442-57.
7)
Ashm inv 47 58 46/E(1-3)b (111); r 448-54, 457-61.
P Ashm inv 103/75(8); A 23-33.
P Ashm. inv 19 28 81/C(j, tn) (11-111); A 31-61.
'3 Ashm. inv 19 28 78/D(9-10)b (11); A 44-61.
T)
1. Ashm inv 32 4B 7/H(3-4)c: A 53-61.
7)
Ashm inv 62 68 76/F(7-8)¢ (111); A 81-95.
7)
Ashm. inv 104/42(1): A 93-9, 121-8.
P Ashm. inv 118/28(4) (1-11); A 113-41.
1.4. The Papyri 119
1 997
1 999 P Ashm. inv. 48 58 30/](3-6);; (11); E 197-207, 323-49, 351-67, 373-91,
396-405, 452-9, 835-53.
1000 P Ashm. inv. 50 4B 33/G(2): E 209-74 (and one fragment).
1001 P Ashm. inv. 89A/27(5) (11); E 213-55, 260-80, 285-302.
E 1002 P Ashm. inv. 5 18 53K(m) (111); E 249-66.
1003 P Ashm. inv. 103/165(6) (11); E 272-92.
1005 P Ashm. inv. 30 48 37/F(1-3)6 (111); E 279-91.
1006 P Ashm. inv. 27 38 43/D(1-3);1; E 287-312.
1007 P. Ashm. inv. 11 18 151/C(g) (11-111); 8 301-13.
120 I.4. The Papyri
1024 ‘J
Ashm. inv 51 4B 19/D(1-2)c (iii): E 564-71.
0
1029 7)
Ashm. inv 2 1B 94/L(c) (Byzantine): E 593—608, 621--9.
O
1031 7)
Ashm. inv 22 3B 19/](a): E 615-35 et al.
O
1033 ')
Ashm. inv 103/19(0) (iii): E 640—54, 681-94.
U
1034 '2)
Ashm. inv 24 3B 74/D(b) (iii): E 705-26.
0
1035 ')
Ashm. inv. 106/ 105(0) (Byzantine): E 714--23, 749-~58.
1037 7)
Ashm. inv. 86/84(a) (i): E 744-807.
1038 ')
Ashm. inv. 102/101(a) (ii): E 748—~79.
1039 ')
Ashm. inv. 89A/93(a) (ii): E 748-68.
1040 ")
Ashm. inv. 14 113 204/H(d, 6, f) (iii): E 769-75, 794-800, 817-25, 841-9,
867-75, 8904900.
1042 P. Ashm inv. 85/81(3); E 818-45.
1043 P. Ashm inv. 24 313 75/12(6) (11); Z 1-6, 10--17.
1044 P. Ashm. inv. 105/98(1)) (i): Z 25»-33, 41-57 with interlinear glosses (and two
scraps).
1046 P Ashm. inv. 114/55(61): Z 30-7.
1049 P Ashm. inv 101/29(6) (ii): Z 54-74.
1050 P Ashm. inv. 112/23(3); Z 89-100.
1051 P Ashm. inv 112/109(6); Z 116-26, 146452.
1054 P Ashm. inv 104/92(6); Z 150-60, 163-4.
1059 P Ashm. inv. 103/208(6) (1); Z 199~210.
1060 P Ashm. inv 31 4B 9/N(5)a + (6-8)a (ii—-iii): Z 206-65.
1061 P Ashm. inv 30 4B 41/F(1-2)a (iii»-iv): Z 292--9.
1062 P Ashm. inv 102/12(1); Z 304-10.
1063 P Ashm. inv 22 3B 19/19(6) (ii): Z 309-21, 323-34, 360-8.
1064 P Ashm. inv 103/4(6); Z 342-9.
1065 P Ashm. inv 30 4B 38/](1—-5)a (iii): Z 373-6, 381-7.
1066 P Ashm. inv 75/69(a) (ii-iii): Z 396--422.
1067 P. Ashm. inv 102/48(3); Z 412-17, 45047.
I.4. The Papyri 121
1072 P. Ashm. inv. 41 513 83/](5-6)a + (1-4)d= H 6.10-66. (061118 24-32, 34-6, 38-43,
47-53)
1073 P Ashm. inv. 25 313 58/L(f) (iii-iv): H 34-67.
1074 P Ashm. inv. 19 213 72/13(1)6 (iii): H 48-54.
1076 P Ashm. inv. 27 313 45/D(1-4)3 (iii): H 136-49.
1077 P Ashm. inv. 47 SB 45/F(1--3)<: (ii): H 150-67.
1079 P Ashm. inv. 100/65(6) (ii): H 237-47.
1080 P Ashm. inv. 37 413 103/F(4)c (ii-iii): H 252-62.
1081 I’ Ashm. inv. 66 6B 27/B(1—2)a (Byzantine): H 287--92, 318-25 and scraps.
1082 P Ashm. inv. 21 313 27/E(3--6)b (i-ii): H 287-91.
1083 I’ Ashm. inv. 24 313 74/G(g) (ii): H 288-91, 358-74, 408-21.
1084 I’ Ashm. inv. 100/89(3) (iii): H 290-313.
1085 I) Ashm. inv. 33 4B 79/N(7-10)a (ii): H 304-8, 310-22.
1086 iP Ashm. inv. 100/159(b]1(iii--iv): H 351-2.
1088 1P Ashm. inv. 102/91-2 (i-ii): H 420-7, 443-63, 465-<3 1 (reclamans).
1089 iP Ashm. inv. 103/48(a): H 443-52, 473-G) 2 (reclamantes).
1090 .P Ashm. inv. 100/183(g) (iii-iv): H 449-58 (616. 452).
1093 I’ Ashm. inv. 66 6B 27/:..(1—-—2)a: (5) 36-43, 79-86.
1094 1P Ashm. inv. 105/112(8) (ii): E) 51-2 vel A 81-2.
1095 I’ Ashm. inv. 102/143(6) (iii): (3 53-71, 74-6, 78, 85-6, marginalia 16 92, 97,
102. Transcribed by N. Litinas.
1096 P Ashm. inv. 21 313 29/D(18)b (1); <3 58-72 (661. 59).
1097 P Ashm. inv. 33 413 88/G(1-2)6 (1)1 0 74-8.
1098 P Ashm. inv. 19 28 76/F(6-7)6 (iii): 8 120-34, 147-61.
1099 P Ashm. inv. 34 413 77/G(1-~2)a: (3 138-61.
1101 P Ashm. inv. 23 313 1/Q(1-3)6 (ii--iii): 1-3 169-80.
1102 P Ashm. inv. 13 18 125/F(g) (ii): 0 196-223.
1103 P Ashm. inv. 123/81(6); <3 206-14.
1104 P Ashm. inv. 106/16(6) (ii-iii): <3 215-35, 258-72, K 115-25, 155-63 (366
scraps).
1105 P Ashm. inv. 127/24 (ii): (-3 237-69.
1107 P Ashm. inv. 17 28 56/D(g) (i-ii): (3 278-85.
1108 P Ashm. inv. 100/7(6) (ii-iii): 8 286-90.
1109 P Ashm. inv. 21 313 28/F(1-3)6 (1); <3 297-306.
1110 ")
Ashm. inv. 71/4(6); 13 358-79.
1111 "J Ashm. inv. 37 313 87/K(14)d (11-111); 8 416-20.
1113 T)
Ashm. inv. 35 4B 71/A(1-2)b= 8 438-54.
1117 '1')
1. Ashm. inv. 106/11(11): <3 544-59 (661. 548, 550-2, 558).
1118 _.)
Ashm. inv. 30 413 40/G(4-6)b: I 1-6, A 519-39.
1119 '3
Ashm. inv. 87/352(6); 1 1-3, 9-25.
1120 P Ashm. inv. 104/9(3) (i-ii): 14-33.
1122 P Ashm. inv. 49 5B 99/D(4—6)a: I 45-65 (and one fragment).
1123 P Ashm. inv. 30 413 37/A(1--3) (i-ii): I 49-62.
1124 P. Ashm. inv. 50 4B 30/C(1-3)a + e (ii): I 60-77 (ancl one fragment).
122 I.-4. The Papyri
12 G. I. C. Robertson, Five Oxyrhynchus Papyrifrom Iliad 12. M.Phil. thesis, Oxford 1993.
124 1.4. The Papyri
637-59, 662-5, 677-84, 686-704, 713-17, 735-171 18, )3 449-80, 483-7, 491-510.
Transcribed by M. ]. Anderson.
1343 P. Ashm inv. 37 38 87/K(14)b: o 584-601.
1344 P. Ashm inv. 3 1B 82/B(1)a (iVi’): O 586-604, 622-39.
1345 P. Ashm inv. 30 48 35/F(1--3)d (11-111); o 643-50.
1346 P. Ashm. inv. 38 38 85/A(3-5)3 (ii-iii): o 671-86.
1347 P. Ashm O inv. 103/90(c) (s. ii): O 691-743. Transcribed by A. Nodar.
1349 P. Ashm inv. 18 28 66/F(5)b (ii): 11 8-31.
1350 P. Ashm inv. 30 48 38/H(1-2)3 (111): 11 21-41.
1351 P. Ashm inv. 8 1B 186/C(a) (Byzantine): H 26-30, 57-61.
1352 P. Ashm inv. 4 18 69/8(6) (111); n 46-66.
126 1.4. The Papyri
1
1 1443 P. Ashm inv. 105/146(a) (i-ii): E 515-21.
1445 P. Ashm inv. 27 38 43/A(1-2)6 (1 8c-1 AD): 2 611-16.
1446 P. Ashm inv. 125/24(d) (i): T 1-5.
1447 P. Ashm inv. 34 4B 77/H(3-6)a (i): T 7-16.
1.
.1)
1448 P. Ashm inv. 34 4B 77/Q(5-6)c: T 21-7.
£1 .
4 1449 P. Ashm inv. 6 1B 8/C(b) (ii): T 31-47.
1?»
1451 Ashm inv. 57/108(3) (ii-iii): T 154-65.
1452 P. Ashm inv. 18 28 66/F(1)3 (iii): T 242-9, 275-81.
1453 P. Ashm inv. 63 6B 62/M(1-3)b (iii): T 309-19, 353-63.
I-1
13
310-24,327-37,356-81.
:51
1464 P. Ashm inv. 34 4B 78/E(2—3)a: T 240-61, 280-304.
1465 P. Ashm inv. 5 1B 38/F(j) (ii-iii): T 253-9.
1466 P. Ashm inv. 22 3B 21/N(1-2)a (ii-iii): T 258-68.
1467 P. Ashm inv. 39 3B 78/K(1)b: Y 286-9, (I) 124-57.
1469 P. Ashm inv. 82/5(b) (ii): I‘ 375-91.
1470 P. Ashm. inv. 14 1B 221/F(e) (Byzantine): T 393-456.
128 1.4. The Papyri
1471 7)
Ashm. inv 67 68 8/H(1)b, 10/c(1-4)6 (111): r 404-13, 437-67.
1472 7)
Ashm. inv 16 213 47/6(6) (11-111); 11> 1-16.
1473 7)
Ashm. inv 27 3B 38/I-{(2-3)a: <1) 29-72.
1474 7)
Ashm. inv 45 58 58/6(7):. (ii): 0 45-61, 80-103.
1475 7)
Ashm. inv 57 (top of box, no further inventory number; i-ii): (I) 68-74.
1477 7)
Ashm inv 87/333(8): <1) 81-9661 (+ scraps).
1479 7)
Ashm inv. 58/8(81) (ii): 0 253-71.
1480 7)
Ashm inv 101/ 108(b) (ii): <1) 273-84, 461-72. Transcribed by R. May.
1481 7)
Ashm inv 81 28 85/56(6) (ii-iii): <1» 283-93.
1483 7)
Ashm inv 75/72 (ii): <5 354-65.
7)
1484 Ashm inv 1 1B 211/B(u) (ii): (P 369-77.
1485 7)
Ashm inv. 47 58 43/F(9-10)6 (11-111); <1> 374-86.
1486 7)
Ashm inv. 13 18 125/H(b) (111); 0 392-406.
1487 7)
Ashm inv 27 38 41/](1-2)3 (111); <I> 403-8.
1488 7)
Ashm inv 100/173(8) (11-111); <8 425-9.
7)
1490 Ashm inv. 67 613 11/D(1): <13 460-71.
1491 7)
Ashm inv 31 48 10/L(1).».. (11-111); <1» 536-51.
7)
1492 Ashm inv 19 213 74/K(b)2 (1) 573-90.
1494 7)
Ashm inv 75/76(3) (1-11): X 27-43.
1495 Ashm inv 44 5B 61/E(1-3)a (iii): X 40--50 (and one scrap).
1496 VJF3 Ashm inv 5 18 38/1(6) (180-1 AD): X 67-81, 87-110, 117-34.
1497 7)
Ashm inv 72/24(6); X 93-103, 140-8.
1498 P Ashm inv 27 38 41/13(1-2)6 (ii-iii): X 100-15.
1499 P Ashm. inv. 44 5B 61/H(12--15)a: X 111-26, 145-6, 187--97, 207-10
295-301.
1500 P Ashm inv 93 Dec. 19/B(1) (i--ii): X 115-30.
1502 P Ashm inv 36 413 99/B(1-3)C, 100/H(4-5)a (ii): X 211-—-17, 282-8.
1504 P Ashm inv 46 513 51/E(6-7)a: X 240-59.
1505 P Ashm inv 101/97(0): X 327-58.
1507 P Ashm inv 37 4B 103/F(5)b (iii): X 354»-65 (and traces of P397-406 P).
1508 P Ashm inv 63 68 72/A(1-4)b (Byzantine): X 369-77, 412-19.
1509 P Ashm inv. 5 18 42/8(6) (111); X 369-84.
1510 P Ashm inv 106/125(6) (1); X 378-81.
1511 P Ashm inv 67 68 14/H(1-5)6 (111); X 380-98.
1514 P Ashm inv 102/28(6): \P 38-44.
1515 P Ashm inv 81 28 85/11(3) (vi): \P 65-82, 103-19.
1516 T) Ashm inv 10 18 161/G(f) (i-ii): W 101-7.
1517 P Ashm inv 14 18 208/G(6) (111); 111 180-5.
1518 "3 Ashm inv 105/41(6): ‘P 222-30, 232-45.
1519 7)
Ashm inv 103/73(8) (iii--iv): W 237-8, 245, 275-84.
1520 7)
Ashm inv 93 Dec. 29/E(1) (Byzantine): ‘P 457--65, 483-91.
1521 7)
Ashm inv 68 68 25/](6-7)3 (iv): \P 514-25, 558-71, 603-6.
1522 7)
Ashm inv 65 613 30/M(3-5)di ‘P 518-21, 524-41, 571-88, 603-7.
1523 7)
Ashm inv 15 2B 40/E(C) (iii): ‘P 539-50.
1524 7) Ashm inv 5 1B 57/H(g) (ii-iii): ‘P 593-600, more on verso (unidentified)
1525 7) Ashm inv 49 5B 97/D(1-2)a (iii): ‘P 614-18, 632--6 (and one scrap).
_., . , .ir, L4. The Papyri
I
l
1544
El
.9
2I;
.-3
1545 Wall inscription from the Via S. Basilio in Rome, edited by V. de Marco, Notz—
-i
zie degli Scrwi 1938 (xvi), 422-4 (iii): Q 171-5.
9
h1 P. Bonon. 6 (inv. 10a; Bibl. Univ. di Bologna no. 122820; Paclrz 1157; iii-iv):
hypothesis to A. Ed. F. Montanari, Arzagenrzesis 2 (1982), 273-84 = id. (1995),
87—95.
h2 P. Berol. inv. 5014 (Packz 1158; school exercise, Panopolis?, v): glossary to A
1-6, 8-12. Ed. U. Wilcken, SPAW (1887), 8181.
h3 P. Achmim 2 (Bibliotheque Nationale, Paris, Suppl. Gr. 1099, 2) (Packz 1159,
papyrus codex, Panopolis, iii-iv): glossary to A 1--21. Ed. U. Wilcken, SP/-1W
(1887), 818--18.
I14 P. Mich. inv. 1588 (i-ii): scholia minora to A 1-9. Ed. T. Renner, HSCP 83
(1979), 313-21. _
h5 P. Oxy. inv. 8 1B 189/E (1--3)(a) (papyrus codex, iii): glossary to A 4-18. Ed.
F. c. Philips, 11., BASP 8 (1971), 9111.
h6 P. Oslo 2. 12 (Packz 1160; school exercise, Theadelphia, ii): glossary to A 5-24.
h7 P.I.F.A.O. inv. 105 (Paclrz 1161; iii): glossary to A 10-12. Ed. Schwartz,
BIFAO 54 (1954), 701.
h8 P. Oxy. inv. 44 5B 61/G(4-6)a: scholia to A 23-40(?).
h9 P. Hamb. 3. 199 (inv. 81) (ii): scholia to A 381. (Mythographus Homericus).13
C1. W. Luppe ZPE 56 (1984), 31--2.
l‘l1O P. Cair. inv. ]. E. 45612 (v-vi): glossary to A 43-5, 48-50. Ed. C. Gallazzi, ZPE
84 (1988), 1-9.
h11a Bodl. Libr. Gr. class e.44(P) (Packz 1217; i): Apollonius Sophista. Ed.
E. W. B. Nicholson, CR 11 (1911), 390--93.
hill) P. Mich. inv. 5451a (Karanis, ii): Apollonius Sophista. Ed. T. Renner, HSCP
83 (1979), 321--31.
h11c P. Oxy. 2517 (ii): Homeric lexicon (Apollonius Sophista? 9o:up0iCe1v-Bpfivvg).
hlld P. Oxy., inv. unstated (ii): Homeric lexicon (Apollonius Sophista?). Ed.
,1. W. Schumaker, BASP 7 (1970), 59--65.
1112 P. Qxy. inv. 46 5B 51/E(4-5)a: scholia to A 53-75(?).
h13 Oxy. 2405 (Packz 1162; ii-iii): glossary to A 58--128.
h14 541°‘ Narm. inv. 69.43 (Madinet Madi, ii-iii): glossary to A 731., 801. Ed. C. Gal-
lazzi, Ann. Sew. 69 (1983), 189-93.
1115 P. Turner 13 = P. Strasb. inv. Gr. 33 + 39 + 40 + 41 (Packz 1163; late iii’): glos-
sary to A 83-361. Sutton divided this between h15 and h17.
h15a P. Berol. inv. 21306: glosses to A 97-9. Ed. W. Brashear, Proceedings of the
XX International Congress of Papyrology (Copenhagen 1994), 285.
h16 P. Koln 1. 46 (inv. 1755) (ii): paraphrase 01 A 12811., T 24511.
[h17: see h15.]
*3 For the so-called Mythographus Homericus see B. Kramer in P. Hamb. III, p. 199; M. W.
Haslam, BASP 27 (1990), 31-6; F. Montanari in ]. G. J. Abbenes et al. (ed.), Gree/e Literary
Theory after Aristotle (Amsterdam 1995), 135-72.
'€
4-.4s -.w, . e
$
2>
I.4. The Papyri 131
'5
I
h18 P. Mich. inv. 1585 (Packz 1222; ii): glossary to A 170-80. Ed. A. Henrichs, ZPE
7(19711 148f£
h19 P. Palau Rib. inv. 147 (ii): scholia minora to A 191-416. Ed. S. Daris, Stud. Pap.
2
5
13(19741 7420. -
3.
h2O P. Oxy. 418 (Packz 1164; i-ii): scholia on A 2631., 399 (Mythographus Homeri-
cus).
i
§< h21 P. Schubart 2 (-P. Berol. inv. 7501) (Paclcz 1165; iii): glossary to A 266-72.
h22 P. Oxy. 3237 (early iii): glossary_to A 302-23.
h23 P. Kriln inv. 2281 (Oxyrhynchusi, ii): glossary to A 318-416, 512-610. Ed.
A. Henrichs, ZPE 7 (1971), 22911.
1
h24 BKT 5. 1. 6 (inv. 10577) (Packz 1166; ii): glossary to A 338-49. Reedited by A.
Henrichs, ZPE 7 (1971), 25211.
h25 P. Oxy. 3238 (early iii): glossary to A 405(?)-538, B 385-93.
h26 P. Ant. 2. 70 (Paclrz 1167; papyrus codex, iii): glossary to A 464--9, 480-6.
1
h27 Brussels, Musées Royaux inv. E 7162 (Packz 1224; i-ii): treatise on Homeric cos»
:
mology. Ed. M. Hombert and C. Préaux, Mél. Boisacq I (Brussels 1937), 493-7;
cf. M. S. Funghi, La Parola del Passato 38 (1983), 11-19.
h2 8 P. Mil. Vogl. 120 (inv. 613) (Packz 1168; ii): glossary to A 525-30, 536--51. Ed.
M. Vandoni, Acme 14 (1961), 2381.; A. Henrichs, ZPE 7 (1971), 2551.
h29 P. Erl. 5 (inv. 3) (Packi 1169; ii): paraphrase o1 A 528-46.
h3O P. Rain. inv. 26221 (papyrus codex, vi): A 6011., 6091. with paraphrase. Ed. H.
Harrauer, Codices Marmscripti 1 (1975), 741.; G. Bastianini, Prometheus 6
(19801 8348.
1
[h3l Sutton is a duplication 01 h1.]
i
h32 P. Ryl. 1. 26 (Packz 1216; Oxyrhynchus, i): Homeric glosses by Apion.
h33 P. Rein. inv. 2088 (Paclrz 1170; papyrus codex, iv-v): scholia minora to B 45-57.
Mentioned by P. Collart, Atti del IV Congresso Internazionale di Papirologia
$1” (Milan, 1936, repr. 1976), 71.
1134 P. Hamb. inv. 7336 (ii): glossary to B 61-222. Ed. T. Vlachodimitrios, ZPE 11
1 (19731 8548.
h35 P. Kéiln inv. 53 (iii?): glossary to B 93--104. Ed. A. Henrichs, ZPE 7 (1971), 2571.
has P. Oxy. 3832 (ii): scholia to B 201-18. C1. Montanari (1995), 131-7.
h37 P. I-Iarr. 10 (inv. 1731) (Pacltz 2628; ii): scholia to B 381-98. Identified and
E reedited by M. Gronewald, ZPE 46 (1982), 951.
h38 Cair. Masp. 3. 67331 (Packz 1171; papyrus codex, AntinoeP, Byzantine): scho-
1ia t0 B 414-21, E 412--77, 552-603.
h39 P.S.I. 1276 (Pacltz 1172; i BC): B 617-70 with interlinear paraphrase.
h40 P. Oxy. 1086 (P8.Cl{2 1173, pap. II Erbse; i BC): commentary on B 751--827. Same
hand and format as h129.
h41 P. Berol. inv. 11518 (i): glossary to I" 148, 433-61, A 2-26, 47-76. Ed. W. Miiller,
Forsc/21/mgerz und Bericbte der staatlic/aen Museere 224 Berlin 10 (1968), 113-18.
h42 BKT 5. 1. 6 (inv. 9960) (Packz 1174; i-ii): commentary on P 59 = Z 333. W.
Luppe, APF 44 (1998), 2131.
4
h43 P. Oxy. 412 (Paclcz 53; iii): ]ulius Africanus’ Kestoi quoting an extended hex-
ameter incantation claimed to have stood in manuscripts in Palestine, Caria, and
Rome.
132 I.4. The Papyri
h44 A. Carlini, Papiri lettemri greci (Pisa 1978), no. 22 (P. Pisa inv.1) (vi): hypothe-
sis to I". C1. F. Montanari, Anagermesis 2 (1982), 273-84 = id. (1995), 87-95.
1145 P. Dura 3 (inv. D Pg. 33) (Packz 2135; Dura Europus, early ii): glossary to A
302-16. Identified and reedited by M. Gronewald, ZPE 44 (1981), 1771.
h46 Papyrus edited by S. Daris, Stndi Triestini di Arztichitti in onore di Lmlgia Achil-
lea Stella (Trieste 1975), 463-9 (pap. IIa Erbse [VII 284]) (ii): scholia minora to
A 164-9.
h47 P. Ryl. 1. 24 (Packz 1175, pap. III Erbse; i): scholia minora to A 3061.
h48 Bodleian Library, Greek inscription 3017 (Paclcz 1176; school exercise, wooden
tablet, ii-iii): paraphrase of A 349--73 with explanations. Ed. M. Hombert and
C. Préaux, Mélanges Gregoire: Anmmires de Plnstitut de Philologie et d’H£sto1're
Orientales et Slaves 11 (Brussels 1951), 161-8.
1149 P. Rain. (inv. unstated; Packz 1177; Fayfirn?): paraphrase of A. Mentioned at Phi-
lologische Anzeiger 14 (1884), 414; now lost.
h50 P. Ryl. 3. 537 (Packz 1178; early iv): glossary to E 5--11, 37-53.
1151 P. Strasb. inv. Gr. 1015 (Packz 1179; school exercise, ii): glossary to E 88-254.
Ed. O. Plasberg, APF 2 (1903), 196--206.
h52 P. Harr. 2. 177 (inv. 46) (iii): scholia minora to E 158-256.
I153 P. Berol. inv. 11636 (Packz 1180; school exercise, wooden tablet, iii-iv): E
265-313 (= no. 181 above) with a glossary to 287-312. Ed. G. Plaumann, Amt-
Ziche Berichte aus den /eohiglichen Kzmstsammlrmgen 1913:10, 220; L. M. Raffa-
elli, APF 38 (1990), 5--12.
h54 P. Amst. 1. 5 (inv. 51) (iii): glossary to E 287-315.
h55 P. Oxy. 3158 (ii-iii): scholia minora (virtually a glossary) to E 655-822.
h56 British Library Pap. 113 (14b) (P. Lit. Lond. 177; Paclcz 1181; school exercise,
iv): scholia minora to E 670-710.
h57 P. Haun. 1. 3 (inv. 314) (Packz 1182; Fayf1m?, iii): scholia minora to Z 1-48.
h58 P. Cair. inv. 60566 (Packz 1184, pap. V Erbse; Oxyrhynchus, ii): scholia
minora to Z 236, 252-88. Ed. W. G. Waddell, Mélanges Maspero I (Cairo 1934),
148-51.
1159 P. Strasb. inv. G 2374 (Packz 1185; school exercise, iii BC): Z 448-55 quoted in
a prose summary. Ed. N. Lewis, Et. Pap. 3 (1936), 4611.; S. West (1967), 731.
Formerly listed as p317.
h60 P.S.I. 135 (Paclrz 1183; i--ii): commentary on Z. C1. A. Carlini, Zetesis (Festschr.
E. de Strycker, Antwerp-Utrecht 1973), 484-6.
h61 P. Oxy. 3159 (iii): hypothesis and glossary to H 411.
1162 P. Oxy. 1087 (Packz 1186, pap. VI Erbse; late i BC): scholia to H 75-83.
1163 P. Mich. inv. 920a-b (ii-iii): hypothesis to H, G). Ed. ]. ]. O’Hara, ZPE 56 (1984),
1--9, cf. 59 (1985), 149.
1164 P. Amst. 1. 6 (inv. 82) (iii): glossary to (5) 1-11.
h65 P. Berol. inv. 11634 (Fayum, second half of ii): ‘Schulpriiparation’ to 6) 361-563, l
I 2-64. Ed. G. Poethke, Forschzmgen and Berichte der staatlichen Mnseerz zu
Berlin 8 (1987), 105-10.
[h66: erroneous entry combining elements from h64 and h67.]
h67 P.S.I. inv. 1733 (Packz 1187; papyrus codex, Tebtunis, vii): glossary to I 58-93.
Ed. M. Manfredi, SIFC. n.s. 27-28 (1956), 501.
I.4. The Papyri 133
h68 P. Washington Univ. inv. 217 (pap. VIIa Erbse; Oxyrhynchus, ii BC?): com-
mentary on I 129( P)-47. Ed. Z. L. Pacltman, BASP 10 (1973), 5311.; M. W. Has-
lam, BASP 22 (1985), 97-100. cf. A. Wouters, ZPE 21 (1978), 271--3.
h69 P. Lit. Lond. 142 (Brit. Libr. inv. 160.50) (Packi 1188, pap. VII Erbse; ii): com-
mentary on I 447. Mythographus Homericus, same roll as h115. Identified and
reedited by R. Pfeiffer, Phil. 92 (1937), 16-18.
h70 Papyrus in the.collection of E. von Scherling, Leiden (inv. G 99) (Packz 1189;
ii): glossary to I 454-68. Ed. B. A. van Groningen, Mnem. 35 (1937), 62-8.
h71 P. Mil. Vogl. 3. 119 (inv. 1031) (i): scholia to K 305-46.
1172 P. Mich. inv. 1315 (Packi 1190; ii): paraphrase of K-P. Cf. ]. G. Winter, Life
and Letters in the Papyri (Ann Arbor 1933), 195.
1173 BKT 5. 1. 6 (inv. 10511) (Packz 1191; wooden tablet, ii): glossary to A 136-263
(with 26 out of order). Reedited by A. Calderini, Aegyptus 2 (1921), 3081.
h74 P. Arnh. 2. 19 (Packz 1192; parchment codex, vii): glossary to A 558-601.
h75 P. Oxy. 574 (Packz 1193, school exercise, ii): narrative based on A 57511. C1. F.
Montanari, ZPE 48 (1982), 89-92 = id. (1995), 97-102; ]. ]. O’Hara, ZPE 59
(1985), 35.
1176 P. Iand. 1. 2 (Packz 1194, pap. VIII Erbse; i BC): commentary on A 677-754.
Reedited by P. A. Kuhlmann, Die Giessener literarischen Papyri und die Cara-
calia-Erlasse (Giessen 1994), 45-54.
1177 P. Med. inv. 72.13 (ii): scholia minora to A 321(?)-53. Ed. S. Strassi, Aegyptus
58 (1978), 1151.
1178 P. Ryl. 3. 536 (Packz 1195; iii): scholia minora to N 198-227, 317-24, 415-35,
530-62. F. Montanari, Mosaico. Studi in onore di Umberto Alhirzi (Genova
1993), 135~48 = id. (1995), 137-48.
1179 P. Oxy. 3003 (ii): Mythographus Homericus relating to N 302, 459, E 319, O 229.
h8O BKT 5. 1. 6 (inv. 10510) (Packz 1196; wooden tablet, ii): glossary to N 63411.
1181 P. Mil. Vogl. 1. 19 (Pack: 1197, pap. IX Erbse; Tebtunis, ii): Apollodorus,
Znrfilltxror vpallporrmoi on E (subscriptio only).
h82 BKT 5. 1. 6 (inv. 10508) (Paclcz 1198; school exercise, wooden tablet, ii): glos-
sary to E 227-521. Reedited by A. Calderini, Aegyptus 2 (1921), 3061.
h82a P. Mich. inv. 1206 (iii-iv): explanations 01 E 316-48. Ed. W. Luppe, ZPE 93
(1992), 183-5.
1183 BKT 5. 1. 6 (inv. 10509) (Packz 1199; school exercise, wooden tablet, ii): glos-
sary to O 17-180. Reedited by A. Calderini, Aegypttts 2 (1921), 3071.
h84 P. Rain. inv. 39940 (ii): glossary to O 320-633. Ed. P. Sijpesteijn and
K. A. Worp, ZPE 15 (1974), 15311.
h85 P. Ant. 3. 150 (ii-iii): glossary to II 100(?)-50.
h86 P. K6111 4. 180 (inv. 7891) (papyrus codex, vi-vii): paraphrase of U 394-418,
459-71.
h87 P. Cair. inv. 60565 (Packz 1200; i-ii): collection of Homeric similes. Ed.
W. G. Waddell, Mélanges Maspero I (Cairo 1934), 145-8.
h88 P. Mich. inv. 48320 (Karanis, ii-i BC): anthology or summary 01 E-T. Ed.
T. Renner, HSCP 83 (1979), 331-7.
1189 P. Oxy. 2397 (Packz 1201, pap. X Erbse; i): commentary on P 4(?), 36, 112(?),
201-10(?), 481 £f.(?), 520(2), 893-5.
134 I.4. The Papyri
i (1993), 9-11.
1;" h116 P. Mich. inv. 1206 (iii-iv): scholia minora to E 316—48. W. Luppe, ZPE 93
;g‘-.
'2-.
(1992), 163_5.
h117 P. Michaelides 5 (Packz 1612; early iii BC): F 425a—9a in anthology. Reedited
by S. West (1967), 70-3.
h118 P. Berol. inv. 16705 and 21253 (iv-v): two fragments of Apollonius Sophista
E'r-- for 0:. A. Henrichs and W. Miiller in Collectnnea Papyrologica. Texts PnZ9lish-
ed in Honor ofH. C. Youtie (Bonn 1976), 27-52; Ioannidou (1996), no. 157.
:.-; .-
[h119 P. Yale 2. 124 (iii BC): perhaps an unknown hexarneter p0em.]
l} h12O P. Yale 2. 125 (inv. 1245; iii--iv): glossary to A 66-74.
:'.l3 '
h121 P. Yale 2. 126 (inv. 1544; i): scholia minora to A 189——223.
'1
§i h122 P. Oxy. 4096 (ii): Mythographus Homericus. M. W. Haslam, ZPE 110 (1996),
iii,i,
115—-17; W. Luppe, ZPE 112 (1996), 25-33; M. van Rossum—Steenbeek, ib.
34-6.
[h123 is Philodemus Poet. and does not belong in this list.]
i
h124 P. Berol. inv. 21306 (vii): glossary to A 97f. Ed. W. Brashear and A. Biilow-
-'2':
Jacobsen, Proceedings oft/ae XX International Congress of Papyrology (Copen-
hagen 1994), 285.
h125= P. Hamb. 2. 136 (inv. 665) (Packz 633; first half of iii BC): B 101-9 quoted in
l
a prose text. Formerly listed as p459; reedited (but incorrectly designated p457)
by S. West (1967), 38f.; cf. Erbse I. xliii.
.<_
h126 P. Mich. inv. 2720 (v—-vi): scholia minora to F 388-416, A 508-18, E 197—~446,
;;i_-.._._
._§l _.
528-41, 620-6, 660-1, 763——6, 782—5. Ed. Schwendner (1988), 3l——98.
-:11:
h127 P. Schub. 21 (P. Berol. inv. 13930) (Packz 1203; parchment codex, v): Mytho-
graphus Homericus on T 326, T 53(?), 146ff. R. Merkelbach, APF 16 (1958),
:17f.; W. Luppe, APF 31 (1935), 11.
i h128 P. Berol. inv. 17598 (i): A 1 with brief perioche. W. Luppe, APP 44 (1998),
l 2l4~15.
i h129 P. Oxy. 4451 (i BC): commentary on A 56--8. Same hand and format as h40.
s
§
h13O P. Oxy. 4452 (ii): commentary on T 347—-57, 383-416.
h131 P. Ashm. inv. 72/ 19(a) (Oxyrhynchus, ii): scholia minora to B 24-40. Tran-
scribed by]. Spooner.
11132 P. Ashm. inv. 51 4B 18/H(1-—3)a (Oxyrhynchus, iii): scholia minora to B
50—109. Transcribed by J. Spooner.
.1»
pi=.[-
h133 P. Ashm. inv. 19 2B 82/K(a) (Oxyrhynchus, iv): scholia minora to B 214-27.
l..
Transcribed by ]. Spooner.
h134 P. Ashm. inv. 70/39(a) (Oxyrhynchus, iii): scholia minora to B 277--93, 307--18.
=1?- Transcribed by ]. Spooner.
. :
.-,1
_i
I-‘II
I
i
136 1.4. The Papyri
W1 P. Lit. Lond. 182 = Brit. Libr. inv. 126 verso (Packz 1539; iii—iv): fragment of
Tryphon, Téxvn vpotpp0t'n1<ri, quoting A 18-20 (<bi?\nv), B 760, K 342, O
208f., (W 461f.?). Ed. Kenyon (1891), 109——16.
W2 P. Berol. inv. 9873 (BGU 4. 1204 = P. Mag. 22ab; Pack: 573; iv-v): Homero-
manteion quoting A 75, 96, B 548, F 40, A 141, P 714.
W3 P.S.I. 1293 (Packz 2707; Oxyrhynchus, ii--iii): Writing exercise quoting A 75 f.,
B 1, 3090), E 215, H 151., 264, A 596.
W4 P. Oxy. 853 (Packz 1536; ii): commentary on Thucydides quoting A 117, B 504,
766, 1" 1, 6 297-3000), 559, H 467, P 5ss?, 2 29s, 492, X 55 (61 r1 s162), n 294.
W5 British Library Pap. 121 (P. Mag. 7. 1--148; Packz 552; iii—-iv): I_IOl'I'l€‘1'OITl3I11I€1"
on. Formerly listed as p101.
W6 P.S.I. 1192 (Packz 1467; Oxyrhynchus, i): Sophocles, O.T. 179—200 with scho-
lia on 196 quoting A 314.
W7 P. Koln 3. 127 (inv. 2693) (ii-iii): Homeric cento employing B 514, F 278(?), I
532 (P), Z 482, and possibly K 83 = 386 and other lines from the Iliad.
W8 P. Flor. 2. 259 (Pack: 623; Writing exercise?, margin of private letter, Theadel-
phia, ii-111): B 1f.
W9 P. Vat. Gr. 11 (Packz 455; early iii): Favorinus, Tl'8[)\1 <i>uyfi<;, quoting B 128, 595,
F 471., A 171, Z 311, E 308, X 213. Ed. G. Vitelli and M.1\1orsa, 1ZPapiro Vati-
cano Greco 11 (Vatican City 1931).
W10 Gizeh Museum, inv. 9267 (limestone inscription, Alexandria, late i BC):
inscriptions with reliefs and epigrams, quoting B 204f., 412, E 31, O 187-91.
Ed. ]. G. Milne, ]HS 21 (1901), 2869990.
[W11 = p372.]
W12 P. Bon. 24 b2 (Packz 645; ii--iii): Homeromanteion quoting B 298, 325, F 51, A
26, E 899, Z 432, I 43, T 107, X 79. Ed. O. Montevecchi, Aegyptas 27 (1947),
1831.
W13 BKT 5. 2. 140--2 (inv. 9734; Packz 2170; iii): treatise on metrics quoting B 698,
710.
W14 P. Hamb. 2. 137 (inv. 657; Packg 1221; mid iii BC): prose treatise quoting B
848a, E 746, I 2301., N 505, E 3491. Reedited by S. West (1967), 59-62.
W15 P. Berol. inv. 13419 (Packz 1357: iii—iv): commentary on Pindar quoting F 196.
Ed. U. von Wi1amowitz—Moellendorff, SPAW 1918, 749f.
1.4. The Papyri 137
W16 P. Lit. Lond. 183 (inv. 885; Packz 2291; ii?): treatise on Attic usage (Phryni-
chus ?) quoting F 229, H 795.
W17 P. Osl. 2. 13 (Pacltz 2148; Theadelphia?, ii): grammatical discussion quoting A
139, A 388. 4
[W18 = ‘Hes.’ fr. 31 and does not belong here.]
W19 P. Hib. 173 (Brit. Libr. inv. 2946) (Packz 136; Hibeh, iii BC): commentary on
Archilochus quoting A 182, E 130, E 66. Reedited by S. R. Slings, ZPE 79
(1989), 1-8.
W20 P. Oxy. 410 (Packz 2295; ii): Doric treatise on oratory quoting A 443, I 381,
385, 389, 404.
W21 P. Hamb. 2. 128 (inv. 650 + 650.16) (Packz 1502; ii BC): Theophrastus, Trepi
Aéiewq (P, I1 612 Fortenbaugh) quoting A 485.
W22 P. Wiirzb. 2 (inv. 19; Packz 2297; ii): treatise on literary tropes (by Seius Nio-
anor?) quoting A 521f., T 222-4, Y 59f.
W23 P. Rein. 2. 83 (P. Sorb. inv. 2093) (Packz 2571; late ii): Stoic treatise or dia-
logue on resignation, quoting E 385 f. + 392 f. + 395, X 117f.
W24 British Library Pap. 2239 (P. Lit. Lond. 193; Packz 2524; ii): sophistic discour-
ses quoting E 531 = O 563.
W25 British Library Pap. 734 (P. Lit. Lond. 175; Pacltz 1223;_iii): allegorical interpre-
tation of Homer quoting Z 235f., H 161,1 169f., O 189(?), 193(?), ‘I’ 252, 648f.
W26 P. Mil. Vogl. 1. 17 (Packz 89; ii): commentary on Antimachus of Colophon
(pp. 79ff. Wyss) quoting Z 268 (= E 402, 1.|J 48), K 394.
[W27 is a medieval manuscript: S. G. Daitz The Sc/aolia in the ferasalem Palimpsest of
Euripides (Heidelberg 1979), With at p.104. 11f. a quotation of I 146 identified
by Diggle, ZPE 40 (1980), 20.]
W28 P. Harr. 1 (inv. 3, Pack’ 1325; iii): Musonius Rufus, fr. 15A Hense quoting I
323f. Identified and reedited by]. E. Powell, APF 12 (1937), 175-8.
W29 P. Oxy. 2429 (Packz 362; ii): commentary on Epicharmus quoting K 511.
W30 P. Oxy. 2262 (Packz 204; ii): scholia on Callimachus’ /-letia (II 101. 27 Pf.) quo-
ting A 21.
W31 British Library Pap. 131(2r) (P. Lit. Lond. 181; Pack?’ 197; Meir, i): sch. Lond.
on Call. fr. 1. 45 quoting N 64.
W32 P.S.I. 1000 (Packz 2463; Oxyrhynchus, ii): Mythographus Homericus quoting
N 217 or E 116. Cf. L. Salvadori Baldascino, Civilta Classica e Cristiana 9
(1988), 259—~62.
W33 P. Berol. inv. 11521 (Packz 200; ii): comm. Berol. on Call. fr. 7. 29 quoting N
685. Ed. U. von WilamoWitz—Moellendorff, SP/1W 1912, 544»-7.
W34 P. Berol. inv. 13876 (Packz 2436; ii): mime quoting H 1, E 112 = T 65. Ed.
G. Manteuffel, De Opascalis Graecis Aegypti e Papyris, Ostracis Lapidibasqae
Collectis (Warsaw 1930), no. 16.
W35 P. Oxy. 3001 (ii): cento Using Tl 93f., P 32 = T 198, X 194, 197, ‘P 106, Q 486.
W36 BKT1 (inv. 9780) (Pacltz 339; I-Iermupolis, ii): Didymus on Demosthenes, quo-
ting Z 56 = 437.
W37 P. Vars. 2 (Packz 2435; ii): mime quoting (I) 198f.
W38 P. Derveni (iv BC): commentary on an Orphic theogony, quoting Q 527f. as
Orpheus.
138 1.4. The Papyri
W39 Paris. Suppl. gr. 574 (P. Mag. 4; Packz 746; Thebes P, iii—iv): magical text quot—-
ing E 385, <9 424, K 193, 521, 564, 572. Formerly listed as P264.
W40 P. Oxy. 3831 (miniature codex, iii-iv): Homeromanteion quoting A 216, 524,
B 247, E 124, 256, o 154, 176, 414, I 378, 482, K 159, 324, 401, TI 514-16, <5
lll,Q 92, 369.
W41 P. Rain. inv. 26765 (ii-iii): romance? quoting N 278. R. Pintaudi in M. Capas-
so (ed.), Papiri letterari greci e /atini (Galatina 1992), 289f.
W42 P. Osl. 3. 177: )(pe'iou of Diogenes, quoting B 24. 1. Gallo, Frammenti biogra-
fia da papiri (R6646, 1975-80), II 363.
W43 P. Berol. inv. 21116v: prose With quotation of Z 483, 485. Ioannidou (1996).
W44 P. Mag. 6 (Brit. Libr. inv. 47; Packz 568; ii—iii): magical text quoting A 37--8
and 39-41. Formerly listed as p113.
W45 P. Ien. inv. 659 (iii): quotation of I 32~—6 in prose text. Formerly listed as p599.
W46 P. Ashm. inv. 30 4B 41/D(2-3)b (i): quotation of Z 68--70 in prose text.
W47 P. Ashm. inv. 106/11(c): quotation of X 319-20, 447--51(—52?) in prose text(?).
5. The Early Medieval Transmission
Manascripts used
The two oldest medieval sources, both fragmentary, are:
X (p568 Sutton) = Sinai, St. Catherine’s Monastery, inv. MF26. Twelve or
thirteen folios of a codex With interlinear prose paraphrase, covering
A 319-38, 477--96, 608-13 15, B 115-42, A 50--69, 111-29, 345——90, 458~—-78,
E 105--27, 169——89, 626-50. Described by I... Politis, Scriptoriam 34 (1980)
14 With pl. 8b, who dates it early in the second half of the ninth century;
photographs of two leaves in H. Charlesworth, Biblical Archaeologist 43
(1980), 26--43; additional information (With references to further imminent
publications) in M. Apthorp, ZPE 127 (1999), 141—-8.
Z (Vei Allen) = Rome, Biblioteca Nazionale gr. 6 (A-—Z 373) + Madrid,
Biblioteca Nacional 4626 (H 89-Q). Mid to late ninth century (Wilson
[1983], 85). This is the oldest manuscript of the D scholia. It contains the
important Prolegomena that are the source of Osann’s famous Anecdotum
and Wilamowitz’s ‘Vita Romana’, after which come the scholia, with lem-
mata in a left-hand column and the scholia on the right. At first the lem-
mata are so abundant as almost to make a continuous text, and as far as A
171 a second hand has actually added all the missing Words. Later they
become somewhat sparser. From the beginning of the Madrid portion to
the end of H the manuscript has a complete text in the left column (each
verse being divided between two or three lines) With the scholia to the
right. Accentuation is erratic. The hypotheses to each rhapsody are in an
elegant sloping uncial very similar to that of Paris. gr. 510 (R. Barbour,
Greek Literary Hands A.D. 400--1600 [Oxford 1981], no. 5), which is
dated between 879 and 883. For further description and references see
Allen (1931), 52f; I-I. van Thiel, ZPE 132 (2000), 8~——10, 25.
From the tenth century we have two largely complete manuscripts, though
certain lost portions have been supplied at later times. These are:
A (A Allen) -= Venice, Biblioteca Marciana, gr. 822 (olim 454), often re-
ferred to as Venetus A. Beautifully written and meticulously corrected,
140 1.5. The Early Medieval Transmission
1 D. Comparetti, Horneri Ilias cam sciaoliis. Codex Venetus A Marcianas 454 ploototypice edi-
tas, Leiden 1901.
2 Mnemosyne 32 (1904), 447-50.
3 The tWo—line deficit on f. 41‘ is better explained by a deletion of B 855al2 in the exemplar
than by one of 848a and 854a, as van Leeuwen supposes, and the one-line surplus on f. 77“ is
better explained by marginal status of E 783 than of E 786. He is incorrect in stating that f. 106’
contains 26 lines: it has the normal 25.
4 Cf. Allen (1931), 141. I have not thought it necessary to follow Allen and van Thiel in
distinguishing this portion of D as ‘d’.
1.5. The Early Medieval Transmission 141
commentary and are related to the scholia found under the name of
Ammonius in a second-century papyrus (P. Oxy. 221). La Roche (1866),
474; Allen (1931), 17, 141; Erbse (1969--88), I Xxi f.
I have made more sporadic use of the following:
H (Vi5 Allen) = Vienna, Osterreichische Nationalbibliothek, phil. gr. 117.
Thirteenth century. ‘P 647-Q 804 are supplied by a fifteenth-century hand.
La Roche (1866), 473; Allen (1931), 54.
M (M1 Allen) = Milan, Biblioteca Ambrosiana, A 181 sup. (74). Thirteenth
century. Scholia (recension h). E.‘ 289411 305 are missing. La Roche (1866),
464; Allen (1931), 23.
N (U4 Allen) = Venice, Biblioteca Marciana, gr. 841 (olim 458).
Twelfth—thirteenth century. Scholia (recension h). Contains only E 419--Q
804. Allen (1931), 41.
O (O8 Allen) = Oxford, Bodleian Library, New College 298. Thirteenth
century. South Italian, probably from Otranto: Wilson (1983), 227--8, 268.
I-Iistorically linked to the Vindobonensis phil. gr. 49 (W Ludwich, Vi3
Allen), see Wilson 228. The Catalogue of Ships (B 494-877) is omitted, and
the page containing N 63--214 has fallen out. Small writing in two columns
with ornamental initials, many interlinear glosses in a minute hand, some
marginal scholia; yp0i<1)sT0tt-variants added by the glossator. Allen (1931),
31.
P (P" Allen) = Paris, Bibliotheque Nationale, anciens fonds, grec 2766.
Fourteenth century. Scholia (recension h). La Roche (1866), 471; Allen
(1931), 35.
V (V1 Allen) = Vaticanus gr. 26. Thirteenth century. A 26--53 is supplied
by a fifteenth-century hand. Allen (1931), 43.
Manuscript relationships
The lemmata in Z reflect a text independent of the main medieval tradi—
tion, yielding a number of good or ancient readings that are found in few
or no other copies. Examples are A 173 éékoctal, 404 fiinv, 424 1<O£T0t
50610: and ‘érrovtou, 464 o1r7\0'ty)(vcx noiootvto, B 801 rrpofi éiotv, 872
rrékeuov Kiev, F 10 fiiits, 11 dueivwv, 51 Katncbein, 212 Ecbonvov, A 93 15’
div, 213 'é?u<ev, 327 Hetedfiio, 344 éoormilwuev, 517 uoipoz Tréonoe, E 534
Aivsiw, Z 18 poi 0'1, H 113 Toiitov, N 300 "r0c7\0ci<bpova, P 572 ixowciou.
Many of these were evidently inherent in the tradition of the D scholia,
but the lemmata do not always coincide with the reading presupposed in
the scholia.
144 I.5. The Early Medieval Transmission
6 In my first volume I occasionally used Q (rather than Q"') even where one or two of these
manuscripts had a discrepant reading, taking the near—consensus to signify a united tradition. A
user justly criticized this practice as confusing, and in the second volume I have more consis—
tently used Q"'.
7 His readings are often isolated in the tradition, for better or for worse. He has unique good
ones at I 394, 397, A 564(?), M 218, 412, P 148, T 259, (I) 191, 611, X 42; with papyrus support
also at X 69; less good ones at A 304, 350, 432, 434, B 258, 266, 682, 751, F 326, 406, A 17
(twice), 282, Z 113, 330, 475, H 41, 230, 304, o 311, 337, I 66, 214, 225, K 127, 397, N 359(A1~**),
O Z52, II 5, 53, 106, 504, F 505, )1 171, 485, 568, T 391, T 156, 303, (D 84, 131, X 431, 475, ‘P
464, Q ZOO.
3 There are further places where a corruption in Q is shared by all other available ancient
sources and may go back to pre-Alexandrian times, e.g. B 291 (i(V¥I}BéVT(X 1/6600011, F 109 5’ (3
yéptov, 373 fiporto xfioog (= Z 165), Z 291 émrrmbg, 409 K(XT(11<T0(V6OD0l, H 434 iiypsro Aaég, I
414'i1<u>110n, A 62 o1’57uog oiorrjp, 413 nerd ocbioi, P 727 uizv voip re, T 229 pnvpivog.
I.5. The Early Medieval Transmission 145
There are many places where the copyist has deliberately chosen to follow
Aristarchus’ or some other ancient doctrine that he found reported in the
scholia?
The same thing happens especially in the family known as la, which goes
back to a source dating from the eleventh century, if not the tenth.‘° It is
represented by N'M P, as well by various further manuscripts of the thir-
teenth to fifteenth centuries, such as Allen’s V6 V15 U2 V3 V19 and others;
at any rate they drew from it some readings and its particular version of
the scholia. The la manuscripts evidently represent a scholarly recension.
The man responsible — it is a pity that we cannot identify him -- quarried
the scholia for Aristarchean and other ancient readings, which he either
put in the text or noted as marginal variants.
The /J family has long attracted attention for the quantity of its ancient
readings. But as most of them were obtained in this way, we should be-
ware of forming an exaggerated opinion of its value as a representative of
the direct tradition. For this reason, and because a single h reading can
often not be identified in them, I have not cited N M P regularly but only
where their testimony seemed of interest.
Nevertheless, this recension remains remarkable for the number of its
ancient readings that were not to be found in the scholia. A number of
them turn up in papyri, e.g. B 202 01506 Trot’, A 15 0’ om., 283 uev Toiag,
318 uév Kev (With F), E 92 0’, Z88 the first Y’ 0m., 821 Y’ 0m., H 370 év
T87\é8UOl, K 493 eufiaivovreg (with W), A 277 Tffip, 525 éivfipecg, M 270 71011;
(with V), N 254 Sovpudtvtbg, 256 1<7\10in<i>1, 329 d1q>i1<o1T0, 587 15101, 623
>u.oi3rioeo0e, E 112 vetérepog (with R), 429 cisipovreg, O 700 ¢>e13E0100011
(with F“W), P 365 div’ iituhov, 541 (iv 0’, Z 39 01001 Fjv (with GV), 200--1
om., 334 éveirto (with Y), T 140 rrapaoxéuev (with W), 195 ts (with CW),
390 népe (with A1”), T 421 1<é>(vr0 X71600 (with ZW) <I> 213 é1<<1>0évE0rr0,
X 271 6’ om., 396 dmbotépw, 450 611 (with D), ‘P 103 1'1 (with Y), 562 éifitog
(with F°R), Q 382 T01 T0106 nap, 413 fi0e (with DRG), 759 1<01t0r1t6¢vn1
9 E.g. A 3, 65, 93, 124, 241, 271, 374, 496, B 339, 506, 702, 801, F 99, 215. 362, A 177,400,
E 160, 164, 333, 677 (With D), H 207, 451/8, O 35, 376, 378, 513, 530, I 112, K 306, 515, A 72,
439, M 204, 286 (With D“), E 223, Z95, O 450 (With T), 626 (With T“), II 227, 379, 409 (With F“),
445 (with E), P 144 with (PG), 214 (with G), 0 323, 417, 530, \P 307 (with YG), .0 ass (with
TV).
1° It carries a peculiar recension of the scholia which was already known to the compiler of
the Etymologicrrm Magnum. See K. Alpers, Das attizistisc/Je Lexi/eon des Oros (Berlin—New
York 1981), 93 n. '36. On this recension cf. Erbse (1969-88), I. lvi—-lviii.
146 I.5. The Early Medieval Transmission
1* B 544.11 231 (with D“), H 407, n 697, P 506 (with Wv), 734 (with ov), T 2s4, 0 321
(with RG°), X 59.
11 n 463, 673, 6ss (F), 732 (V), P 191 (W), 2 s (A40), 176 (YRW), 299 (A5), T 403 (W), r
17s, 215 (WV), 363, 0 92, 101,400 (DYPF), 433, 455 (At W), 493 (F), 317 96 (o), n 17, 723.
1.5. The Early Medieval Transmission 147
376 drreipéoz qawvniv, 383 K(XT87\8iTf8T6, 400 ‘ucéofiai, 578 oiqnpov, 635 sq, 642
uefiénxa.
19 was a good source in which many ancient readings were preserved, for
example: F 318 Beoig, tat (Nicanor), E 363 5' Eip (with G), 394 Kori, Z 96
dig KEV, 400 1<6?\Trov (with T), H 387 eirrépsv, G) 217 K’ évéitpnoe(v), 270
[3e{57\1i1<o1, 458 om. (with G), I 132 Koiipn, 698 pi] 6’, A 263 §[30rv (with T),
O 253 (and <1) 461) Aibr; viocj, 626 T8, 745 éfiéi )(or7n<o31 (la), P 379
Treoévtog, 478 Koiluwe, 729 T’ om., <I> 533 K?\.OVéE3l, 588 eipvoépeofior, 606
rroivteg, ‘P 225 orovorxilwv, 761 ;<e1p‘i, Q 300 1-es’, 428 Tdiv (sch‘”’).
Some variants were recorded in the margins of Z9, as appears from cases
such as A 190 ti B“ CC E“; T 348 )(0t?u<6v B C E, -6g B5 Cs ES; F 112
itouioeofiou BS E; ® 108 iuiotwpcx BS E‘; 109 KOp8iT(.0V BY” C‘ E; 267 int’ BS
C“ E“; N 667 <i>6eio60u BS C Es. This may be one reason why B C E do
not always agree. But it is probable that they occasionally turned indivi-
dually to other exemplars. This is especially the case with C.”
13 B and E, as mentioned above, are gemelli, written in my opinion by the same scribe. In
analysing the relationships of these manuscripts in the scholia, Erbse (1969—88), I xlviii—l and
lviii, found that C was the most independent. But his finding of a special affinity between B
and F does not correspond to my own experience with regard to the poetic text.
l
\I' 112 1<?1101ét0v (with Y; -1o'100v R), 348 rétpacbev (with F), Q 231 1<017w'1
(with F and a papyrus), 440 értofifiac (with F“ and a papyrus).
Often these manuscripts agree with one another in twos and threes, in
varying combinations: D R W D R G, D W/, R W G, R G, W G, and so
on. The most conspicuous pairing is that of D (or Di) with G. They fre-
quently share errors,“ but sometimes good or ancient variants, as at A 145
‘imrwv (with Y), 303 irrrrooiivmoi, E 343 uévot (with W), 756 Zfiv, Z 349
0161:-zicufipotvto, I-I 349 dtvtfiyet (with R), (E) 149 6’ int’, I 170 01160003 (AW),
241 du<p01<6pu11(301, 615 1<’ £2112-3, 628 rrponééyuevot, A 417 56 T’, M 369 ozii,
N 318 1<eiv0>(1), 485 éunhttinv, O 232 récbp’, T 336 rrponfiéyuevov, X 195
ciiifieofial, Q 399 56 Tot Their common source had some affinity with
Z, with which they share distinctive readings at A 106 i§8lTfCtQ, B 126
61011<0oun6z-zinuev, 163 uetét (with F), Z 367 wimp E'r’(1), T 276 Aattpnpfiv.
The next most common pairing is of R and W: I-I 367 (and I 95, O 285)
fig o<1>1v, (9) 80 Néotwp 0’, 144 <|>épt01r6<;, I 5 1" ex, 127 500’ éuoi, 401 épfig,
588 minor [3017t7\e1"0, A 329 Tfépt, N 337 fipxe, O 79 6’ ti (= Aristarchus), 274
ti ocbt, 701 Emrero, H 406 ei)u<e, 456 tapxfiowot, 488 614* (As TY”), 719
tbpovinv, P 178 uocxéeofioti (/9), 365 1161/ov, 408 T660, 509 0'1 Trap (desunt
AD), 534 1"’ om., 611 A6100, 705 évfirev, Z 349 Cé0(0)av (Y), 364 iéuuevott,
388 fivrioato (Y), 549 0011311’ érétvxto (Y), 550 62-: T1061 (Y), <I> 108 1<0t‘1 om.,
600 voip 6’ (PC), 604 vézp i-51-)e)tyev, X 307 131r01‘1, ‘P 76 vsioottou (cf. Y), 308
Kév (Y), Q 166 fitiiuot, 277 évteoloupvoug Of these readings, those at I
401, o 79, 274, n 406, 456, 4ss, 719, P 365, 509, 611, >3 349, 549, 0 600,
and ‘P 76 have ancient attestation.
F, as I have said, vacillates between the two main lines of tradition. It drew
often on Z7, whose version of the scholia it adopted. It shares b’s distinc-
tive variants at, for example, F 411 Tropouvéouoot (+W°?), 434 Trouisofiou
(= Aristarchus), E 394 Kai Tfi-Sp (ancient), E 222 vri 611 Oev (with T"'°W and a
papyrus),)I 496 7rpo615pn(1)o1v (Y), T 29 ufi 1101, (I) 158 interpolated (Gm),
'4 E.g. E 902 (schD"’ Ts), Z 155,402, 43s (with W), H 23,97, 129,215, 240-1, 293, 294, 300,
0 29, I 310, 343, 367, K 185, 280, 286, 302, 40s, 510, A 102, 132, 151, 323, 592, 727, M 11, 277,
3s2, "5 348, o 121, n 453 (AYP), 628, <1» 19s, X 17s, ‘P s1, 402, 632.
I“ I"
.6
._
150 I.5. The Early Medieval Transmission
300 Kexvpévoto, 376 60110|.1évr| 60110301, X 441 0p601, \P 653 0‘Ee6?t0v, 875
péoong, Q 361 éfifipeto.
The eclectic nature of its adherence to Z9 appears more clearly from
numerous cases where F has been corrected to the Z9 reading, or the Z2 read-—
ing is recorded in F as a ypoitbetotvvariant: A 258 fiovhfiv, B 558 present,
E 182 ytyvoimcwv, 394 1<0ti, Z 96 Céc KEV, 400 1<67\'rr0v (T), I 132 Koiipn, N
486 1<?\t-flog, E 62 éiupi, H 290 66161 ¢6[3n6ev, 299 i=§<1>0o/ow, Z 430 dvoioxsto,
T 153 0366, 396‘i1r110101v, Y 77 I-3, <I> 586 yoip 6’ 0'1, X 491 1'r01pc10"t, W 172
rtupi, 854 fiv Zip, 877 fi om., Q 673 ttb, 758 1" om., 764 ciiyoiv’ écg.
No less often F goes with the D R W G group. But it may have had
access to a separate source of good quality (or more than one). For some—~
times, almost alone, it agrees with A in its good or ancient readings: B 48
rrpooelifioeto (T), 505 '1"7r00fi{301c; (F°), 506 ’Ov;(r|ot6v, 646 Kvtooév, A 277
iévtt, 539 015 1<é Tl (F°T), Z 269 ye (W), H 56 i?5p130no01v, I 632 cbovfiog (T),
686 yoip éfiev, K 579 ticbuooéuevoi, A 690 yoip 6’, 747 étrépovotx, N 443
Irekétulev, H 237 riufioozcg, Z 209 rcpivtovtou, Y 259 61v031, 394 érttootétpotg,
<I> 246 fiivng (G), 411 ioocbcxpileig, 426 érri (G), X 36 éorfitcet, 128
d7\?tfi>to1'1'v (R), 272 Oviwv (W), 306 eipiioootto (W), 310 01p7r0i‘<§t0v,
380 €ppeEev (cf. W), 412 116w; (G), \I’ 535 0’ éip év ’Apvsio1g (G), 735
izpz-;i0e00ov (G), Q 731 éxfioovrai (ZD).
Or it agrees with variant readings noted in the margins of A: (9 496
Trtep6ev1" civépeue, I 602 éuiptov (G), K 278 pev, A 230 icbv, 380 [3é[3?1e011,
M 161 [50t7t7\0t1év60v, 350 011101 01160603 (W), O 330 i<01p"rep0013pt0v, 540 fiae’
i-:rr011115vr03p, 639 d1é671t0v (G), <I> 105 ouprroivtwv, 121 and 427 1t'rep6ev'r01
rtpoomiéoc, 416 ¢17topus10fig (G), 467 1t01u0u3t1s(o)001, 493 61161101 (/0), X 374
évé1tpn0e(v) (W), ‘P 662 veéofiw (RG), Q 785 po6oi50'i1<'rv>1o<; (T).
Even without A, F has the capability to proffer ancient readings. Some
of them are the readings of Aristarchus, and are doubtless derived from
the scholia, even though not all the relevant scholia are transmitted in F
itself: A 204 tekéeofiou dim, 214 fituv, 484 1<c1t01 (with RG), Z 157 1<o11<01
tniootto, I 324 5’ éipot (with T), O 563 5’ om. (with D). Some Aristarche-
an readings in F had also been taken up in /9: ‘P 287 iivepfiev, Q 663 vtitp.
F has certain other notable readings apparently derived from /J: A 318
Kev, T 393 cbdt’ éireuxéuevog, <1) 493 iitreitot (Aw), 310 1<’, 773 é1t01'iEs00c11
(W), Q 219 1<0t1<fi, 492 Tpointie 1107t6v1"01 (Y), 574 re 1:011 (RG). Others again
are shared‘ with Z: B 367 dhcnroiletg (Z“?), 633 Kpo1<t5>1nv, 1" 51 1<0rrn<i>ein,
E 127 5’ om. But there remains a residue of good variants that F is alone
or almost alone in presenting: A 238 7r01?\cit1r|(1)c (YV), 393 éoio (= Zen-
odotus, with HO, cf. Y), B 278 0 om. (known to Didymus; RG"‘?),
766 <I>npi111, 769 cbéptepog (V), F 78 '16p156no0tv, E 190 rrp0'1'0i1,u011, K 561
tpe101<011€5é1<01t0v, A 91 év 0’, 272 éfiei’ é615vn 5i3vev. The last item is par~
§
i‘ .
i
E
i
Z
i I.5. The Early Medieval Transmission 151
i
ticularly fine, having the support of two papyri. It might possibly be a
i
.4
Byzantine conjecture, but if so, it is above the usual standard.
i
-r.
i
composite nature of its scholia is matched by its inconstant textual allegi-
ances: it sometimes stands with I9, sometimes with A, sometimes with
-_»6m~W<wt -W-~¢<\_. _~;. _
i both, or neither.
§ Now and then the copyist, like that of A, adopts Aristarchean or other
I
ancient readings from the scholia: A 15 Ai00£-ITO, 608 ‘JT01I‘|(I6V ifiuintoiv,
i
6-) 275 Xpopiov (with R“), I 154 nohfipnveg, K 146 6115’ (with W), A 368
i
éfievoiptlev, M 11 £10169, Z 14 éitp ea vfiac_;’i'uev (with HV), 34 6146666618,
? T 30 a7\a7t1<éuev, 376 T6 re (with G), Y 28 06 rt (cf. W).
:-
Ii I
Alignment of T with [9 may be illustrated by the following instances:
i.
i A 365 civop:-215060, F 235 Kai 1<’, 283 om., 443 rrpcitepov, A 115 apxov
{I
’A>(a10'3v, A 263 Efaav, N 12 arporarni 1<0pv<]>fi1, H 745 iig, P 192
i2' :
1to7\u5ar<:pt5tou (with D), 751 01386 re (G), Z 463 ufi Tl T01, T 319 izpoi,
§
5.
I.
1{J
Y 182 )(s1p‘1, 338 Tpuieoot, X 75 aioxtivovot, 236 ting (G), Q 446 Btipag (T"),
r- 583 ii5T|(1).
)3
.‘\-.
With R W G: E 269 Bnkéag, E 376 §)(81, O 24 001100, P 455 interpola-
it
ted (CZF), Z 86 dig (F, deest D), T 75 atroettrdvtog (Ti with F, deest D),
s
Y 260 66<6>1<6 (/0 E2), X 217 Iron, 357 €\/6061 (schhlp F), 415 éfi ovouaidtfifinv,
E:
‘P 500 poiotivt 6’ (F), 733 avaiijavte 7r0'O\a1ov (FY), Q 83 661"’ (F, deest
D).
T’s closest single association in this group seems to be with R: B 475
4
61a1-spivouotv, E 491 tn7te1<7\e1"tt5v T’, 853 1'61‘:-3, H 419-20 om. T‘"*R, (E9 415
‘i
assimilated to 401, 507 5’, A 767 56 T’, M 9 Kev, 430 02-;, N 36 Epfiake (Cc),
i__
Q. 223 ywvoioxw, 415 éévra (G1), P 65 T’, 295 fipiae, 757 iiottg 1111<pfi1o1,
»<
T 135 d5>1eo1<ev (conjecture, = C“), 328 i;67\7re1, T 263 peia 51s?\e15oe0(-lat,
1
i 365 iuev, 471 évértpnoev (As), <1) 130 iituv, \I1 792 ’AXi7tfi'1', 864 om. (N),
E
Q 80 u0>t1[30a'ivn1, 768 évioitoi.
E
152 I.5. The Early Medieval Transmission
We have noted that R commonly goes with the D W G group, and that
it sometimes shows special connections with T or C. On occasion it has
the readings of A, especially readings recorded as variants in A: I-I 465 0’
(AS), I 7 iéxevev (T), K 323 Tr00661<ea (AYP), 342 ei (As), 452 ’e11fi10’, A 400
051<é01<; 11111000 (Aw), M 377 évavriot (As), 382 Xeipi ye Tfit étépnt 1136001,
E 453 uaxpa 131[?>0'100wv, O 54 1<é71ev00v (A18 TY") 240 éoaveipero (G),
P 42 ii T’ fi ts, <I> 234 5110115011; (A5), X 2 aveqnfixovro (As = Ar), ‘P 192
é1<01isT0 (DY), 539 110’ i:1<é71e1>0v (cf. AP’), Q 81 éuusuavia (Alp).
At other times R adheres to 5, for example: I 137 )(a711<013 1<a’1 Xpvoofi,
H 838 i;TréTe7171’ (with W), 863 Ttp0130’1c, (G), Z 506 T0101 0’, X 231 0Té011ev,
339 11’ Ea, 358 T1 T01 (TAW), ‘P 129 1<é71e0s (Fa Y), 254 1<7110in(1) (5 W), 629
06, Q 765 1101 viiv.
In a number of places R presents good or ancient readings either alone
or with few other copies: F 235 om. (=1 p3“; merely accidental?), E 387
Tpe101<ai0e1<a, 842 eEsv01p1Cev R“ (P), I-I 326 om. (accidental?), I 100 T03,
K 334 715711010, A 86 06011017, 709 and 725 11010000111, N 541 Aiveiac 0’, 572
divtootv (G), E 101 011r01taTrTavé01>01v (H), 222 0501 71e01<0371ev0<;, O 224 K8
(FG"), 272 é00e13avT0, 387 d11to[30’1vTec;, H 247 ’11<é0010 (5), 386 iite 011 1'3’, 707
1T(.0 (55), P 349 om. R“ (accident?), 440 01u<1>0Tép106ev (G“’), 467 Aiuovifiao
(G), 662 avtiov (GO), 681 10010 (5 G“I—I), 746 0'1 ye ue11a03Te (YG), Z 48
’A110’16e1a R“ (WHV), 427 om. (GH), <1) 195 om. Ra, Q 724 iTtTr000i11010 (D).
In a couple of places the copyist produced a chimerical text through the
unintelligent conflation of variants. At K 285, finding T0061 with
suprascript 0651', he wrote T00ei01-ii; at K 336, finding Tr0T’1 with suprascript
Ttpo, meaning that it should be Trp0T’1, he simply put Trp0 in the text. And
at A 108 he combined the vulgate 1<aT01 0Tfi00g B01715 50001 with the vari~
ant found in D G, ['50i71ev )(a711<fipe'i 00upi, to make 1<aT01 0tfi00g 1501716
)(01711<fip5i 00vpi.
With the text of A: E 806 1<apTcp0v, 909 "Apn, Z 269 vs (Fa), 478
1311117 T’ 0’1y0100v, I-I 186 011 1')’, 0 6 om. (DV), I 381 T10T1v10cTa1, K 65
0171711j710101v (Aa), 211 Te (T), M 340 T1010011, N 422--3 written on one line
in W, 422 om. A, 435 <bae1v01 (F), 799 q>0171r|p10evTa (AA), E 420 in mg.
(V), O 82 éinv (D4 Gm), 114 ~ 398 0’ i-51100 r|1’50a (F), 256 T150 (T), T 79
011<00e1v (ZG), ‘P 814 apcborépw.
With AV” (or Am, A‘): E 288 01T1oT1a150e00a1 (ZG), I 493 é<EeTé71e00av (Z),
500 7101[.’>fi(1)c Te 1<v10(0)n(1)1; Te, 512 01T10T1vr|1, 674 01110vé11eva1, K 41 ein
(D), 183 00003p1i0t0vTa1, A 455 E11’ ei, M 363 0111a 01160010, N 630 017171’
0111001 (Z), E 185 71a11T1p0v, )1 239 11’ev i§T1e1Ta, Y 317 xatouévn 1<aito01, 454
viiv 0’ a0 T001; 017171000 (F), 496 é1’>’Tp0>(01711.01 (F), X 403 Teprrucépavvog, ‘P
155 ’Avat1é11v0va (V), 854 fig yap, Q 20 1<017101pe, 28 apxfig, 165 11317111101.
Even independently of A, W is able to produce a fair number of good
or ancient readings, such as: E 288 r1p’1v ii (G), 808 om. W“, H 345 0’ om.
(Z), G) 240 1011011011, 317 £311’ (lo), 492 110717101 110’100v 1<a’1 710717101 uévnoa, K 53
Aiavre (Ts), 252 1Tap03)(101<1~:v, 413 011700015010 (DT), A 135 C1060 W“(?), 381
01110, M 280 vetcbéuev, 34243 A’1avTc (Ts), 433 115,15 E 294 i-ipog, H 80
01p0v0v, P 260 oiivoua W1, 396 15’ om. (with Y, desunlt AD), )1 552 Tr1T1Tev
W“, 571 apaprfi, X 452 v0'1a, Q 17 T017 06 T’.
Many of these, but not all, Ioannikios would have been able to extract
from the scholia, or to achieve on the basis of his considerable grammati~
cal expertise. I-Ie had an above—average understanding of orthography, pro»-
sody, and metre, and tended to apply it by accenting prepositions to show
anastrophe, geminating initial consonants that are lengthening the preced-
ing syllable, writing Kauav0p~ for Z1<au01v0p-, and using spellings such as
v1yvt001<t0 (A 651), Boppéng (I 5, ‘P 195), tint (I 245, Z 88), Al0U1TiTC1O
(N 427), 01va1t11.0Tc1 (P 497), 011001011 (E 28), 0111011 11e71é1.0v (H 131, cf. K 22,
132, 465), and the like; even ’10a’1 710119011 for ’10i5 710¢>ov (K 573). He made
an incorrect metrical emendation at K 230 and A 333: seeing that 000p’1
I<71i5TOQ Mevé71a0<; / A10ufi0ng (as most copies had it) was wrong, he wrote
000p’1 1<71e1v0g, the correct form being 1<71e1T1’)g.
G’s closest association, as I have said, is with D (or D“). But it often
picks up ancient readings that we find recorded in the A scholia or in AS,
such as: A 335 and 567 0100017, 406 000’ ET’, A 506 11éya’i01)(0v, E 118 TOV
06, 288 01Tr0Tr01150e00a1, 656 auaprfi, 661 [3e[37111'1<e1v, Z 174 Eeivtle, H 73
0’ izv, (E) 423 00, 505 0'ii;',s00e, 526 E71T10;.1a1 e0)(011ev0c_;, I 210 11i0T071s, 602
0113011311, A 254 0’, 509 pera1<711(v)0€;vT1-zg, 629 a0Tfi1, N 61 1<e1<0¢>1i3g (V),
0 370 61 06, 403 6116 (with R).
Two further passages confirm that the scribe of G must have had access
to an exemplar closely related to A. At (*3) 240 A gives Exqov with a
1.5. The Early Medieval Transmission 155
I:‘f.
‘c
the page has 26 verses instead of the ‘usual 25, suggesting that this one had
=6
been in the margin of A’s exemplar,” especially in view of the note in A
I-i
“i-iv Tlotv 013 1<6iT0t1”. The omission in G, then, reflects a direct relationship
not with A itself but with A’s exemplar or another cognate copy.
G has various other good readings agreeing with A: A 216
eipiioootofiou (AA), B 95 otevotxilero, 1" 63 ooi, 78 0m., 119 §ll<éM-ZUEV, 279
If-I iittg, A 94 err rrpoléusv, 308 énépfieov, 424 uév T8, 433 nohvndpovog, E 183
eioépotov, 744 rrokiwv, 797 teipero, 852 é7\é060u, G) 140 ii 5 1'01, 315 in
mg., 539 dwfipwg, I 449 nepi Xuioozto, A 38 ouitofi, CD 409 inv, 520 Trétp Zqvi,
526 izotrizcet, 554 K7\OVéOVT(X1.
i It also has a few good variants that were not to be found in A: A 238 5'
om., 528 Tf7‘\8l5|.l0V1 G“?, E 200 duvet, 288 rrpiv ii.
Distinctive agreements of G with Z9 are not frequent for the greater part
of the poem, but become markedly so in X: H 72 Eiotusints, (9 458 om.,
J
O 179 trtohsnilwv, 570 oiirs 0136’, P 136 i<cx7\t'51r'r0v, T 86 oipeifiéuevog,
X 265 oiité ts, 282 Aotfioiunv, 302 nap, 370 Ké i=;, 378 fiptoz-:<; Actvotoi
Bepoinovteg ”Apno<;, 388 Cwbg év ’Apveio101 (i)tAOTfTO?\él.1010l ueteito (A"'°),
398 §0t00ev, 422 éikyea Gfitcev, 425 "Aioog 6%, 458 uévev, \P 695
oiuqaéototoav (Y).“’
Q:-..
E,
156 I.5. The Early Medieval Transmission
-,--,1-----.i-i--_.i
*7 See the apparatus at E 585, N 222, 287, 465, 710, 783,11 636, 682, P 506, 630, E 39, 200-1,
485, T 231, 426, 484, (D 185, 194, 258, X 416, 468, 476, ‘P 42.
I.5. The Early Medieval Transmission 157
The next task would be to bring into the picture other manuscripts of the
twelfth and thirteenth centuries such as [Allen’s sigla]:
Et Eton College 139, s. xii: A-—E 84.
V10 Vat. gr. 903, S. Xiii A 62-13 368, T 359--E 278, E 342—M 299,
M 365--0 280, Q 333-587.
V32 Vat. gr. 1315, S. xii! A 23--I\I 21, N 235~——<I> 387, (I) 605-0 468.
M11
Ambr. gr. L 116 sup. (502), s. xii—xiii: A-0.
P10
Paris. gr. 2697, s. xii—xiii: A-~M.
V13 Vat. gr. 1316, s. xii—xiii: A 109-0 216.
L16 Laur. 32. 47, s. xiii: A-0.
M5 Ambr. gr. F 101 sup. (355), S. Xiiil A 574-—-‘I’ 895.
M7 Ambr. gr. I 4 sup. (450), AD 1276: A--0.
Mc Monte Cassino S 94, s. xiii: F 220-~\P 782.
P3 Paris. gr. 2681, s. xiii: A-0.
P13 Paris. gr. 2768, s. xiii: A-0.
P5 Paris. gr. 2894, s. xiii: A-0.
P17 Paris. suppl. gr. 497, S. Xiiit A 215--365, A 588--B 493, T I-299, F
458--A 754, A 802—I\I 67, N 145-307, E 112-265, O 244-413, IT
319--565, P 266—-349, Y 188--X 469, T 36-0.
R‘: Flor. Riccardianus 30 II. 10), s. xiii: A 69-‘P 482.
V6 Vat. gr. 31, S. xiii: A 449*-Q 636.
V14 Vat. gr. 1317, s. xiii: B 532--0.
V15 Vat. gr. 1318, s. xiii: B 494-0 685.
The questions to be asked of them are, firstly, do any of them show any
ability at catching ancient readings independently of the manuscripts stud-
ied above, and independently of the scholia? Secondly, can significant
affinities be established between any of them and any others?
Positive answers to either question would be a gain. But it is only too
likely that enlargement of the data-base will confuse rather than clarify a
picture that is already confused enough. The prospect is of diminishing
returns.
6. The New Teubner Edition
The objectives
The essential features of a modern critical edition are a considered text and
an adequate critical apparatus. What should the editor of the Iliad aim at
in each case?
As regards the text, I conceive the aim as being the best approximation
that may be possible to the Iliad as its original author left it. There is no
reason to suppose that he did not leave it in a finished state. In some pla-
ces it is plausible to recognize lines or passages that he himself inserted
into an earlier version that lacked them. These are of great interest for the
compositional process, and it is sometimes helpful to point them out in a
critical note as an answer to difficulties that the reader may feel. But it is
not essential to the enterprise. We have to choose one version of the text
that we are aiming to establish, and clearly it should be the poet’s last ver-
sion.2 I
On the other hand, the edition must also provide the user with the Iliad
that was read throughout antiquity. It cannot, for example, omit the Dolo-
neia or other interpolated passages that were clearly current in all or most
texts from an early period. They must remain in the text, though bracketed,
so that they can be seen in the contextsin which they were interpolated and
in which they have been read ever since. Interpolations that were not so
widely current and that appear in only a few sources will be better relegated
to the apparatus. Some of them are linked with ‘wild’ variants in contiguous
genuine verses, which could not in any case be displayed in the text.
Another requirement is that the edition should use a text format fami-
liar to the modern user. It will differ from an exemplar of the Hellenistic
or Roman era, and even more from the author’s original, in making full
use of word division, punctuation, accents, and breathings, as well as being
in minuscule characters of modern type, with majuscule initials distin-
guishing proper names. It will use the late Classical orthographic conven-
tion of representing € and 6 by t-:1 and 013.3 All this may seem too obvious
to be worth stating; but it is a necessary qualification of the declared ambi-
tion to restore something like the text of the Iliad as its author left it.
How realistic is that ambition? We know that in the early centuries the
tradition was unstable, characterized by so-called ‘wild’ variants, and we
have only glimpses of the total extent of the textual diversity that may have
existed.4 We must concede that it may be impossible to establish what lies
2 Not many will support Nagy (2000) in his protest against ‘privileging’ a single version of
the text. He considers that Homeric poetry, deriving as it did from traditional oral epic, conti-
nued to be an essentially multiform thing, every version of which should be regarded as having
equal validity. He calls for a ‘multitext’ edition in which, wherever there are variants, the text
would divide into parallel streams to display them all without privileging any one. The editor
would then discuss them in a commentary ‘from a diachronic perspective, making his or her
own considered judgments about differences in the chronology, dialect, historical provenance,
and so forth’ -— only not passing any value judgments. Nagy fails to see the difference between
a genuine creative oral tradition and the reproductive tradition of the Homeric rhapsodes,
whose business was to perform excerpts from specific poems conceived as fixed entities. Of
course their interpolations and ‘oral’ variants are valid and interesting as manifestations of their
art, but they have a quite different and inferior status to the original, normative text that the
rhapsodes were supposed to be presenting. My edition is for those whose greater interest is in
that original text. For those who wish to study the rhapsodes’ variants, the evidence is there in
my apparatus. For criticism of Nagy’s position see most recently M. Finkelberg, CPI? 95 (2000),
1-11.
3 Cf. chapter 1, p. 29.
4 Cf. chapter 1, pp. 14f.
160 I.6. The New Teubner Edition
on the far side of that ocean. Some scholars have concluded that we should
renounce the ambition, and attempt nothing more than a reconstruction of
the Athenian text as recited at the Panathenaea after Hipparchus, or at any
rate as known to Thucydides, the orators, and Plato, or alternatively the
Alexandrian text as Aristophanes and Aristarchus had it. But it is clear that
there was neither a single ‘Athenian’ text (and we have not the material to
set about choosing the best of Athenian texts) nor a single Alexandrian
text, even if some agreement was reached after Aristarchus as to which ver-
ses should be included and which excluded. We know of many variant
readings current in the Alexandrian period, and we have to choose from
among them. In any case, why limit our ambition to reconstructing an
Alexandrian text, stopping five hundred years short of the originals, when
we have at least a modicum of evidence that takes us further back?
No, let us hold to our aim of establishing, so far as our means allow, the
pristine text. We know we cannot achieve total success. But let that be our
objective?
As for the critical apparatus, its primary purpose should be to provide
the user with adequate information about the sources for the text at every
point, the variants known to exist, the strength of attestation for each, and
such conjectures as are judged to be of interest. With many ancient authors
it would be sufficient to say what I said in the Preface to my edition of
Hesiod’s T/aeogony, that my aim was ‘to include (a) anything from any
source [manuscript or conjecture] that might be or point towards the true
reading, (I9) the principal corruptions in the principal sources.“ With
Homer, however, we cannot make such a clear distinction as in most auth-
ors between ‘the true reading’ and ‘corruptions’; nor is it easy to say what
are ‘the principal sources’. Do they include all the papyri, all ancient quo-
tations however careless (and how ancient must they be?), and all records
of ancient scholars’ readings? Given the great quantity of data available
under these heads, and the large number of medieval manuscripts, there is
a danger of information overload. The apparatus of Ludwich’s edition
5 I make no arrogant claims for my edition of the kind implied in Nagy (2000), as when he
writes, ‘West is relatively optimistic about his own edition. He seems confident that he has reco-
vered the closest thing to the putatively original Iliad.’ I have no such confidence. But I do assert
the right to the active exercise of my judgment. Nagy seems to think that an editor should sim-
ply marshal the evidence in a non—committal way. He complains that ‘the text of West’s Iliad
contains many editorial judgments that go beyond the manuscript evidence and that flatten out
the textual history of Homer.’ He and I evidently have very different concepts of the editor's
role.
" West (1966), v.
1.6. The New Teubner Edition 161
often takes up more than half the page, and that was when only a tiny frac-
tion of the papyri now published were known. Allen’s 1931 edition pro-
vides vast amounts of trivial (and frequently confused) information about
the readings of manuscripts down to the sixteenth century. The Budé
Homére, like P. Von der Miihll’s Odyssea, is far more succinct and gives
more value per centimetre, but at some expense of precision: often vari~
ants are cited without any indication of source.
In my view the ideal lies somewhere in the middle. I have aimed to iden-
tify and report all non—~trivial variants that are ancient or have a good
chance of being ancient: not because all ancient variants have a chance of
being true (i.e. the reading of the original text), but because they are of
sufficient interest in their own right as evidence of the nature of the trans»-
mission. For the same reason I think it appropriate to report all readings
and conjectures of ancient scholars, even when they are obviously absurd.
i
I
164 I.6. The New Teubner Edition
ten down, and we know of no process likely to bring Aeolisms into the
text at a later epoch.
Faced with such a relatively rich manuscript tradition and such an abun-
dance of information about the stateof the text in later antiquity, there is
a danger of the editor’s succumbing to a kind of moral paralysis when it
comes to going beyond what is directly transmitted. Few have had the
nerve even to print vterritfiytov at B 264. But in view of the evidence that
some archaic morphs, while still leaving traces in the tradition, have been
largely replaced by more recent ones, we should not be too hasty in dis-
missing all the modern proposals for restoration of older forms. How
many more choice survivals might we not find, if we had a papyrus or two
from the sixth or fifth century?
As regards the layout of the text, nearly all previous editions agree on a
feature that seems to me unfortunate.“ They start each new ‘book’ — or
properly speaking, each rhapsody — on a new page, with a prominent head-
ing. This invites the reader to pause, perhaps to go and make a cup of
coffee before continuing, or to conclude the reading session altogether. But
these divisions are not original, and although whoever introduced them
made some attempt to have them coincide with natural breaks in the nar-
rative, this was not always possible, and in some cases they clash visibly
with the poet’s own scene-division (nightfall).15 One should read straight
on over these artificial divisions. I have accordingly marked no break in
the text, or at most a new paragraph where appropriate.
The testimonia
I have devoted much effort to the collection of ancient quotations from
the Iliad as a potential source of new ancient variants or of support for
those known from the direct tradition. Many, of course, were already
recorded in older editions, particularly those of La Roche and Ludwich,
though not all subsequent editors have taken much account of them. Lud-
wich made a systematic collection of the earlier quotations, down to the
Augustan period,“ but much remained to be done for later antiquity. I
have aimed to extend the collection down to the ninth century: after that
we have a rich direct tradition and quotations are increasingly unlikely to
contribute anything better. This chronological limit meant drawing a
perhaps somewhat arbitrary line through the Byzantine grammatical
tradition; I included Choeroboscus, Photius, the Epimerismi, and the
Etymologicum Gemtinztm (so far as I was able on the basis of published
information), but ruled out the later Etymologica, the scholia to Oppian,
and those to Dionysius Thrax. A line had to be drawn somewhere.
I further restricted the collection to verbatim quotations and excluded
mere allusions and general references from which nothing could be
inferred about the reading in the text. Anyone studying the later reception
of Homer would have to take them into consideration too, but for my
purpose they were superfluous.
If I had followed Ludwich’s practice of appending quotation references
after the manuscript sigla cited for each reading, they would have taken up
an inconvenient amount of space and unbalanced the apparatus. Instead I
have listed them in a separate register below the text, indicating the limits
of the quotation in each case; a line number followed by ‘a’ or ‘b’ stands
for the portion of the verse preceding or following the main caesura.” In
the apparatus it is then normally only necessary to refer to the author’s
name or, where no ambiguity arises, to use the siglum t (singular) or tt
(plural). When the testimonia are divided between two readings I specify
the authors who support the one variant and put simply ti‘ or tt"' for the
rest. By referring to the quotations register the user will be able to work
out which ones constitute t"' or tt"' in each case. Where there are several
overlapping quotations, I gather them together in a block with semicolons
between the entries, so that it can more easily be seen how far back one
must look to find all the references to quotations that cover a given verse.
It often happens that an author or lexicographer quotes a word or
phrase that occurs more than once in Homer, without indicating from
which context he has taken it. In such cases I list it only at the place where
it first appears in the Iliad. I do not record instances where a verse, or part
of it, is quoted in a scholium on the line itself, though I do so when it is
quoted in a scholium on a neighbouring line.
Under the general heading of testimonia we must also include - insofar
as they can be related to specific Homeric models -- parodies, such as that
of Matron of Pitana (SH 534), imitations, and centos. In the last category
‘7 To avoid any confusion, I note here that in the apparatus line numbers followed by ita-
lic ‘rt’, ‘b’, etc., denote plus-verses.
I.6. The New Teubner Edition 167
represented among them, and always to cite these manuscripts when vari-
ants are mentioned. This is not because we rely anxiously upon precisely
these exemplars for knowledge of the text; it is rather in order to give a
balanced picture of the state of the tradition between 900 and 1200. In the
Iliad particularly, where we have such a wealth of evidence from scholia,
papyri and quotations, our dependence on the medieval tradition is much
reduced.
Occasionally one or more of the copies outside the selection presents a
reading which is not in the select manuscripts and is either superior to
what they give or is shared with an ancient source. In some cases this may
be due to fortuitous coincidence in error. But it seems advisable to take
note of such readings where they occur, and of others that look as if they
might be old,” even if we do not wish to cite the manuscripts in question
regularly. A few, such as I-I O V and the principal representatives of the /9
family, are dignified with individual sigla, for the not very compelling but
practical reason that van Thiel has provided them with sigla and cited their
readings. The rest I have referred to without distinction as ‘rr’ (recentio-
res), or in the singular ‘r’, though it should be noted that a few of them
date from the twelfth or early thirteenth century and are thus no later in
date than some of the select copies. I have taken their readings from Lud-
wich or Allen, and anyone who needs to know which manuscripts they
are at a given place should consult those editors’ apparatuses.
If the details of a reading given in my apparatus are modified by a paren-
thesis, the qualification applies only to the last source mentioned. For
example, at N 229, “’i5ncu 9 10 (~fllOLl) 47 Q""’, it is only papyrus 10 that
has tomou, not 9. Sometimes, instead of detailing insignificant differences
among a group of sources, I have used the adverb ‘fere’ to indicate that
they all more or less agree on the reading cited. In stating that a group of
manuscripts has such and such a reading, I generally ignore trivial diver-
gences in such matters as breathings, movable nu, and the presence or
absence of accents on prepositions and of iota in the long diphthongs,
except in cases where some interest attaches to the question.
On the same principle, when I have had occasion to cite conjectures of
older scholars, I have not scrupulously recorded whether or not they prin-
ted digammas, if the significance of the reading lies elsewhere. Sometimes
I have used a formula such as ‘Fick post Payne Knight’, meaning that
Payne Knight basically had the right idea but an incorrect form, and Fick
i_
1° I mean a variant such as Ii] pisv éip (ix; eirroiio’ at A 428 for 63¢ éipot qxovrjooto’ of the vul-
gate (and four papyri). It would not be surprising if this were to turn up in the next papyrus.
170 I.6. The New Teubner Edition
corrected it. For example, at E 333, where the paradosis is or-:6 iiotepog, it
was Fick who restored oé’, but Payne Knight had already posited an eli-
ded, uncontracted form, albeit a false one (oe/7 ’).
' Other similar passages such as Eur. Hec. 1077, Ion 505, and Tim. Pers. 138 (cited by
R. Renehan, A]P 100 [1979], 473f.) may derive from Aeschylus and are not necessarily inde-
pendent witnesses.
174 II. Notes on Individual Passages
T0101 5’ dvéotn
uoivng ®eoTopi5fi€. oiwvonéawv Eix’ éipioroc,
Kci)\)(oTc_;, 5v ®éoTwp Trot’ év Etifioinr eupeini
veivorro, "I5u0v0c vibe ’ABorvTioi5cxo &vou<Toc;.
The extra lines may have been a Euboean interpolation, or a borrowing
from the Cypria, where Calchas played an important role in ordaining the
sacrifice of Iphigeneia. j
A 170. 0135:-'3 o’ dim etc. The pronoun must be dative; but the enclitic
form of the second-person pronoun in Homer is T01, not oot. So we must
either write T’ <’)'it0 (or T01 oito), or allow that the pronoun is emphatic. ooi
0’iuJ (following Bentley) would enable it to assert its accent.
\ t I , \
A 185/207. In 185 all sources give To oov yepotg, but Payne Knight s Teov
is plausible in view of 138 ii Tebv yépotg and 207 where T60V and To obv
pévog are variants; cf. 282 Teov uévog, Z 407 To oov uévog, N 287 and
<I> 340 Teév ye uévoc, <1) 305 and X 459 To 5v uévog (v.l. Teov in both).
A 344. The objections to u0t)(éo1vTo ’A)(0uoi (Q, also sch“ [Aristonicus?]
and sch. Theoc. 3. 24a) are the hiatus, the un-Homeric form -o1vTo, the
use of u0t)(e- as a present stem, and the syntax of the optative. Thiersch
preferred |.1(X)(é0VT(X1, which has since been found in Z and in five out of
the six papyri covering the passage (corrected in one from —ovTo, which is
what the sixth has), so it is at least as well established as the other as an
ancient reading. It is tempting to adopt it, as do Leaf in his commentary,
Ludwich, and van Leeuwen. But iiirwg + future is a Classical construction,
rare in Homer and lectio facilior. The potential optative is good Homeric
syntax for the unfulfilled consecutive, cf. or 254, [5 53, 5 167, 560, 1 126
(compared by Hentze). It is hard to see why uot>(éo\/Ton would have been
corrupted to -o1vT0. The use of u0c)(e- as a present stem in the optative is
paralleled at 272 u0t)(éo1To. Bentley’s -oi0tT’ neatly removes the other two
difficulties. Cf. the variants at Z 376.
A 531-3. See West (1998a), 103. The argument that Zeus did not leap
into his mansion can be met by comparing other passages where, following
a more specialized verb in the first clause, a more general one has to be
understood in the second: K 334f., N 584f., 0 107.2
At B 4 Ttokfig was the reading of Zenodotus. The scholium stating that
at A 559 he had Tt07\e'ig should not be emended (Lehrs) on the assumption
that his §1<50o1g must have been consistent.
B 15. Papyri and the medieval tradition give the line in the same form
as in the repeats, 32 and 69, where the clause Tptéeoot 5:-3 1<fi5e’ é<1>fiTrTou is
continued in the next line by etc Atdg. Aristotle, however, twice alludes to
a sophism of one Hippias of Thasosi which presupposes the reading
5'i5ouev 5é oi eiixoc dtpéofiou (cf. <I> 297); by accenting 5156uev (infinitive)
he tried to avoid having Zeus state an untruth. Ps.-Alexander (Michael of
Ephesus), commenting on the Soph. el. passage (33. 27), could not find the
quoted hemistich in the context of the dream and was mystified, but it
must have stood in line 15. No editor has adopted it, so far as I know,
though it was hailed as genuine by E. Schwartz (1908), 10f. It might in
theory be some rhapsode’s variant inspired by (I) 297, but that is a differ-
ent and distant context. It fits perfectly here after the reference to all the
gods consenting. If it is original, we can see why the poet had to change
the expression in 32 and 69: 5i50uev could not metrically be converted into
the third person. We see too how easily Tpuieoot 5i-3 i<fi5e’ éofirrtou would
have displaced it.
B 111. Aristarchus, Callistratus, Ammonius, and Dionysodorus had
uévotc — so too one testimonium and a late papyrus - while other sources
have uéyot. The latter looks like the lectio dzfficilior, but is also difficult in
a bad sense, as it does not sit well with ditni évé5n0'e [3cr.peir]1. On the other
hand it does go well with the idea dotoe, cf. I 533, A 340, H 685. Perhaps
the poet said uéyot with that idea in view and then changed his phrase.
uévotg does occur as an epithet of Zeus in the nominative, and emphasis on
his power is in place here; but to occupy the iambic slot before Kpovi5ng
we find rather Trcxrrjp, <11 508, X 60.
B [206]. The verse (a modification of I 99) is poorly attested, and unsa-
tisfactory here because oqnot has no reference. It was no doubt added
because 5tii1<e seemed to require an object; for this motivation of interpo-
lations cf. above, p. 12.
B 216. See above, p. 22.
B 258. We have the remarkable situation of four concurrent ancient vari-
ants for the end of the line. vii Trap 0356 (3 Q) is the best attested, and
looks the best in itself. Presumably rhapsodes had difficulty in remember-
ing this unusual phrase (vu Tffip (3 320 v.l.; not elsewhere, at least in Homer),
and improvised substitutes.
3 Osann conjectured ’H7\eioc;, HAEIOC = GACIOC, and the passages appear in Diels-Kranz at
Hippias 86 B 20 under the heading ‘Zweifelhaftes’, with bibliographical references on the iden-
tity question. Cf also RE s.v. Hippias nos. 9 (a Hippias of Thasos in Lys. 13. 54, 61) and
16. For another Mioig by him using reaccentuation see ‘P 328. Cf. above, pp. 24f.
176 II. Notes on Individual Passages
4 There is a superficial parallel for tivtnfiévta véeofiou in Y 117, where Nestor tells Telema-
chus that it would take years to apprise him of all the Achaeans’ sufferings at Troy: Trpiv KEV
civmfieig ofiv rrottpioot yofiav 'ii<o1o.
II. Notes on Individual Passages 177
lines added when the Aulis catalogue was adapted to the Trojan setting.
The Boeotians were early members of the Delphic Amphictiony, following
the First Sacred War. An alliance of Boeotians, Phocians, and Locrians is
presupposed in [Hes.] Scat. 22-6, where Amphitryon leads an army made
up from these three peoples.
B 527. In the epic language ‘the son of N’ is not normally expressed by
the simple genitive without Trotig or 816;. The only example quoted is
Hymn. Herm. 145 Aloe <5’ éptotfivtog 'EDHfi§ (vioc M. Schmidt); others
appear in Hipponax and later.5 Ilgen, commenting on that passage, saw
that ’O'i7\fiog A’iot§ might conceal an Aeolic-type patronymic adjective, like
Tehaptévtog A’i'otc;. Nn7\fi'ioi'irnrot, Hoioivrtog uiog. and in Stesichorus rcotio’
’Of>uo”e'iov. This seems extremely probable, at least as regards the origin of
the formula, though in the Iliad it belongs to a system which includes A’iotg
’O'1'7tfioc TOL)(iJC_§ viég (N 701, E 520) and Méotov ’O'1'7\fio<; v60oc uiécg
(N 66), v66oc; uiog ’O'i7\fiog Oeioto (N 694, O 333). So the poet probably
already heard ’O'i7tfio<; as a genitive, whether its primary form was ’O'i7\fi1o<;
(for unrnetrical -fjiog) or ’O'i7\e'iog (’O'i7\éF1oc_:, Pick)"
B 535. Again an inorganic line is added after the ship number. We do
not expect this general remark about the location of Locris from a poet
who has already listed its principal towns. Perhaps it was added by a rhap-
sode who thought these East Locrians should be unequivocally distin-
guished from the Ozolian Locrians.
Irépnv iepfig E1.’>[3oinc; is nevertheless a curious way to define the posi-
tion of Opuntian Locris, as a glance at a map will show; it would apply
more fittingly to Boeotia or Attica. The line is half parallel to 626, vfiowv
oii votiovot "rréprjv tiO\oc ”H7\16oc; oh/Tot, and indeed some manuscripts have
been affected by interference from that line. Was the parallel perhaps ori-
ginally closer, and did 535 once read Aoxpdiv, (ii votiovot rrépnv dkog
'E7\7\0i5og tiivrot? If so, the motive for change at a subsequent date is
obvious.
B 546-58. It is generally accepted that this section has been subjected to
Athenian rewriting, presumably in the time of Pisistratus or his sons. What
..,.....?..........,.....,. .
7 Cf. R. Hope Simpson and ]. F. Lazenby, The Catalogue of Ships in Homer’s Iliad (Oxford
1970), 56; E. Visser, Homers Katalog der Sc/Jiffe (Stuttgart-Leipzig 1997), 437 f., 445 f.
8 For the status of the pseudo-Hesiodic fragment as witness to a more original tradition than
that of the Iliadic Catalogue see M. Finkelberg, CQ 38 (1988), 31-41.
II. Notes on Individual Passages 181
where in the Iliad we find the phrase veo3v itv dvdfivt meaning ‘the assem-
blage of the ships’ (O 428 = H 500; H 239; T 42; cf. Y 33 pet’ dwlivot vedfiv).
It has always seemed odd that Ajax should have ‘brought (his ships) and
stationed them where the Athenian ranks were stationed’; and the present
participle does not go well beside otfi Ge. The parodist Matron of
Pitana, certainly, understood ciivwv: Supp. Hell. 534. 95-7, Trotig 65. rig i:1<
Zowxpivog éivev tpe1or<oti5ei<ot vfioootg I dc; 6 ucivetpocg I Bfire oépwv,
‘iv’ ’A6nv0titov xaréiceivto cbdkavyeg. Cf. Hymn. Dem. 384.
B 572. (Cf. West [1999a], 188.) The reference to Adrastus is curiously
pointless, and seems to be merely a way of filling out the line, the pur-
pose of which was to give Sicyon a place in a document from which it was
missing. The want of such a line would be felt most keenly in the sixth
century, when Sicyon rose to prominence under Cleisthenes. The cult of
Adrastus was important at Sicyon at that time, and indeed it was then that
the Delphic Oracle pronounced him to have been a Sicyonian and not an
Argive king (Hdt. 5. 67). Cf. K. O. Muller, Gesclaiclate der grieclais-clJen
Lll€7’6lll/£7’, 3. Aufl. (Stuttgart 1875-6), I 90 n. 69, ‘Hier zeigt insbesondere
der Vers Il. 2. 572, worin Adrastos der erste Konig von Sikyon heisst, ver-
glichen mit Herodot (5, 67-8), recht deutlich die Ansichten des argivischen
Rhapsoden.’
The only other reference to Sicyon in the Iliad, at ‘I’ 296-300, is also
suspect.
B 585. K051 ’O'iTv7\ov was already proposed by Wilamowitz (1884), 324
n. 38.
B 703. After this line, 708-9 appear as a pointless repetition. Payne
Knight and Wolf deleted them. This illustrates the common phenomenon
that when corruption has left two things in the text that are singly accept-
able but cannot coexist, critics tend to leave the first alone and to attack
the second, because it was when they came to the second that they became
aware of the difficulty. But the chances are even that it is the first that is
wrong.
There are two reasons for thinking so in the present case. Firstly, 703
has an obvious source in 726. Secondly, and more damningly, it does not
make good sense after 702, because ‘they were not leaderless either’ implies
a previous mention of another contingent who were without their original
leader but had a replacement one. The point is made by Leaf, who does
not bracket the line in his text but says that it ‘is simply copied here from
726’, where it is fully at home. When 703 is eliminated, the reference to
‘them’ in 704 may seem a little awkward following the four-line digression
on Protesilaus’ fate; but they, the men of Phylake and the other places
182 II. Notes on Individual Passages
listed, are the main subject of the entry, and the poet could easily revert to
them in this way.
B 749. The first of the two tribes mentioned in this line were called the
Ainianes. The name could not have been written with ’E- instead of
Ai- until the diphthong came to be pronounced like s, that is, not before
the Hellenistic period at earliest. The medieval tradition is largely agreed
on the spelling ’E\/ifiveq, but we now have four papyri, all of which write
0uv—. Cf. Waclternagel, Glotta 14 (1925), 40 = (1955), II 848: ‘Ebenso haben
bei Homer B 749 alle byzantinischen I-Iandschriften 'E\/ifivecg; erst aus den
Papyri der ersten zwei Jahrhunderten n. Chr. ist das einzig mogliche
Aivtfiveg wiedergewonnen worden.’ In the quotations of the verse in Stra-»
bo and Stephanus of Byzantium the manuscripts are divided between the
two forms; in that in sch. Soph. El. 706, the scribe of the Laurentian wrote
otivetfiveg and then altered it to évlfiveg. In Hymn. Ap. 217, where the
people appears again, the best manuscript gives fie’ étyvlrivotg, clearly
implying Aivtfivag, and the others have the same further corrupted
p0tyv1r'|v0t<;, ii uotyvniootq, etc.). The consonantalization of the iota to
accommodate the awkward name in the verse is as in B 537 'IoTi0u0tv, I
382 Aiyvrrriotg, Hymn. Dem. 266 ’E7\evoTvitov, Tyrt. 23. 6 Meoonvitov,
etc.; see West (1982), 14.
That the false form ’Ev1fiveg still stands in the texts of Allen (1931) and
van Thiel (1996) shows to what an extent editors still take the medieval
vulgate as their starting—point and look askance at more ancient
manuscripts.
B 765. Eumelus’ steeds are ‘equal—backed to a T’. The tradition and the
editors accent 233751 vtbtov as preposition + case. But while this is appropriate
in a phrase such as ‘black on the back’ (PMG 848. 5 5:131 vtbtot |.1é7\0uv0t),
one cannot say ‘equal on the back’; what is wanted is the plain accusative
of respect. Nor is the bare dative 0Ttx<bL5?\r|1 easy. We should join the éiti
with it and accent 0'r0t<bL57\n1 Em. Cf. 6 245 al. étrti otoifipnv ’{6vve, of the
craftsman who planes his wood and makes it straight to the ruler; in that
expression the accusative is appropriate because of the sense of conver-
gence.
B 797. Here three papyri agree on a reading different from the medie-
val vulgate, dig ré Trot’ eipfivng as against dig Trot’ en’ eipfivng, and a fourth
exhibits a conflation, we T8 TFOT err etpnvqq. Schwartz (1923), 62, argued for
the adoption of the older-attested variant, explaining eipfivng without pre~
position as a genitive of time analogous to vu1<1'6g. But a closer parallel
would be welcome. Elsewhere the poet uses the preposition, in the for-
mula to rtpiv err’ eipfivng, 1rp‘1v é7t(-)e'iv uiotg ’A)(0ut3v (I 403 = X 156).
II. Notes on Individual Passages 183
B 811. Here and in (P 567 we find the phrase Trportdpotfle Tr6P\1o<; with
the noun scanned as a disyliable, whereas in the thirteen other instances of
this form in the Iliad it is always trisyllabic. Synecphonesis of to would be
anomalous, and consonantalization of the iota (as in Aivifiveg, above)
hardly compatible with the maintenance of short quantity in the first syl-
lable. The minor variants 1r67\r]0g, Tr6?\eo<; do not help; the first is imply
unmetrical, the second is not a Homeric form. It seems worth noting the
possibility that, as at A 559 (above), the explanation of the oddity is to be
sought in an ancient, originally disyllabic form that no longer existed in
the Classical language. E. Schwyzerg suggested that what lies behind it is
a genitive *Tr67\e1g corresponding to the Sanskrit i-stem genitive —e[a -< "Y-eis.
Apparent survivals of such a form in Asiatic dialect inscriptions, however,
are very dubious.”
B 865. Already in antiquity readers were struck by the oddity of Talai-
menes’ sons being born from the Gygaean Lake, and some critics pro-
posed reading Fuyotint kipvm as a local dative. According to Eustathius,
some assumed the loss of a following verse in which the mother was
named. This is not necessary, as it is common enough for Tiicrw to be used
of the father (B 628, E 547, Z 155, 206, etc.). Still, ‘whom he fathered at
the Gygaean Lake’ is an odd way of conveying the information that that
was their home, which is presumably the poet’s essential aim. Aristarchus
read Fvyotin hiuvn, and that was the generally accepted ancient version.
In theory there might have been some myth about a supernatural birth of
two boys from a lake, but the ancients evidently knew nothing of such a
story. The figurative expression in II 34 y7\0tvKfi E36 oe time Eioiikotooot is
clearly of no relevance.
Van Leeuwen suggested replacing Aiuvn by v13u¢>n, referring to 672
(Nireus son of Aglaia) and Y 384-91 (Iphition born to a mortal hero and
a Tmolian Naiad beside the Gygaean Lake). The latter parallel, however,
only reinforces our feeling that the Gygaean Lake should be mentioned as
the place where these brothers were born and fostered. Hence I suggest
that Tére may have replaced 1'pét|>e. Cf. [I-Ies.] fr. 181 ii’ 0’in<v> Tpin
Botwrin E'rpeq>e 1<o1$pr|v (if that is the right reading); Aesch. Supp. 281 1-can
Ne'i7\0c; (iv Bpéweie rotofrrov <|>vr6v; id. fr. 155 "Iotpoc roiouiroccg rrupfiévovg
éfieuxerui I 1'p%ew ii 6’ dtvvbcg <I><'ioi<;.
9 KZ 65 (1938), 247; (1953), 572 n. 4; cited by Chantraine (1942), 217 n.1; cf. V. Pisani, Acme
4 (1951), 224.
*° Thumb—~Scherer (1959), 188.
184 II. Notes on Individual Passages
used in a similar sense to KOVi0OO\OQ (cf. H 374, ‘P 366), but the combina-
tion of both words in a nominative + genitive phrase is suspect; I should
be happier with 1<ovio0t7\0c_; tité)\7\r| or éie7\7\0t, making the first word
adjectival. '
Van Leeuwen’s do7\7ujg restores a good Homeric word, but its usual
sense, ‘all together’, ‘in a mass’, is not entirely appropriate here. The éie7\?t0t
root is, as we have seen, not out of place in the context of a rising cloud
of dust. If further conjecture is justified, one might consider the dative plu-
ral dékhntg (or -orig), ‘so from under their feet a dust-cloud rose to the
breezes’, i.e. for the breezes to carry this way and that. This would be close
to what is transmitted, and although such short dative plurals are rare in
Homer, they certainly occur (Chantraine [1942], 201f.).
F 41. The omission from a papyrus of a line that recurs elsewhere may
be accidental, but in this case the verse adds nothing to the wish forceful-
ly expressed in 40, and its absence seems a distinct improvement:
0t’i6’ 6(i)$7\8Q éiyovég T’ euevou éivotuoc T’ 6uro7téo6ou
F1’ oiirto Méfinv 1"’ ’é|.1t-SVOL1 Kai éiréijnov éi7\7\tov.
The use of if in a wish to mean ‘rather than’, without an expressed
comparative, is idiomatic; cf. A 117 [5o67\ou’ eve Aotov oéov Euuevou
Ii’ dfroitéofiou, ‘I’ 594, v 234, etc. But it may have prompted a later rhap-
sode to interpolate a line containing the explicit comparative Képoiov.
F 144. This line which identifies Helen’s two attendants as Aithra,
daughter of Pittheus, and Klymene was apparently present in all ancient
copies, but Aristarchus rightly queried it, saying that if this was meant to
be Aithra the mother of Theseus, she would be implausibly old, and the
line should be athetized, whereas if it was just a homonym, the line could
stay. He cited other examples of two people having the same name in the
Trojan saga.” The bT scholia record an absurd attempt to meet the prob-
lem by punctuating after A’i6pn and so separating the name from Httfifiog
Ouvoirnp. This will not do; and it is impossible to believe that ‘Aithra
daughter of Pittheus’ was meant to be anyone other than Theseus’ mother,
as a well-known legend told how she came to be Helen’s servant. Theseus
and Peirithoos, having made a common resolve to marry daughters of
Zeus, kidnapped the seven-year-old Helen, brought her to Aphidna in
Attica, and left her in the charge of Aithra, Theseus’ mother, while they
set off to Hades to try to win Persephone for Peirithoos. The Dioskouroi,
»»--~»--.»1
in pursuit of their sister, ravaged Attica and took Aithra captive. Theseus,
according to Hellanicus, was fifty years old at the time of this exploit,
so that Aithra would have been at least sixty years older than Helen.
Theseus’ sons Demophon and Akamas joined in the expedition against
Troy in the hope of recovering their grandmother.”
The mention of Aithra as Helen’s servant in this passage of the Iliad
need not imply knowledge of that story in all its details, but it surely does
imply the myth of Helen’s early abduction by Theseus. There is no hint
of this elsewhere in the Iliad, and if the poet had wanted to allude to it
here, he would surely have added a line or two to explain how Aithra came
to be at Troy. We have had one clear intrusion of Theseus at A 265, a line
absent from most sources, and F 144 is evidently another, added by
someone who knew the Aithra story and thus felt able to identify at least
one of the 0’tu¢iTro?\01 who accompanied Helen to the wall. In the original
poem they were anonymous, as is normal; cf. X 461, or 331, C 18, 84, u 18,
O 198, 303, T 601, ti) 61.
F 147. This line, which increases the number of Priam’s party from five
to eight, recurs at Y 238 in the genealogy of Aeneas, and there we learn
what is not mentioned here, that Lampos, Klytios, and Hiketaon are
Priam’s brothers. L. Friedliinder proposed deletion, and he was followed
by Wilamowitz, who argued that the poet was not likely to introduce the
trio here without identifying them as brothers of Priam, and that it was
equally unlikely that ‘der Dichter des Y’ would have made three ordinary
pensioners from F into brothers of Priam; ‘vielmehr diirfte der Vers im F
aus Y zugesetzt sein, weil es einern Rhapsoden passend schien, die Briider
des Priamos unter den Ratsherren um ihn zu versetzen’.‘6 Against this, it
must be noted that all three are mentioned individually in O (419/427, 526,
546/576) as fathers of Trojan warriors, Lampos being identified as a son of
Laomedon, that is, as Priam’s brother. They were, then, established in the
poet’s cast-list of elder Trojans, and it is only to be expected that he should
include them here in his list of Trojan elders.
I‘ 188. The three papyri available here agree with the medieval tradition
in reading e7\é)((-inv, and editors all adopt this. Our oldest witness, how-
ever, Strabo, twice quotes the line with éaévunv. This is the older form,
and it is found elsewhere in Homer (1 335); it is parallel to the third-per-
*5 Iliou Persis ap. Procl., cf. fr. 6 B. = 4 D.; Alcm. PMGF 21; Stes. PMGF 191; Pind. fr. 243
+ 258; I-Idt. 9.73.2; Hellanicus FGrHz'st 323a F 18-21 with ]acoby’s commentary, IIIb Suppl. i.
39-42.
16 L. Friedliindenjb. Phi/., Supp. 3 (1859), 821; Wilamowitz (1916), 294 n. 1.
II. Notes on Individual Passages 187
A 36. Zeus has asked Hera why it is that she is so hostile towards Troy:
what have Priam and his sons done to her that is so harmful? (31-3). Then
follows this sentence about her assuaging her wrath by going into the city
and devouring them raw, and the rest of the Trojans too. This has always
been taken as a statement, ‘then (by doing that) you would assuage your
wrath’. But as he professes to find her wrath inexplicable, it seems unna-
tural that he should be able to say confidently what would assuage it, with-
out even an ‘I suppose’. I therefore read it as a question: ‘if you were to
go into the city and devour (etc.) ..., would you then assuage your wrath?’
A 66 xetpéiv 5’. Ciig Kev Tptiiec
95 itéioi 62’: K6 Tptéeooi
The weight of manuscript evidence is for KEV in 66 and for K6 in 95.
Manuscript evidence does not count for much in such a matter, and it may
seem inconsistent of me not to print the same form in both places. They
are not entirely parallel, however, because in 66 the particle stands in the
biceps of the foot and in 95 it stands in the princeps. The biceps, being of
slightly longer duration, requires greater filling and, except in the first foot,
is seldom occupied by a short final vowel before mute and liquid.“ Kev is
preferable in 66 on this ground, whereas in 95 the lighter form is perfect-
ly in order.
A 115/195/205. In each of these lines Mevéactov tipfiiov ’Arpé0r; u’16v and
Mevé7\0tov dtpriiov citpxov ’A)(0u(I>v are variants, the weight of the tradition
being for the former in 115, for the latter in 195, and evenly balanced in
205. The first formula recurs at P 79, and Menelaus is called ’A'rpéog uiég
also at I‘ 37, Z 46, P 1, 89, 553; dpxog ’A)(otufiv does not appear elsewhere
in the Iliad. In A 115 cipxov ’A)(ouo3v would be awkward, because the pre-
vious line ends with u‘iec_; ’A)(0t1u3v. I assume that ’A1'p6o<; u‘16v is correct
there, as most manuscripts and p1 have it, and that I9 T have imported the
variant from 195/205. In 195 there was reason to change the formula
because Agamemnon is speaking, and it would be odd for him to, refer to
his brother as ‘the son of Atreus’. In 205, where Talthybius repeats Aga-
memnon’s message to Machaon, the wording must be the same.
A 132. The line is usually read as otutfi 6’ otiir’ ’i0uvev, 561 Ctoorfipog
oxfieg. (x1’>'rI’| 5’ ot1"iT’, of a goddess, begins a line at F 383. But I can see no
reason here for the emphatic aurfi, ‘she herself’, nor for oti’iTe ‘again’, and
I therefore reinterpret it as oturfit, ‘in just that direction’ (correlative to
501) and 0031(6), ‘it’ (the [3é7\og of 129). For a directional adverb in -111 with
’161’5va1v cf. Q 362 mix rroirsp o'36"i1r1rov<; re K(X’1 fiutévoug ’161’5ve1g; and T 479
‘ivcx re Evvéxovot Tévovrsgl dqnctfivog, Tfil 'r6v ye ¢>i7\ng 61631 Xstpog Enetpev.
A 159 (= B 341) del. Koechly. It makes the sentence into a specific refer-
ence to the truce between the Achaeans and the Tro'ans l rather than a
general statement about the efficacy of treaties confirmed by sacrifices. But
160-2 is a general statement about Zeus’ punishment of treaty~breal<ers;
the g nomic aorist in 161 cannot be reP laced b Y or understood as e q uiva-
lent to a future, as Zenodotus seems to have wanted.
A 160 f. z~;’i Tl'6[) voip re rot‘: ou’>ri1<’ ’O7u5umor; our érékeooev,
€1< Te 1<a‘1 amt 'rs>ie'i, cniv T6 l.l8Y(i7\UJl drréreroow.
The first T6 in 161 is supported by all sources. Bekker printed aé, and in
his note says ‘cf. 262, A 82, M 246, T 165’. Two of these passages, A 82
and T 165, may be held to support "rs in the apodosis to a condition intro-
duced by éi Tl'6p yoip (T6), but in both cases it stands in the combination
d7\7\di T6. The other two passages clearly favour the adversative oé.
A 170. off K6 Bdvntg mt uoipav/Troruov (iV(1Tf7\Ti01']1(_I, 6161010.
Both variants were known to Didymus; Aristarchus had Tréfuov, but most
copies had uoipav. A 263 provides apparent support for the former:
Tl'6T|10V dvanhfioavreg Eouv ociuov ”A'i60c_; éiow; cf. Pind. Nem. 10. 57 (the
Dioskouroi) Trofuov dtpnirrhoivreg ouoiov. Elsewhere in Homer Tr6Tu0<; is
more or less synonymous with ‘(fated) death’. dvomiurrknut in combina-
tion with it will be analogous to the use of this verb with 1<0u<ov oirov,
1<0u<6t 1ro7\7\o’i, or Kfioscx. The difficulty in the present passage is created by
the addition of [516To10, as rréruog [516To1o might be expected to mean one’s
fortune in life, the way one’s life turns out, as in Eur. LT 913i. ‘rivet nor’
’H7\é1<Tpon iroruov | éiknxs fitérou. This does not suit the Homeric use of
1r6T|.10g, and it is not something that one can ‘fulfil’ (civantunldvat) by
dying.
With the reading noipozv we could understand the phrase as ‘complete
your allotted span of life’. Cf. Hdt. 3. 142. 3 1'[07\1>1<pdiTn<; uév vuv
izljérflinoe uoipav rfiv éwuroii; 4. 164. 4 ’Ap1<eoi7\ewc; uév vvv éfiénhnoe
uoipav "rfiv éwvrofi.
A 177. This verse may be a rhapsode’s addition. It adds a gratuitous cru-
dity to a sentence which is perfect without it: the Trojans’ glee over the
failure of his expedition and the loss of Menelaus is a sufficiently dismal
prospect for Agamemnon, without his needing to imagine their dancing on
his brother’s grave-mound. Also it is suspicious that Menelaus is referred
to in the third person although Agamemnon is speaking to him. Lines
introducing Tu;-speeches (duos at Tu; e’iTreo|<ev, <’5<1)pa T16; 635’ einnt, ufi Trot:-5
Tu; eirrnot, and the like) are normally followed directly by the speech,
190 II. Notes on Individual Passages
without another line intervening (there are fourteen examples in the Iliad),
though there is one exception at H 87, where, as here, a participial clause
is appended giving the circumstances of the envisaged remarks.
A 337/369. The omission of this formulaic line by four papyri in the first
passage and by one in the second (although other papyri have it in both)
provides strong confirmation of Nauck’s suspicion that it was interpolated
by someone who thought that a more explicit verb of speaking than
veiiceooev (336/368) was needed to introduce the omtio ream. It is not: see
B 224, I‘ 38, E 471, Z 325, K 158, (I) 480. The interpolation is of a com-
mon type; see above, pp. 12f.
A 343. The genitive ooutcig has caused much difficulty. I believe the poet
wrote it under the influence of the phrase he had used in the previous line.
“It is proper for you uoixng KO£UUT81[)fiC_‘, CiWTlf)07\fi0(1t, rrpufrruo vérp 1-coir
ooufoq -- ” ((i(VT1f)07\E,iTOV) is the verb in mind, until a replacement expres-
sion is formulated, 0’u<ou0iCeo60v izueio.
A 483. The transmitted Irecbtixtzt is impossible as a pluperfect (contrast
the tense of Tr0(I>t5(xo1 in 484). Hermann’s Tre<1>151<r|1 (Opusc. II 44) is the
obvious solution; it was perhaps written in W before correction. Such is
van Thiel’s reluctance to depart even so slightly from the manuscripts that
he pleads in a defensive note, ‘aut coniunctivus brevivocalis aut forma per-
fecti in similitudine, ut H 633 opuipei (cf ]anko), P 435 éorfi K61 et saepe
dvuiyel in fine versus, cf Chantraine 312’. A short-vowel subjunctive is not
to be expected in a K-perfect (we should need metrically guaranteed
examples to support it); and the idea that perfects with present-stem
endings should be admitted especially in similes has just as little to recom-
mend it. dtvtéyz-:1 is a special case. In P 435 a papyrus gives éoffixni,
and this is to be accepted. At II 633 too -m is perhaps to be restored. A
scholium in A reports that Aristarchus read opuipel as against Eipwpev, but
there is no guarantee that the scribe has not adapted the reading to that of
his text, and that Aristarchus’ reading was not really optiipni (Thiersch).
E 5. All sources give iig T8, but this, followed as it is by a subjunctive,
would only be appropriate to a non-specific antecedent, ‘like a star of
autumn which most of all shines bright after its bath in Oceanus’. The
reference is in fact, as Aristarchus saw, to a specific star, the star of autumn,
i.e. Sirius, as X 26ff. make clear beyond any question.” A more satisfac-
tory reading would be ei’i'rs: ‘like the star of autumn when most of all it
shines bright, after its bath in Oceanus’. Cf. Sappho 34, ciiorepeg pi-3v duqii
I I ?\ 9 /' / T 5/ I
xcodxotv oekavvav I cup OUfUl<{)UTl'T01Gl cbotevvov s16og omrorot 'rr7\n601oot
udimora ltdiurrnt net». I make a similar suggestion below on E 399, though
there the verb in the dependent clause is in the indicative.
E 36/49/77. Alexander of Cotyaea, followed by some Byzantine scribes,
avoided the awkwardness of the syllable left short before Z1<otuotv5p- by
writing Kotuotvop.-..""’ Allen notes ‘cf. Nonnus iii. 39, xxiii. 222’, and he
could have added ‘xxii. 387’. However, it is far from certain that Nonnus
used this spelling, because for his text we rely on a single thirteenth-cen-
tury manuscript, Laur. 32. 16, which is given to conjectures and which
makes the same alteration at Hes. T/0. 345, where the rest of the tradition
has Zlcduavopov.
E 118. For Tcivoz-3 cf. 175, 219, 228, 238, and for the construction with-
out 56<; cf. Chantraine (1953), 317f.
E 141. For the spelling dtv)(r|0"r1v0u see West (2001), which supersedes
West (1998b), xxix.
E 178. Eat stands for €TtsoT1. Cf. Hes. Op. 749 and 754. The phrase is a
brachylogy for )(ot7\s1ri\ at 06013 ufivig, iitotv érrfit.
E 272. See below on (*3 108.
E 293. Zenodotus’ reading éfieotifin may seem plausible in the light of
661 aixufi 6% otéoovto (cf. O 542); but the aorist i;0(U)150n is first attested
in tragedy.
E 313. Aeneas has already been introduced as the son of Anchises and
Aphrodite at 247-8, and this inorganic line, which repeats the information,
may be a rhapsode’s addition. Certainly the point has to be made that
Aphrodite intervened because she was his mother, but this is done imme-
diateiy in 314 by saying that she threw her arms about ‘her own dear son’.
On the other hand, one cannot make it a principle of criticism that the
Iliad poet never composed a redundant line.
E 338. In this case there are other grounds for suspicion besides redun-
dance.
1. It is omitted by a fifteenth-century manuscript for no obvious
mechanical reason, though this cannot be taken as a significant pointer. It
is present in three papyri.
2. While warriors often receive body wounds through a garment or a
piece of armour, Aphrodite’s wound is in her hand, which we should not
expect to be covered by her 1réTr7\0<;, even though she is holding a fold of
it in front of Aeneas (315f.).
..
22 [Hes.] fr. 23a. 21. According to the Meydkott ’Ho"10u (fr. 260) Endymion, in love with
Hera, was deluded with a phantom of the goddess. The lines at 7\ 602-4 which say that the Her-
acles whom Odysseus saw in the underworld was only a phantom, not the real one, are also
intrusive, as was recognized in antiquity. On the phantom motif cf. West (1985), 134f.
194 II. Notes on Individual Passages
Kafifiuevog fiuara rrdvtot, I norrpog éuoio rrorrfip (E 118), 5 u’ €i5 tpéqpev fi5’
dt1'iTot7\7\ev. After that, it would make excellent sense to add ‘As for Ty-
deus, I do not remember him, because I was still little when he went off
to fight at Thebes.’ As it stands, the couplet seems to require us to fill in
that information mentally. Can it be that some genuine verses were lost
from the text at an early stage?
Van Leeuwen, who regards 222-3 as interpolated, writes: ‘Qui versus
hos inseruit, pronomen inv non ad 5é1r0to sed ad Foivt-3150 rettulisse vide-
tur.’ If the verses were truly interpolated, that would be a shrewd diagno-
sis: that a rhapsode, mistaking the point of 221 and taking it to mean ‘I left
him (Oineus) at home’, saw fit to add a note about what had happened to
Tydeus. But the reference to the Theban war is very much in keeping with
the Iliad poet’s interests, and if there were no mention of Tydeus, the
explanation of the Eeviot Trottpodicx would be even more elliptical than it is.
Z 237. The reading mipvov is supported by three papyri and by A.
tbnybv may well come from I 354 ~ A 170 Zicouoig re 'm37tac; Kai om/ov
’iI<0tvev/ikovro. It is less suitable here because the tree was outside the gate;
its place is in narrative about fighting close to the city. If the Trojan women
swarmed round Hector when he reached ‘the gate and the tree’, the impli-
cation would be that they came out of the city. But surely the poet imag-
ined them surrounding Hector as soon as he passed in through the gate.
‘The gate and the wall’ is therefore more appropriate. For the conjunction
of the two cf. 1" 149/153.
Z 252. Interpreters since Aristarchus have supposed that since Hector
meets his mother on arriving at the palace, and she is coming towards him
(izvotvtin), she must be on the way out. What, then, was the meaning of
Aoto5i1<nv éooivovocx? Aristarchus’ forced and artificial answer was that i-IQ
was to be taken with Aoto5i1<nv, the éivovoot being intransitive: ‘making
her way to Laodice’. Aristonicus’ laconic note in schA cites H 312 eig
’Ay0tuéuvov0t to illustrate the Homeric use of sic with a person, while
schbT cites a (Koine) idiom Troi eiooivetg; equivalent to Troi eioépxm; Lid-
dell-Scott gives no other reference for this usage, and there seems to be no
parallel for the simple tiivetv being so used. Besides noting this, modern
critics have wondered what Laodice is doing here anyway. She was men-
tioned at F 124, where Iris came in Laodice’s likeness to summon Helen
to watch the duel between Paris and Menelaus.
Ao:o5i1<nv éooiyovoa can only mean ‘bringing Laodice in’, that is, into
the palace. The sentence begins at 242 with ‘But when (Hector) came to
Priam’s fine mansion’, and there the poet goes off into a description of the
palace with its dozens of chambers. When he has finished that, he conti-
II. Notes on Individual Passages 197
nues ‘there his kindly mother came up from the opposite direction’. He
has not said a word about Hector’s passing through the hall, or about
where in this extensive building he is. What the poet has in mind is sure-
ly, not that Hecuba is coming out from the interior, but that she is arriving
at the palace from elsewhere at the same moment as Hector; it is on the
street that they meet, in front of the door. It is in accord with Homeric
technique that some explanation should be offered of where Hecuba is
coming from, or what she has been doing, and 252 provides it: she is escor-
ting one of her daughters home. Laodice was in the poet’s mind from F,
and although in that episode it was not strictly speaking the real Laodice
who accompanied Helen to the wall, the impression may have remained
that Laodice was out and about in the town. Or we may say that Iris dis-
guised as Laodice merges with the real Laodice, as in B 786ff. Iris dis-
guised as Polites merges with the real Polites.
Z 285. Aristonicus’ note explains Aristarchus’ reading cbpév’ éitep rtou,
and notes iivtoi at oivvoriooivreg vpo1¢>o0o1v “dTép1rou”. This reproduces
Aristarchus’ criticism of an §1<5oo1<; in which, in his opinion, an accent had
been wrongly placed. It is probable that this was Aristophanes’ text, which
was the first to introduce accentuation; we know of no others between him
and Aristarchus, and any that there were would for the most part have
been copying Aristophanes’ accents.
tiitsp TFOU is unacceptable, both because of the meaningless and mis-
placed rrov and because é1<7\t-:7\ot0éo0ou is left without a genitive to govern.
The objections to cirréprrov are that the formation is anomalous (for
drreprréog) and that it gives feeble sense; o'i?.',15g is by its nature dtreprnicg, to
say the least. I have therefore adopted Zenodotus’ <i)i7\ov fitop, which, if it
were the only reading attested, would be subject to no query. The
difficulty it leaves is that of explaining how cbpév’ drréprrov came into the
tradition in the first place and came to dominate it. All five papyri that
cover the passage had it: otteprrov 480a (Ptolemaic) and 413, dtépnou 1,
O.‘I'8lf)l71'0lU 400, qlrlepniov 582. In the cases of 400, 413, 480a, and 582, we
cannot be sure whether drréprrou or oitep TFOU was intended, but in the
absence of lectional signs the former articulation, as being the more
obvious, should perhaps be assumed. In the case of 1 (the Ilias Ambro-
siana), where accents and other lectional signs were added at a later date,
there are acutes on each of the first two syllables and a hyphen under
D71’. In other words, one corrector accented éitep rrou, a second restored
citéprrov, adding the hyphen to confirm that this was a single word.
Z 356. In all three places where there is reference to Paris’ misdemea-
nour (F 100, Z 356, and Q 28) ’AM=.E<iv5pou iivex’ dtpxfir; and dim; are vari-
198 II. Notes on Individual Passages
ants. The explanation is probably that each noun is original in one or two
of the three passages and was then introduced in the remainder by conta-
mination in some sources. r
In F 100 Menelaus speaks of the sufferings that both Achaeans and Tro-
jans have undergone éivek’ éufic; §pi5oc; Kori ’A7\eEo’tv5pou i-iv:-3K’ citp)(fi<;
(éirncg Zenodotus only): here the two nouns complement each other, ‘my
quarrel (with Alexander) and Alexander’s initiating (of it)’. In Z 356 Helen
speaks of the burden that has fallen upon Hector éivex’ éueio Kuvbg Kori
’A?\e<Eo’w5pou iiveic’ éitng (so Aristarchus[?], 190‘, Cosmas, (2: dpxfig 190*‘
400, Ap. Dysc., R): here the principle of parallelism favours éirng, whether
we take it as ‘me, who am a shameless bitch, and Alexander’s losing his
head’ or as ‘me the bitch and Alexander the dim’. The latter construction
may be possible in view of Soph. Ant. 533 Tpécbuov 515’ éitoi Kcimavototoioetg
Bpévtov, Eur. Andr. 103 f. ’I7xiw1 ozirreivéii Hoipig 013 voiuov d70\0’i riv’ éirotv
I fivoiyet’ 8‘0V(Xi(1V sic Bahtiuoug 'E7\évotv; see further my note on Hes. Op.
191 (West [1978], 202). The former, however, is more suitable at Q 28,
where ditng is again the better attested variant and the better sense.
Z 388. The statement that ‘she is arriving in haste at the wall’ fits badly
in the housekeeper’s speech; her role was to tell Hector in which direction,
or to what destination, Andromache had departed from the house, and she
has done that in 386. She cannot know that her mistress is now arriving at
the wall uouvouévni eixuiot, even if that was her condition when she left.
Payne Knight condemned both 388 and 389 (Heyne having thought that
386-7 might be a later variant of 388-9). But the information that Andro-
mache has her child with her is essential as preparation for the ensuing
scene. I therefore bracket 388 alone. There is no difficulty in uouvouévni
eixuicx being appended to 386 étrri mipvov E511 despite the intervening clau-
ses, but the interpolator presumably felt that a nearer main verb should be
provided.
Z 433-9. Against the authenticity of these lines Aristarchus made the
valid points that tactical proposals are inappropriate in Andromache’s
mouth, that her remarks about the vulnerability of the wall near the fig-
tree are unsubstantiated, and that the beginning of Hector’s reply, 441 ii
Kori iiuoi Tci5&; Ttoivtot uéiu-:1, viivou, relates to her representation of her per-
sonal worries up to 432; it is not a variant of 492 Tr67\etlo<; 5’ (iiv5psooi
I.1s7\r’joe1. Kirk adds, ‘it might also be argued that 431f., taking up 407-9 at
the beginning of Andromakhe’s speech, may suitably bring it to a ring-
form conclusion’, though he then discards this as ‘inconclusive and sub-
jective’. The fact is that the completion of a ring may mark the end of a
section of a speech, but in no way precludes a continuation of the speech.
Kirk proceeds to endorse M. M. Willcock’s conclusion that ‘probably the
II. Notes on Individual Passages 199
25 So Lesbonax, De figm/is 11: 1'o'5v 51$ Tfig ’Ao10't6og 'EMrivwv tori to 1'6: fniporrot -ta onw-
Tocooti evoz ootncofi Q Jrrtéoeow aitiarmai oovro'tTrs1v' éiol dip “Gov 'vr oé s” citvf: T06
7 I \ I \ J \ /- I I Y1 I I h
epoi, icon 'Opnpo<; “Kori 6’ Axtkeug rourov ye ]J0t)(I']l evi I<U5t0tv6lpr]1 epply’ do/r1fio7\no0u”
tivri rofi rolittoi.
200 II. Notes on Individual Passages
.,i-»-s»--_i-___-_
® 59. The absence of the line from several papyri suggests a concor-
dance—interpolation from B 810. The suspicion is reinforced by the consid-
eration that although the poet was composing hastily here, he was not
likely to repeat the half—line 'iro7\iJg 5"<’)p1Ju0tY§oc; optiipel so soon at 63.
(>3) 108. oiig Trot’ drr’ Aivaiow éképnv iuiotwpe/1 cbéfioio.
The passage clearly stands in a close relationship to E 272, Ttl) 5% 515’
Aiveiou fitimsv ufiotwpe/1 <]>6[5o1o, which records Aeneas’ acquisition of the
horses now in question. The main tradition in both places gives ufiottope
<b6[3o1o, so that the phrase refers to the horses. But elsewhere the formula
is used only of heroes (Z 97 and 278 of Diomedes, M 39 of Hector, ‘P 16
of Patroclus), and in E 272, at least, the verse seems more naturally filled
out with a formulaic phrase appended to the name of Aeneas than with a
characterization of the horses chosen for him by Anchises. In ® 108, on
the other hand, the emphasis‘ is on the horses’ quality; they have just been
described as émotoiuevoi rreoioto I icpcmrvét uoik’ €v0oi K051 iivficx Swaxénev
r’|ESIa cbélieofiou, and ufiottopt-; ¢>6[.’)ot0 can be seen as picking this up. If the
paradosis had given priotwpot here, no one could have found fault with it.
But as the dual is strongly established in both passages, we should probab-
ly infer that -~ if it is not original in both - it is original at least in one, and
was then introduced by assimilation in the other. If that is the situation,
® 108 has the stronger claim to tuiotwps. The formula will have come into
the poet’s mind because of the connection with the account of the horses
in E 265ff., but then he adapted it to refer to the horses instead of to
Aeneas.
(9 114. ’i<i>6iuo1 is the lectio dzfficilior, as there was a natural inclination
to assimilate the adjective to the following name, as indeed we have
’i¢»6ipog Zfiévt-:?\0g at W 511. Also the position of the Te is very strongly in
favour of ’i<i>91t1o1: see Denniston (1954), 517; Fraenkel on Aesch. Ag. 229f.
The only Homeric passage I can quote to parallel the order ’i¢6n1og
Zfiévekég T6 is Q 473 f., T(i) 62-: 615’ oiw, I iiptoc Aiironéotov Te Kori ”A7u<u1og
5C0; "Apnog, where I propose a conjecture below ad loc.
icbfiiuw would also be possible, continuing the duals of 113.
('3) 116/137. fivia otv0O\6ev1'ot is the usual formula (E 226, 328, A 128,
P 479). The appearence of the variant <1>o1vu<6evt0t here and at 137 is hard
to account for unless it is original; it was natural for OlYOt7\68VT(X to in~
trude. The poet’s motive for varying the epithet is obscure.
('5) 183. It is surprising that a plus-line found in a Ptolemaic papyrus but
absent from two papyri of the Roman period and from the whole of the
earlier medieval tradition should reappear in Eustathius and in manuscripts
of the twelfth and thirteenth centuries. It should warn us that ancient
202 II. Notes on Individual Passages
variants can surface in such quarters. Hector’s speech ends well at 182, cf.
L‘ 47. Leaf suggests that the line was composed in order to gloss otutotig
with ’Apyeiov<;.
E) 185. Aristarchus rightly observed that the four~horse team is un-
Homeric, as well as conflicting with the following duals (186, 191), and
that the names are borrowed from other Homeric horses. Leaf’s defence
of the line seems to be motivated by his feeling that ‘the speech would
begin very badly without’ the vocatives. This may be countered by citing
\P 403 (Kirk) and 443, where Antilochus and Menelaus address their hor-
ses without vocatives, beginning directly with dual imperatives, just as
Hector does here in 186.
(E9 189-~90. Aristarchus followed Aristophanes in athetizing 189, arguing
that horses do not drink wine, Kori iitt “9vu<’)g do/05v01” sic, péfinv yr-;7\o'iov.
Leaf writes: ‘the line in this case will be a singularly unfortunate expedient
on the part of an interpolator who was offended at Hector’s being made
to eat grain, as is the case if we omit it The wholepassage is too hope-
less to be remedied by a single omission.’ Kirk follows E. Delebecque, Le
c/aerial dams Plliade (Paris 1951), 59, in the eccentric interpretation that the
wine was mixed with the grain and not with water. ‘Willcock cites Colu»
mella 6. 30 for such special treatment for medical reasons.’ Kirk explains
Trteiv Eire Bvpog divufiyoi as a formula that is ‘here probably half-humorous-
ly reinterpreted so that the horses remain subject of Ttteiv (as though they
were drinking wine by itself), whereas the Bvutig belongs to Andromakhe.’
This is frankly preposterous. The second hemistich can only mean that the
mixed wine was for the horses to drink (so it was separate from the grain)
when they were thirsty. Now, it may be conceivable that a Greek horse-
lover might think to give his horses strength (cf. Z 261) and gratification by
lacing their water with a measure of wine, and that horses might acquire a
taste for it. Whoever composed the line evidently thought so. But there are
other objections to it: the awkward syntax, and the increased separation of
Trpotépoioi from ii’ éuoi. On the other hand, if it is excised, we are left with
the clumsiness of Andromache’s apparently feeding grain to her husband
after she has fed the horses. It may be that both 189 and 190 should go, 190
being an addition designed to clarify Trpotépoiot.
For wheat as horses’ fodder cf. Wilamowitz (1916), 46.
9 359. Diintzer’s suspicion of a rhapsodic interpolation may be well
founded. Not only is the line otiose as a whole, but év mrpizst votint is
especially feeble and irrelevant.
® 371-—-2. The grounds given for Aristarchus’ athetesis are that the details
of Thetis’ supplication are not needed here, especially as Hera, the addres-»
II. Notes on Individual Passages 203
see of the speech, already knows about it; nor do we ourselves require the
information. But the lines are not duplicated elsewhere in the Iliad, and
371 in Athena’s mouth has a fine touch of sarcasm. Aristarchus’ suspicions
may have been aroused initially by the absence of the lines from Zenodo-
tus’ text. But we know that Zenodotus’ text was characterized by a cer-
tain degree of abridgment. Absence of the lines from a papyrus would have
carried greater weight.
(*3) 385-7. Here again we have an athetesis by Aristarchus (and in this
case by Aristophanes before him) corresponding to an omission in Zen-
odotus. Aristarchus argued that the lines stood well at E 734--6 but are
pointless here; Athena rrpbg ofaoizv (i(V(X7\(X].l[3(JiV8l tfiv T[(XVT8U)(iO(V. At
390-1, again, he diagnosed an inopportune borrowing from E (746-7). The
whole passage certainly shows extensive repetition from E, and it may well
be that some of the repetition is due to concordance-interpolation by
rhapsodes or copyists. Diintzer’s deletion of 393~—~6 is also worthy of con-
sideration.
(E) 410. The omission of the line in a major part of the medieval tradi-
tion might be an accident due to the partial homoeoteleuton. Without it,
Iris’ arrival on Olympus is very abrupt. There is at present no papyrus evi-
dence.
® 420-4. Aristarchus athetized these five lines on the grounds that 420-2
are an unnecessary extension of Iris’ message, copied from 406-8, while
423-4 are inappropriate to her character, 013 votp oiv éirrev “1<6ov dt55eég”.
As to the first point, we expect Iris to repeat what Zeus said to her, and I
see no reason why she should not round off the threats in 415-19
(~ 402-5) with 420 (~ 406). But it does seem bathetic for her to add that
Zeus is not so cross with Hera because she is only behaving as she usual-
ly does. Those two lines in Zeus’ speech (407—8) could be taken as a com-
ment for Iris’ benefit rather than as an integral part of the message she is
to deliver. As to Aristarchus’ second point, it has some force. 423 f. are a
quite startling outburst from the messenger on her own responsibility, ‘in
sharp contrast with the courteous tone of Iris in O 200-4, where again she
speaks on her own account’ (Leaf). I have therefore bracketed 421-4 but
not 420.
In 423 Didymus, as reported in A, notes oiirwg “U15” 5161 T013 ii, T0 at
“dZ5e§:g” 51’ évbg 5 (5 ’Apio'r0ip)(o<;. This must mean ‘mi, not ooi’, and we
can infer that Aristarchus did not read Y’ after the pronoun. This is
confirmed by T’s version: “G15” éiveu T013 yé, Kori “(it5eé<;” 51’ évog S. Most
manuscripts give 0'15 V’, but C and G lack the particle. Hiatus after GL5 is
legitimate, and the insertion of y’ a common event in the tradition in such
cases.
204 II. Notes on Individual Passages
.,i_..,».--»»».---it
(5) too: ‘doch zeigt dieser Fall von “Entlehnung” einmal, wie wundervoll
auch die an sich weniger “richtige” Umdeutung wirken kann.’
I do not share this opinion. In the present passage the punctual aorists
bx T’ i='§<]>(xvsv and imsppoiyn are inappropriate, as the simile does not refer
to an event that brings the mountains and stars into view, but to a night
that is cairn and clear from the start; mountain scenery is irrelevant. The
lines obscure the picture rather than enhance it. They were absent from
Zenodotus’ text, though present in Aristophanes’ and Aristarchus’, and
perhaps in that of Apollonius Rhodius.” 558 is omitted by a papyrus. See
further Bolling (1925), 115f.; Nickau (1977), 120-2.
I 5 Bopéng is the only surely attested ancient reading and the normal
epic form, but the scansion is anomalous. W and some later manuscripts
offer Boppéng, probably from conjecture rather than tradition. Boppiicg
appears in the editions of the anonymous treatise Trepi Tpoirwv (iii. 211. 4
Spengel) where this verse is quoted, but the editions are not based on a
good knowledge of the existing manuscripts and it may turn out that the
paradosis here too had Booting, as in the parallel rhetorical texts. In the
similar case at ‘P 195, Bopéni Kori Zacbfiptot Kori i’>1rio)(s'ro it-3p(’x 1<00\oi,
Boppém again appears on the fringes of the medieval tradition, while in a
Ptolemaic papyrus (p12) Bopeou was roughly altered to Boppou, with an r|
added above the oi. These seem to be no more than sporadic attempts to
eliminate the metrical irregularity.
In the Hephaestion scholia the line is taken to have U U - in the first
foot and is classed as acephalous. But Wackernagel (1916), 151, points out
that this cannot be defended by the other Homeric instances of acephaly,
as this ‘kommt nur bei Wortern und Wortgruppen vor, fiir welche
Stellung am Versanfang gegeben war, wie bei den Vokativen ”Apeg, (bins
zcotoiyvnte, bei den imperativisch-konjunktivischen Ausdriicken iouev
KM501 |<?i15'rs, dem satzeinleitenden étretori, und iiberhaupt nie bei anapiis—
tischen Wortern’.
Since a dialect form of Boreas’ name with a long first syllable is known,
namely the Attic Bopp('iQ,” we must assume that the poet who composed
this verse and ‘P 195 scanned the word in this way, that is, with a length-
ened p, though he no doubt gave it the Ionic form of the ending, -fir; or
-ér|g.32 Whether or not he wrote it with a double pp, we should do so as
the proper way of representing the intended pronunciation.
from them I would hand over to Agamemnon; and he just sat there and distributed a
small amount while keeping the larger part for himself.
Line 320, pointing out that the doer and the non-doer alike must die, is
irrelevant in this context: what concerns Achilles is that they do not re-
ceive duly differential rewards from the chief commander. The verse is evi~
dently a gnomic interpolation of the sort not uncommon in tragedy. The
rhapsode responsible failed to follow the tenor of the argument, and was
probably misled, as van Leeuwen remarks, by the reference in 318 to an
ion poipa awaiting the man who fights and the man who stays aloof from
the battle (cf. Callin. 1. 8—15).
I 327. At E 486 we had ciuuvéuevou tiipeoow, with a papyrus variant
ooipeoow, meaning ‘defend their beloveds’. Here we have what is evi-
dently the same word in the genitive, appearing as ooiptov in the main tra-
dition, but with a variant (i)p€oJ\/33 recorded in the scholia in A and T. The
accompanying possessive adjective ocbarspoiwv indicates that the noun is
feminine. The variation between ooc and w in both places is simply a mat»
ter of uncontracted versus contracted.
Aristonicus’ note reads: on Eicxpoi on ttbv drv25pu'3v Kai vvvotuctfav
éufltiou. vfiv 615 Aéyei ttfiv vuvcxucdfiv. Aristarchus (if we assume his teaching
to be reported here) apparently took ooipwv to be genitive of 60£pOl, the
same word as elsewhere means ‘lovers’ talk’ or the like.“ That word,
however, is otherwise masculine, and it could not have the dative in -6001
which we find at E 486. ocipcov and ooipeool together imply a stem
6(xp- with nominative singular iiotp. The Hesychian entry 0 8, iiotpotq V01“
pong. oi at vvvoiikcxg, accords with this, and so does the accusative singu-
lar tiipot in a Hellenistic choliambic (Powell, Coll. Alex. 214. 12).
The variant tiapétov, however, does not. It would point to an a-stem,
nom. iidipn (< "'o0ip0t), or an s-stem, nom. "'ti5pnq; the dative in ~eoo1 would
be compatible with the latter, and Hesychius’ iiotpotg, if reaccented
paroxytone, with the former. But in that case the ooiptov read by Aristar-
chus would have to be taken as an error for "'ootpéwv.
The weight of the evidence is for nominative 601p or contracted 63p. This
suits both variants at E 486 and the ooiptov of the paradosis here. Only the
variant in the scholia, tispéwv, is anomalous. It should presumably have
been dipwv.
33 So T; A gives opéwv, but the etymology given, time T013 ouvwpeiiv, K0186 Kai ouvtopifiotg
érfi tdiv 'imrwv, confirms the expected omega.
3‘ Hes. Th. 205; Hymn. Apbr. 249; Hymn. Ham. 23. 3; later poets. From it is formed the
verb ootpicto. '3
208 II. Notes on Individual Passages
Argives: do not find fault with their message or their feet; before this,
iriowever, you could not be blamedfor being angry.’ ‘Their feet’ is supposed
to stand for ‘their coming here’, but such a usage is alien to the epic lan-
guage, and reads grotesquely. The rest of the line = X 59, where it better
fits the logic of its paragraph. This is surely a rhapsode’s addition.
I 654. Manuscripts and testimonia are more or less unanimous in giving
the synaliphe "rfiufii; of the copies I cite individually, only R has Tfi éufi,
while O has T’éufi. Editors since Wolf have generally printed Tfil éufii.
There is a similar case at A 608, where the evidence is heavily in favour of
'rri>uu31; there we have a papyrus (pl), which writes Twiuwt. See also West
(1998b), xxxv. I see no reason why the contractions should not be attri-
buted to the original text. See the collection of material on Homeric cra-
sis and synizesis in Chantraine (1942), 84 f.
I 657. Aeiqravteg and orreiooivreg are ancient variants here, and the tra-
dition is similarly divided at at 0 426. Both verbs are guaranteed as Home-
ric in the sense of pouring libations,” but as ofrévoto is the one usual in
post-Homeric authors, the other is to be preferred as the dzfficilior lectio.
K 54. Aristarchus’ reading Trotpot vfiotg, ‘along the line of ships’, is what
the sense requires; cf. I 657, A 805. Naber conjectured the same at K 136.
K 231. 6 T7\I‘i].l(.0V ’OE3uoe15g is transmitted here and in 498. The use of the
article is suspect, but we find it similarly in 363 r’]?5’ <5 TrT07\iTrop9og
’O5uooe15c;, 536 Kori 6 rparepog Aioufionc, B 278 ave 6’ <3 TFTOMTFOQQOQ
’O51J0"oe15g (ancient variant). In all these places it is easily eliminated, except
at K 536 (rat Kpatepog? Nauck). In the present formula some ancient
readers may once have articulated as 5’ othfiuwv, cf. Eitkog, ofltéto.
K 258. Opinion diverged in antiquity on whether (’Ji7t(7\)0c|>ov should be
written with one lambda or two. A single one would be appropriate if it
is a case of simple metrical lengthening of (ii-, as in dmoiuottog; but dkocbog
need not involve a sequence of three short syllables, and on normal prin-
ciples is therefore not entitled to metrical lengthening. More likely the poet
lengthened the consonant, on the false model of €7\7\0t[3e, etc. In this case
our general orthographical convention calls for double lambda.
K 321. Didymus reports oiirtocg, dvoioxeo 6161 T013 6 ai ’Apiotdip)(0v. No
variant is attested, but Didymus must have known one which did not con-
tain an omicron. It is usually assumed that the alternative was drvcioxeu,
but it is at least as likely to have been dvoioxeg. The active is common in
Homer (as elsewhere) in the sense ‘hold aloft’.
K 341. Most copies known to Didymus seem to have had TlC_I,, but he
also knew the reading T01, which later dominates. It may be due to the par-
allel of 477, where the deictic particle is more in place. Tu; is appropriate
here; cf. v 380 éi?\7\o<; 6’ 0ci’$"r€; Tu; o1‘5'r0<; dvéorn uoivre15so6ou. .
K 451. 7tTo7\euiEwv is strongly attested here. Normally the formswitll
7TTOA- are used only where metrically necessary, which is not the case here.
But perhaps the Tl'TOA~ form was inseparable from the future/aorist stem in
-1’§—. Cf. the apparatus at B 328, O 179, Y 85, Q 667.
K 489. Aristarchus and the vulgate agree on the impeccable optative
irfitfifiele, but the ancient variant Tr7\ri<Ԥ0to1<e represents a genuine Homeric
construction (cf. Q 752) which is clearly the dzfficilior lectio. It is remark-
able that it surfaces in the excerpt-manuscript Y (cf. above, p. 152).
A 13-14 (= B 453-4). Zenodotus’ omission of these two lines need not
be significant. It may have encouraged Aristophanes and Aristarchus to
athetize them. Aristarchus argued that they were in place only in B, where
the army had been intent on going home. But, as the scholar behind the
bT scholium observed, iii-3 véeoi-)ou is apposite here too, as the previous
evening, following defeat in battle and the failure of the appeal to Achil-
les, there had been talk of abandoning the war and going home (cf. I
esoft).
A 35. Some grammarians before Herodian (Ptolemy of Ascalon, Tyran-
nion ?) interpreted the transmitted AEYKOI as Aevroi’. The genitive certain-
ly makes for a more logical antithesis, and the hiatus after nominative
7\8UKOi is weak. But the genitive should appear as -01>; even in the eighth-
century ‘Nestor’s cup’ epigram (CEG 454) we find |<ot7t7t1oTE;l<1>0'tlv6
’A<1>pooi1"e<;,, not Kamtiorecboivoi’. The nominative gives the sense ‘It had
twenty bosses of tin, which were white, while in the middle there was
(one) of dark cyan.’ This is better than ‘bosses of tin, which was white’.
A 51-2. In 47-50 each warrior tells his fivioxocg to keep the horses back
there at the trench, in good order, and ‘they themselves’, fully armed,
moved forward with vigour and much shouting. They are surely out in
front of the trench and the horses, on their way to meet the Trojans, with
whom they engage in 67ff. They are evidently the subject of <])66tv
Koounfiévtsg in 51, and ’tTfTl'fi8Q in this line and the next must refer to the
fivioxoz. Cf. sch“ ‘P 502, commenting on Diomedes’ being called a
fivioxog: (inc “0potoi>v fivioxov q‘>opéov1'eg|"Ei<'ropoi” ((5) 89f.). ti ofav éitortov
ti TO0Q I’]Vl6)(01)C; 1-tori “i1r1rfi0tc;” év Tfii A 1<oi7\ei;
The meaning seems to be, then, that the warriors were ready at the
trench before their drivers came up with the horses and chariots; the dri-
vers came a little behind them (52). If uévot is read in 51, it can only be
II. Notes on Individual Passages 211
understood as ‘they were far ahead of them’, which, besides being an odd
use of uévot, stands in awkward contradiction to okivov. Nor is a genitive
of comparison after rbfidivto found in Homer; it is rare anywhere, but I find
one example (in Ionic) at Hdt. 3. 71.5. However, it seems we must swal-
low these difficulties, as the alternative reading uefi’ imniwv cannot be
understood at all.
A 62. Other editors have followed the paradosis and read oiihtog
ciorfip. oiihlog does not occur elsewhere in Homer, though later poets use
it as an equivalent of the Homeric oi’§7\og or 67to6g, ‘destructive, baleful’.
The ‘baleful’ star would presumably be Sirius, the bringer of the fever
season (X 26-31). But balefulness is not so appropriate here as in X - the
poet is describing how Hector looked to an admiring ideal observer, not
how he looked to the Achaeans -- and if Sirius were meant, it would sure-
ly have been identified more clearly.
With the early variant AiiA1OC;, it becomes the still more brilliant Even-
ing Star, which the poet also uses in simile (X 317f., of Achilles’ spear).
This too would be a Homeric hapax, but one yielding a more precise and
apt sense. It is echoed by Call. fr. 177. 5f. = Supp. Hell. 259. 5f., oiotfip
(x1’5?t10Q.l i5<; oufiufiv eioiv fin’ fiekiov (the restorations are certain), and Ap.
Rhod. 4. 1629f. dorfip I ou’57\1og; cf. Rengakos (1993), 133f. Bergk (1872),
860 n. 162, accepted this reading, and offered a shrewd explanation of the
variant: ‘Volksmiiliig ist auch die Benennung des Abendsternes otiihog
dotfip, der in demselben Sinne auch éTri<1)0iTv1og genannt wurde; wenn
Andere oiihtog oiotfip lasen, so verbirgt sich vielleicht der durch Krasis
verschmolzene Artikel wi57uoc;.’
A 109. Trotpot oiic; with hiatus is suspect, especially as at Y 473 we find
Kort’ oiig. Heyne proposed substituting cxi’§re nap’ for otii rrapét. Wilamo-
witz (1916), 213 n. 1, more boldly suggests ’Avfl(b<iTnv at Jtotp’, on the
assumption that ”Av'r1(|>og (101, 104) is a short form of ’Av'r1cb0i'rng and that
the two may alternate. Observing that an Antiphates is killed by Leonteus
in M 191 immediately after a Hippomachos son of Antimachos, he as-
sumes the names to have been inspired by the Antiphos or Antiphates of
A 109 and the Hippolochos son of Antimachos of A 122. ‘Man sieht, wo
der Verfasser des M seine Namen her hat.’
A 146. Aristarchus’ Tr?\fiE0tc; seems the older reading, Tufiiotg a moderni-
zation. Cf. the variants at ‘P 120, 6 507, K 440. The two are graphically
close: HA ~ TM.
A 272. The vulgate reading eta’ oouvou involves an unparalleled elision
of nominative plural -011. The Ptolemaic papyrus 432 produced étu’
ooiivn (with a different verb, teipev for oiivov), and 1205 has
212 II. Notes on Individual Passages
3‘ S. West (1967), 96, came to a different conclusion, preferring to follow Cobet in deleting
the line and the particle 6’ in 269.
37 This was first noted by Hentze ap. Fick (1886), 81 and 482.
33 Perhaps better ’iv or iv (Schwyzer [1939], 570).
i
5
‘i .
i
E
II. Notes on Individual Passages 213
15
fr
i
sis on itué; but ‘if I die’ for ‘when I die’ seems objectionable. Wackernagel
1 (1955), I 261, cites Q 768 (and some Latin uses of 52'), but this hardly
justifies the present instance. It was open to the poet to write ot1’>Totp iip’,
ei’Jts Boivua. But we cannot introduce that by conjecture.
A 488. The series of parallels cited in the apparatus‘ favours the thinly
Et supported i‘.7\u5v, but perhaps the imperfective main verb Efiotve made i-ixuov
>
‘?
the poet’s choice here.
1’
1
an.
A 540-3. It is sufficient to quote Lachmann (1847), 39:
A 521-~39. Der dichter des liedes liilit den Kebriones, der das wiiten des Aias sieht,
,E...
3:
Hektorn vom anderen ende her in seine niihe fahren. er springt hinab und beginnt den
1
1
kampf.
Doch wohl mit Aias? o nein, sondern in drei oder vier versen 540-543 erfahren wir
>._ dali Hektor den Kampf mit Aias vermied. wenn doch alle interpolationen so deutlich
und auf richtiges urtheil gegriindet waren! der interpolator fiihlte was hier folgen
’Y
muste und doch noch lange nicht kommt. auf den letzten dieser verse, welchen die
handschriften nicht haben, ohne zweifel weil ihn die Alexandriner tilgten, passt der
folgende echte wie die faust auf das auge. ‘er mied den Aias: denn Zeus gonnte ihm
E
nicht mit einem besseren manne zu streiten: Zeus sandte dem Aias flucht.’ Nicht
€.
I-Iektor vermied den kampf, sondern Aias floh, sagt der dichter.
A 559. é0ivn(1) must be the subjunctive of the perfect Ecivot found in
Hesiod (Op. 534) and Sappho (fr. 31. 9), not the aorist indicative. Cf. West
(1966), s2.
1
i
3’ On Hes. Op. 174 (West [1978], 196); further examples are Eur. Or. 1498, Call. fr. 75. 20.
214 II. Notes on Individual Passages
cooped up (38, 72); at 89-90 we heard that the division led by Hector was
eager to break through the wall and fight at the ships, and at 106--7
that the whole army was confident that they would no longer be held
(oxrjoeofiat) but would fall upon the "ships. It is certainly disconcerting to
read, a mere five lines later, that Asios, still in his chariot, néaaosv
vfieoot Bofilotv. If Naber’s emendation Tra7t0io0u is accepted, Asios’
approach to the ships is presented as the prospect present to his mind, and
this will be on a par with the uéuocootv of 89 and the itbotvto of 106. In
113ff. it is made clear that he is not going to succeed, and indeed that he
will not return alive.“ If the emendation is rejected, one might account for
Tf6?\Ot06V on the assumption that the passage is a remnant of the earlier stra-
tum of composition, before the wall and trench had been introduced.”
However, we read almost immediately that Asios found a gate still open
in the Achaeans’ fortification (120).
In 127ff. we are told about the defenders of the gate, the two Lapiths,
Polypoites and Leonteus. At 143 ff. they charge out in front of the gate to
fight. At 153 a verse awkwardly appended (and adapted from 135) informs
us that they fought trusting in ‘the army above, and their own strength’.
The reference to ‘the army above’ is clarified, though not made inoffen-
sive, by 154—-8, which explain that the defenders on the wall were hurling
rocks at the enemy, as thick as snowflakes. Thus far all is in reasonable
order, except for the clumsy line 153, which is better bracketed.“ And
from 182 onwards, where we are told of Polypoites’ and Leonteus’ suc-
cesses, the narrative is straightforward again, except that Asios has disap-
peared from view, not to reappear until the middle of N.
But between 158 and 182 we encounter a series of three unhappy para—~
graphs, all condemned by critics ancient or modern. Firstly there are the
three lines 159--61, in which the snow-storm of rocks descending from the
wall is transformed into a two-way bombardment, mounted by both Ach—
aeans and Trojans.“ This spoils the emphasis on the vigour of the Achae-
an defence, which is what leads Asios in 162-74 to complain to Zeus that
he has deceived the Trojans.
Asios’ speech is itself problematic, firstly because it is not obvious to
what promise by Zeus he is referring ((5) 170-—6?), and secondly because he
‘“ Line 117 tells us that he will meet his death at the hands of Idomeneus; but this does not
come about until N 384-93. Cf. Leaf (1900--2), I 524. Perhaps 117, or 116-17, is a secondary
addition.
*2 cf. CR 19 (1969), 25st.
43 Diintzer (1872), 492-3, condemned both 152 and 153. But there seems nothing wrong
with 152.
"’ Diintzer, loc. cit., condemned these three lines too.
i
218 II. Notes on Individual Passages
merely complains and does not ask Zeus for anything; this is not very well
followed by 173 t‘b<; Eoott’, o1’>oi-; Aloe "rre'i6e cbpévot tou3t’ dwopeiiwv. How-
ever, it is right that something should be said about Asios, indicating that
his attack has been brought to a standstill, before he disappears from the
scene. And if 159-61 break a logical connection between 127-~58 and
162 ff., this means that 162-74 at least belong to an earlier form of the text
than do those three lines. I incline therefore to retain 162--74, despite the
faults.
Zenodotus’ text apparently went straight from 174 to 181 or 182 with-
out the intervening lines. As often, his omission prompted Aristophanes
and Aristarchus to look askance at the verses in question. Aristarchus’
athetesis, according to Aristonicus, was of 175--80, and his grounds were:
1. What are they doing fighting at gates, when they have not crossed
the trench?
2. Line 176 (dpyakéov 56 pa is silly, as the poet has not yet said any-
thing about the teichomachy.
3. What is the fire (177) doing, since it is only at O 718 that Hector,
having at last reached at the ships, calls for it?
4. It is eiinfieg to say that the pro—Greek gods were distressed at their
setbacks.
The first three of these arguments, at least, are of some weight. But it is
hard to see why a rhapsode should have interpolated lines that seem so at
variance with the situation. I suspect that we have here a fragment of com-
position originally meant for another setting, in which the trench had been
crossed and the several gates in the fortification were being attacked with
fire. The poet put it in here to keep his audience in mind of the general
situation round the wall before he came to the Lapiths’ titv6po1<1'(xoi0n.45
The line and a half about the gods’ distress look as if they were to lead on
to a speech by one of the gods and perhaps some attempt at intervention
(cf. N 10ff.?). In their present place they serve the purpose of reminding
the hearers of the pro-Achaean gods’ existence; they cannot come into
action until the Trojans are really on top.“
According to sch 175-—81a (Didymus?) and b (Pius’ reply to Aristar-
chus), Aristarchus also athetized 181, objecting to the poet’s use of the
term Lapiths for Polypoites and Leonteus, although in 128 they were cal-
‘5 Von der Miihll (1952), 206f. He remarks of 176 that it is odd, but that ‘geracle dies sieht
nicht nach Rhapsodeninterpolation aus’.
‘*6 Schadewaldt (1938), 114 n.1.
II. Notes on Individual Passages 219
led sons of Lapiths. To us what seems offensive about the line is its
abruptness- after what precedes. I suspect that it originally stood following
the snow-storm simile:
143--53. Polypoites and Leonteus charged out from the gate.
154-8. At the same time the defenders on the wall hurled rocks,
which fell like the snowflakes that the fierce wind, dispel~
_ ling the clouds, sheds thickly on the earth:
159 63c; 'r<i3v er pqetptibv [3é7teot béov <éEorret0t 1r0>Otda'
181 oinv 6’ iifiakov Aontificn trohenov mi E>n'i0tfita.
If this is right, it confirms the impression that the whole passage has been
worked over, with blocks of text (162--74, 175--80) being repositioned and
breaking the original continuity 159 —> 181, whereupon 159 was remod-
elled and developed into the present 159--61.
M 283. Motoiivtot, correctly contracted from -oevtot, was mis-diagnosed
in antiquity as showing Attic contraction of -éovta, and this was replaced
by the neo-Ionic contraction -~e13vT0t. It is interesting that this has invaded
almost the whole tradition.
M 318. I have retained dkhneig, the spelling of the vulgate. Some edi-
tors adopt o’u<7\eée<; (M V), but the original text is unlikely to have had
three E’s, nor are rhapsodes likely to have pronounced them. Prehistoric
"‘tiu<?1eFé(l1)eg may have gone to "'d1<7\e/-_€c, necessitating compensatory
lengthening of the second syllable. This might have appeared as du<?\e1e’ig
(C G°), but -1<7\r|- was favoured by various analogical pressures, such as
the case—forms ’Hpou<7\fiog etc. (though they are anomalous themselves). At
K 281 we apparently have é1'>'K7\e‘i0tg (but with the last syllable in Kongo
before a vowel), and Lobeck"7 on this model arrived at tikkeieg. But
whereas in ”’815I<2\€(F)é(XQ the as might contract before the t-:01, in the
nominative ’i‘du<?\c/-_t-‘:(l1)et; we should expect the the to contract sooner than
the 1-3]-5. So the outcome might well look different.
M 342--3. Aristarchus and the vulgate are for the singular A’iotvT0t in
both lines, but clearly the dual is right, not only in view of 344 (pace
Hainsworth; cf. Denniston [1954], 368) but also of 353 f.
M 349, 362, 364. The variant in p60 (and at 349 also in D), Tshauufivtog
(6i7\1<1p0<;) uiog for Tehauuivtoq (oihctpoq) Aiotg, is interesting and may be
right. We find Tc7\0ttlt5v1ov uiov at A 563, 591, N 67, and P 115, always
with Aiotvtot in the preceding part of the line. In M 349 = 362, however,
the Telamonian Ajax is being distinguished from the Aiante as a pair,
and Aiotg seems easily dispensable. Besides, this is the only place where
Tchautévtog is followed by dmctuog: the influence of the formula
Msvottiou (vel sim.) dlxtuog u‘t6c; might have caused the poet to coin for
the nonce Te)\ott1u5v1o<; tii7u<tuoc; vicic as a substitute for an unmetrical
Tekautfivog éikxtuog viég. Te7\otuu5v1og éi7u<1poc; Aiotc is anomalous inas-
much as éi7u<1uoc; does not occur elsewhere in the Iliad agreeing with a
proper name. It may be that uiog is the right reading in 349 = 362 but not
in 364 (where p60 in fact has otog corrected to AIOLQ).
It must be borne in mind that p60 is a text especially full of wild mis-
takes; and the variant could be explained from the frequency of the
tii?u<1po<; viog formula. Nevertheless, it would be surprising if mere negli-
gence had produced such a plausible result. ‘
M 372. (Cf. West [1999a], 187.) Pandion seems to be an Attic,inter—
polation, similar to that of Aithra and Klymene at I" 144. Cf. Wilamowitz
(1884), 245 n. 7. He performs a quite needless service,“ and is disregarded
in the following narrative. ‘
M 373-7. Lines 375-6 are explanatory of érrstvouévoiot 5"ii<ovto; 377 is
the real apodosis to the temporal clause. So perhaps otiv ta’ i-:[30‘0tov1'o, as
in A 447, (5) 61.
M 426. The line was justly condemned by Payne Knight. Bociotg in 425
must be taken as a noun, as we cannot have an adjective at line-end wait-
ing for its noun to come. So dtorrioag and 7\0uorj'1'ot are in apposition; the
pattern resembles others listed by Bekker (1863), 231, where a more gener-
al term in one line is amplified by more particular ones in the next.” But
does Boeiotg have the status of a general term covering the two particulars?
What is the point of the addition? It is surely a gloss by a rhapsode to
whom the substantival Boeing was unfamiliar. The line also occurs in simi-
lar circumstances at E 453, but in the interpolation about the phantom
Aeneas (449-53). The poet who added those lines may also have added this
OI1€.
48 Not justified by Teucer’s wound (E) 324ff.). If he is sufficiently recovered to draw his
bow, he will certainly have no trouble in carrying it.
4" E.g. Z 47 f. KE.1[Jrj?\tOt :<e“1tou,| )(oz?u<6g re Xpoodg re rroittiicpnrég ts oionpog, 321f., I 660f.,
K 75f., II 766f., X 468 f.
II. Notes on Individual Passages 221
5° Lachmann (1847), 46. Actually the line is more fully identical with E 304 and T 287; at M
383 8 at pw oéot rroimte K011 oiog is a variant (AW). Nitzsch (1862), 132 n. 2, condemned both
449 and 450.
5’ For which, however, cf. P 725 oiipovrotg; s 393 dtpfieig.
222 II. Notes on Individual Passages
write Ai, the contraction of Ari. This is not otherwise found in epic, but
occurs in CE G 419. 2 (Melos, late sixth century) and regularly in Pindar;
the conjunction Ai ¢>i?\ov would recall the common name Aic|>17\og. The use
of an otherwise un-Homeric form would add to the linguistic and stylistic
peculiarities of the whole section (cf. 225 dvouetotl, 285 értetooiv).
N 234. A subjunctive is expected, and the spelling uethniot in p10
implies this interpretation.” No subjunctive of the form in(t)o1 occurs
elsewhere in Homer, unless perhaps at u 253 (simile), though it is certain
at Theognidea 94 and there seems no reason to exclude it. It is best writ-
ten without the iota in the suffix (cf. West [1998b], xxxi), which intruded
after the 11 of -r| 01 was identified as the subjunctive -r|1. The accent will be
properispomenon, as from *ie‘:—no1.
N 251--2. For the accentuation of fit-'3 fie in the disjunctive question
see Liddell—Scott s.v. ii (A), A.II.
N 424. The construction (loousvevg nominative, uevog accusative)
troubled ancient grammarians. The scholium in T and G informs us that
some read ’I5oucve13g as a contracted genitive; we know that Choerobos-
cus took this view, and he may be the ‘some’ referred to. Others, the scho-
liast continues, took uévog to be an accusative doing duty for a genitive,
and they quoted H 471 odixev Tnoovionc dyéuev uéi-)0 Xeimu uérpot as a
parallel. A third interpretation is laconically represented by the words di
at u€;vc<‘ieto, uevdhwg érrefiuuet. Some, in other words, detached uévot from
uévocg and made it adverbial with the following‘ie1'o. But how did this help
with the problem of ’I5otu-zvz-313g 5’ 013 Afive uévog? Presumably because it
allowed uévoc; to be taken as a real genitive; and I suppose this was done
by invoking the theory of the dpxaia onuotoiot, the old script in which
MENOZ could stand for uévovg. That this was an un-Homeric form of geni-
tive (for uévsog) was overlooked by the ingenious critic responsible for
this explanation.
N 495/542. The dative pronouns 0'1 (when not enclitic) and 5:01 are
usually accented perispomenon. As regards oi, at least, this was the doc"
trine of Apollonius Dyscolus (Pron. 81. 3) and Herodian (sch. I 392; II
845. 6 L.; Io. Alex. 23. 9; Chandler [I881], §731). Apollonius, however,
records that Dionysius of Sidon wrote o’i, bringing the accent into line
with éuoi and ooi. Dionysius seems to have relied on theory, Apollonius
and his son on tradition. But why should oi have come to be so
*~m~w»—;
52 The reading of p9, uefiemoi, would give an aorist subjunctive, and this is accepted by van
Leeuwen. W appears to have psfiénot.
224 II. Notes on Individual Passages
N 610. The variant Kori in p10 has been known now for a hundred years,
but editors have stuck with the familiar bis. Allen—-Monro and van Thiel do
not even mention the variant. But while many (indeed most) papyrus
variants are mere aberrations, sometimes they can produce something
superior to the medieval text. So it is here. With the medieval text (and
p60), the ‘he’ in 609 who rejoices and anticipates victory must be taken as
Peisandros. But it is absurd that he should react in this way when Mene-
laus’ shield repulses and indeed breaks his spear.“ With the papyrus’ 52$,
the correct sense is restored: ‘and he rejoiced, did Menelaus, and drawing
his sword ...’.
N 832. The line is repeated from (*3) 380 after the cue of 831 ~ ® 379.
The repetition could be due to the original poet, but I think it more like--
ly due to a later rhapsode or copyist, because the phrase ‘Ireotiw érri vnuoiv
’A)(ot1t‘6v has much less point here, in relation to Ajax’s anticipated death,
than in (9, where it refers to Trojans. The expression Tpoiwv Kopéeig Ktivotc,
fie’ oitovolfig does not need to be filled out with the datives squat Kori
ooipxsooi, as is clear from P 241 ESQ, re toixa Tpofiwv icopéei mivotg fits’
oiwvoiicg.
E 37. Ancient grammarians seized on oipciovtt-:c; as the paradigm Home-
ric example of a desiderative form in —oeito. It is the only one in Homer,
and it is suspect because there is an older attested variant which makes bet-
ter sense: the Eitp’ diidvreg reported from Zenodotus.“ The poet’s need is
to justify the fact that the wounded heroes, Diomedes, Odysseus, and
Agamemnon, are only now taking note of the Trojan incursion through
the defence wall. He has spent six lines (30~—6) explaining that their ships
were the furthest away from the fighting. To continue ‘So they all came to
see the shouting and the fighting, leaning on spears’ is simply a non sequi-
mr. With Zenodotus’ reading we get what the logic requires: ‘So they were
late in hearing the shouting and the fighting, and (now) came ...’. There
53 Faesi is forced to refer the reaction to the previous move in the encounter: ‘er (Peisand-
ros) aber hatte sich schon gefreut und auf den Sieg gehofft, als Menelaos (605) ihn verfehlte’.
This is imP ossible. anko su PP oses that the foolish Tro'an J ‘is P leased at his hit on Menelaos’
shield and antici P ates victor Y Peisandros’ rnis P erce P tion makes it almost comicall Y a P t that
his eyeballs pop out.’ (1)
5‘ Aristarchus was unsure whether to articulate Zenodotus’ text so, or as owe: icivrag, but
the latter is out of the question: there is no such form as owoi. Ptolemy Epithetes, who was a
pupil of a pupil of Zenodotus and made a business of criticizing Aristarchus, claimed that Zen-
odotus’ readin 8 was 0:3 \l1 otuovteg, and that this
H made sense which it does not . It is P ossible
to imagine a visual ambivalence as between i and Y, though hardly between O and OY. Zen—
odotus’ reading was surely <’§t|J' dfiévrsrg (or a corruption of this).
226 II. Notes on Individual Passages
55 Wilamowitz and Plaumann reported the reading of p60 as errsoxoteg (SPAW 1912. 1216).
ertso)(- is certainly written, but from the microfilm the ending appears to me to be -org.
228 II. Notes on Individual Passages
5° i<at<ot7t>s1B6pevov Bekker, cf. Hes. Th. 786; perhaps better <to> Korretfiopevov (van
Leeuwen), cf. O 30, e 185.
5’ Leaf says that its genuineness had been questioned, but I do not know by whom.
58 It will be recalled that I made a similar suggestion on E 5.
3?;
_.
1::
.1
I2 II. Notes on Individual Passages 229
>-.
,;_H-.r M_-4.,r_. ;
ii
remove the jingle |51Trfig éfiezpirrnl’. This is fanciful; such motives play no
E
part in ancient or medieval emendation. His ‘parallels’ for olnfig are no
.¥
-f
l
l
\ parallels, because in them the accompanying genitives specify (as we
i
should expect) the physical thing that has an impact or exerts a pressure,
.>
2’
the boulder and the fire. So of the wind in O 171. Zeus does not barge the
as
‘ll
oak in person but strikes it with his thunderbolt. For Aloe Trknyfi cf.
O 117; Hes. T/9. 855, 857; Aesch. Ag. 367; Soph. Aj. 137.
i
E 427. Zenodotus’ T’ is superior to 5’ because it is precisely the right par-
E
ticle for the additive function required here: ‘and in addition’.
1
j<<‘ '1
'.<
E 474. 1<s<1>0t7u‘|‘v, although attested only by an isolated papyrus, is well
paralleled by Homeric usage (cf. or 208 aivtiic; vdtp I<€(i)(){?\T‘|V TE; mi iipparoc
Koc7\(3t §ou<0tg | Keivwt), and has a special point here, where Archelochos’
la
II:
head has been severed from his body and fallen to the ground before his
15
1;,
body has had time to collapse: it would be in keeping with the nature of
1%‘:
lg. Ajax’s mocking speech to refer specifically to this head. Aristophanes’ [301
<1>ur‘|v would be acceptable (cf. B 57f. pdthora Biz Néoropt aim | éiaég T8
<‘-I
§._.
pt-‘zvefiog T8 dpvfiv 1" dvxlorot éuiucz-:1); the vulgate ysvefiv, on the other hand,
.
is barely intelligible. ‘Family’ cannot stand for ‘family resemblance’, as
it>
.,._
Willcock and ]anko suggest.
§,.
O 33. Ameis—Hentze note laconically, ‘iiv Akk. des Inhalts; sonst
].1lYfiV0tl ¢17\6Tr|T1 I<or’1 stivfit.’ Leaf’s ‘cognate acc.’ is another Way of say»
ls T. .j ing the same thing. ]anko acquiesces, while noting that ‘the unique con-—
i
ii
5%;
struction seems clumsy (a sign of embarrassment?)’.59 It would imply that
£1.
rt. one could say ptvfivon suvfiv or <|>17\6rnrcx. There is no theoretical objec-
.' $1-_
tion to this. But since epic does regularly say tuyfivou <|>r7\6rnr1 Kori euvfir,
IE;
ii’-
:-
W6 should surely expect the poet to connect q>n6m<; Te Kori e1‘>v1‘|‘ with
éuiyng by means of a dative relative pronoun. We should also expect a poi,
for Without it Zeus speaks as if Hera had been having intercourse with no
one in particular, while deceiving him. I suspect, therefore, that the origi~
-:
..;
nal text was fit ti’ euiyng.
O 43. ou’>Tov is paralleled at K 389 and by the (ancient?) variant 0u5To1’Jg
I lé
-r at Z 439. In the latter place, however, the vulgate is 0n’>T03v, and the geni~—
J.
iii
ar
2-l 5" He seems to mean embarrassment over the explicit reference to the sexual act. He thinks
"I
that this was Zenodotus’ and Aristophanes’ motive for omitting the line. I fail to see why
either they or the poet should be embarrassed about what has already been fully narrated and
§
is here referred to in the usual decorous phraseology of the epic. If he means that Zeus was
embarrassed, I do not see Why the poet should choose an awkward construction as the means
of indicating it.
i
230 II. Notes on Individual Passages
i_d
1
Lines 64-71 were composed, I suggest, as the conclusion of a A100
1
0i7r0i1'r| recited as a self-contained lay. The rhapsode responsible rounded
1 off his tale with a brief prospectus of the principal military events of the
1
1
rest of the Iliad, and a more distant sight (changing into the optative) of
i
the [lion Persis. From his manuscript of the Atog duroirn someone copied
the extra lines into a manuscript of the whole Iliad.
O 88. If we could feel confident that cbépouoot was a real ancient vari-
ant, there would be some temptation to adopt it, and to explain 9600001 as
a false reminiscence of Z 394. Themis is pleased to see Hera, but is she so
pleased, or so eager to be ahead of the others, that she runs to her with the
cup of wine? Andomache naturally runs to meet Hector at Z 394, and in
the other places where the formula dvtiog ii>\6e Béwv is used (Z 54, O 584,
i
E P 257) there is real urgency.
1 However, rbépouoot is presented in sch“ simply as a ‘better’ reading,
without any clear indication that it had any manuscript authority. Béouoot
is objected to as introducing a ‘disorderly’ element (ciu<0tT0‘t0'r0z1'0v), espe-
cially inappropriate to Themis, the embodiment of good order. Greeks in
the Classical period considered it undignified to run in the street; cf. Soph.
El. 872; Alexis fr. 265 K.-A.; Plaut. Poen. 522. If the poet wrote Oéouoot,
it was perhaps to explain why Themis was the first to reach Hera.
O 179. See above on K 451.
O 240. The aorist éootyeiptxro, favoured by ]anko, does not go well with
the present participle ytv05o|<t0v. Hector ‘had just begun to collect his
.-_1
wits’; he was still in the process of collecting them, and recognizing his
comrades who stood anxiously around. Cf. <I> 417.
O 263-8. The lines are repeated from Z 506-11. ‘This simile, so fine
when applied to the vain and handsome Paris, loses much of its force here,
232 II. Notes on Individual Passages
where it is inserted to illustrate not the exultant beauty but merely the
speed of Hector’ (Leaf). This was already observed by Aristarchus, who
athetized 265-8, seeing that 263-4 were required by the apodosis 269f.
Diintzer (1872), 501f., went a step further and deleted the whole simile
with the apodosis (263--70); 270, however, answers to Apollo’s instruction
in 258. There is no reason why the poet should not have repeated at least
a part of the simile in Z. But the truism that ‘repetition inheres in the oral
style’ (]anko ad loc.) should not be used as a defence of the whole passage,
since it is equally true that repetition from other contexts is typical of
interpolations. A poet who mechanically repeated all six lines of the Z
simile must have been one for whom that simile had lost the flower of its
meaning. ]anko’s statement (after schT 263—4a) that Hector ‘may well exult,
since he is aided by two gods’ does not adequately meet Aristarchus’ cri-
ticisms.
The scholium on 265 in T, Znvoooroc roiirov pévov vpo’t<|>e1, ’Api0rotp-
Xog at Kori tong tiihkoug Y‘, means that 265-8 all stood in Aristarchus’ text,
although he athetized them, whereas Zenodotus 0155?. iéivporcbev 266--8 but
did have 265.63 What is the significance of Zenodotus’ omission? We have
seen that abbreviation is characteristic of his text and not necessarily a
pointer to a genuine early version of the poem. On the other hand, we
must eschew the customary assumption that any peculiarity of his text is
the result of a critical decision on his part. As the lines absent from his
exemplar are precisely the ones which spoil the re—used simile, we must
acknowledge the possibility that Zenodotus here preserved (rather than
restored) the truth. Aristarchus rejected 265 as well, whether because he
wanted to get rid of as much of the simile as possible, or because he saw
the horse’s love of bathing as a feature relevant only to the dandyish Paris,
like the luxuriant )(0£i'r0t1.°"‘ In the context, however, the line serves to
explain what drives the unleashed horse to gallop out over the plain: he
rushes to do what he loves doing daily. S0 Hector rushes to the fray. On
this interpretation, the line will be more appropriate to him than it is to
Paris in the original.“
i_
"3 So Von der Miihil (1952), 230 n. 11; Nickau (1977), 118f. Bolling’s emendation of ’Api-
orupxog to ’Ap1oroq>0ivr|g ([1925], 49, 157) is possible but does not seem necessary.
6‘ Cf. Semonides’ woman whom 'i1rrr0<; dfipfi Xoutéeoo’ éyeivottoz she 71.oi3rou two or three
times a day, otiei at Xaitnv éxreviopévnv ¢0pe'i | Baflsiav (7. 57-66).
"5 I have no compunction in speaking of Z as ‘the original’. ]anko’s assertion that ‘neither
context is “original”’ no doubt reflects the belief that it is some sort of oral poets’ common~
place; as if we could never hope to identify a simile created for the poem in which we find it.
t
1
O 290-3. As Hector is the object of the verbs in 288-9, line 291 reads
like a gratuitous addition. Leaf thought its omission would give 292 (.iJ§ Kori
i
i
viiv §00e0(-)0t1 (’)'i0u0u a more natural reference, namely ‘as I deem some
“€‘~f\- l~\
god will again protect and save him, if the need comes’, instead of ‘so it
will happen again that he will lay low many a Danaan’. There is little in
this argument, but the obtrusive awkwardness of 291 remains. 292-3 are
.¢;. . also somewhat awkward, with their specific reference to Zeus so soon after
Tic; Getiiv in 290. I suspect that all three lines are interpolated. I find that
,. .
¢..
if
ii
§..
Pick bracketed them, while leaving them in his Aeolic ‘Erweiterung der
.3-.
i: Menis’ rather than removing them to his ‘Ionische Redaktion’. He regards
i
E
the whole passage 281-305 as a ‘jiingerer Einschub’ ([1886], 230), and pre-
.§-
Ii
sumably means to indicate by his brackets that 291-3 are a secondary in-
‘Z
sertion within this Einsc/anb. On the metrically suspect Kori eooiwoev in
1
1
290 cf. below on Q 570.
O 486. The omission of this line in p1339 raises a query about its auth-
enticity. It is supported by the parallel passages Z 111, ® 173, A 286,
P 184, and dvépeg iiote ¢>i7\01 is preceded by a vocative line elsewhere
(O 733, H 269). On the other hand we have tii ¢>i7t01, dtvépeg Eore without
Ii:
minent figure in the city (F 147, cf. Y 238). In N 176 there is a reason for
Imbrios to lodge with Priam, because he was his son-in-law.
O 607. With the augmentless form vivero, the only caesura is between
Trepi and oT6u0r. This is possible (cf. A 53; West [I982], 36), but a more
natural break is given by Bekker’s i3yi(v)ve1'o, which is also attested in p48.
O 610-14. These lines were absent from Zenodotus’ text and athetized
by Aristarchus (perhaps already by Aristophanes). Aristarchus’ arguments,
as reported by Aristonicus, may be paraphrased as follows: (1) "E1<1'0p0g
in 610 is unwanted, as Hector has been the subject of the last five lines. (2)
The insertion destroys the wonderful momentum of the narrative, which
is restored if we read straight on from 609 to 615. (3) We have already been
told that Zeus was spurring Hector on, and what his motive was
(596-604); why return to the subject? (4) This passage contradicts the ear-
lier one, where it was explained that Zeus was granting Hector temporary
success in order to fulfil his promise to Thetis; to say now that he was
doing it because Hector had only a short time to live is to introduce a new
and untimely idea. (5) Zeus was not operating ‘from the sky’ (610 dot’
01106000) but from Ida.
These arguments (apart from the last, perhaps) have some force. Cer-
tainly "E1<r0p0c in 610 has all the appearance of an addition to the original
text, and that must apply to the whole passage 610-14. ]anko writes that
‘The foreshadowing of Hektor’s death is typical and effective A warri-
or’s short life is a standard pathetic motif (1. 352, 1. 505, 4. 478 = 17. 302,
21. 84f.).’ Yes, but it is inappropriate to the present context. The references
to Hector’s approaching death only begin in the context of his killing of
Patroclus and appropriation of Achilles’ armour (H 799f.; P 198ff.; X 131-3;
O 68 is interpolated), because that is what condemned him to death.
There exists the theoretical possibility that 610-14 are an ill-considered
insertion by the original poet. But the absence of the lines from Zenodo-
tus’ text suggests that they were not in all manuscripts from the beginning.
O 626. Critics have not known what to make of Zenodotus’ apparent
preference for a nominative tiixvn: ‘quod quomodo intellexerit, vix dicas’
(Diintzer [1848], 90); ‘very weak’ (Leaf); ‘nonsensical’ (Ianko). Indeed.
And the inference is that he cannot have intended to read the nominative.
The AXNH in his text represented dixvm. Casual omission of the iota would
be surprising in a text (especially a literary manuscript) of Zenodotus’
time.“ But where a long diphthong is in correption before another vowel,
»i
1"‘ For n1/n in Attic texts see Threatte (1980), 355, 360: n for n1 is common only after 200
BC, though isolated instances occur in the fourth century.
i
i
i-: omission of the iota is common in Ionian inscriptions from the sixth
i. '
1- E century on; cf. Thumb-Scherer (1959), 254. If the iota had not been sup-
29 .:
pressed, it is hard to see how the eta could have suffered shortening. Zeno-
5 .
dotus’ 0t)(vr| uTte1<p1:1c|)0n is on all fours with CEG 167. 4 (Chios, 0.400) owrn
i.
erreornoev. This adds colour to the hypothesis that his was an Ionian text.
:
II12,13,17.Herodian accented (12) fie ...; fj ...; (13) fie ...; (17) fie ...;
§
3'
1
taking the four questions as independent, not a set of alternatives, which
‘Z
would require (according to his own doctrines) fié ij rjé fie ...."7
So far as the sense goes, one may be inclined to approve his decision, as
1
the possibilities that Achilles suggests are hardly exhaustive. Herodian did
1
E not realize, however, what C. Priitorius established,“ that the pure asse-
verative and interrogative particle ii does not have a by-form fie. The disyl-
labic form belongs to if ‘or, than’, and appears only in such questions as
3
v:.
are disjunctive; see Monro (1891), §340. We must accordingly write fie in
12 and 13, while fie is correct for the last member of the disjunction in 17.
111
1»,
.>.
>.
1-.
1: -
1;‘.
1
It is not necessary to take all four of the suggestions as equal branches of
1;.
g.
4.
1-.
the question; the main disjunction is between 12/13 and 17, ‘Are you cry-
1 ing because you have some news to tell us, or because you are upset at the
way the battle is going?’ So not utmm an an an ..., but rather
1
z/strum vel "uel an
1
H 31. Editors generally punctuate with a comma at the end of 30, attach-
1
ing the vocative to the preceding sentence and relating it specifically to the
.=,.-
if
.1’ pronoun 015. I have preferred to place a stronger stop at the end of 30 and
..
:1;-3 1 -
113:.
to make otivotpérn a new expostulation. Cf. 33, 204; A 552; 1 351, 494,
'1 . .
.;I'<-III.
E‘ . 21 474. The (probably ancient) variant otivotpérng (nominative) might
1
be defended on the analogy of A 231, 0np0[36p0c 150co171e1$<;. érrei
E 01’>'r100<v0'io1v oivciooeig.
1
1
i
II 134. The variant 1<0t1<t0v |3e7\t—‘:uJv dacwpijv recorded in sch?‘ looks per-
fectly plausible (cf. O 533), and may be thought to be parodied at Ar. Vesp.
615, as there is no closer model in the Iliad or Odyssey. The reading of the
main tradition, however, 7r000$1<e0g Ai0t|<i00t0, may be supported by the
311
parallel arming scene at I‘ 330ff. There too the 0t6pnE that the hero puts
on is identified as belonging to another:
1;-
Iii' .1
*3, 0s15rep0v (xii 0u5pn1<01 Trap) 0rfi0eoo1v i-§0uvev
i
-.-1E11 oio xaoivvfiroto Aulcoiovog, ijpuooe 0’ ut’>ru31.
ii
21$
-1;
1:
:11. . .
.-g
1
1+
E
141 "7 Sch. 12-13b rroivte Q are P ieorroiofi octv oi o:5v6eo 01 l<0('{'0( 7&6 ov rtfav é wt cxrucriw.
-11
_--;.'.'.1.'
.1 ‘*8 Der homerische Gelmmch eon rj (ric) in Fmgeszitzen (Gymn.-Progr. Cassel, 1873).
.3 1
1
:"1
I:
'1
.3,
1
2
236 II. Notes on Individual Passages
The influence of that pattern might naturally lead to (and account for) the
reference to Achilles as the owner of this one item in the whole panoply.
H 156. Editors are so prejudiced in favour of Aristarchus and/or the vul-
gate that they hardly ever give a moment’s consideration to any variant for
which Zenodotus is the sole source. It is true that most of his peculiar
readings are inferior. But his exemplar was after all, except in a very few
places, the oldest of which we have any knowledge, and it would be
perverse to take it as a working axiom that he could never offer a better
variant than the rest of the tradition. His 'JT()iVTl‘|t here does seem superior
to Troivrotg in view of the parallels; cf. A 384, E 495, Z 81, K 167, and H
496. In this last passage the vulgate has rroivrm, a first-century papyrus
gives rrlcxvlrorc, and Troivtotcg is also reported from one medieval manuscript.
H 158-65. Hentze pointed out that the simile was intended to illustrate
the Myrmidons’ courage and eagerness for the fight, and that 160-3, where
the description untypically moves on to a second scene and the purpose is
lost sight of, might be an interpolation. Leaf adds that the lines contain
several strange expressions; ‘and one cannot but feel a reluctant suspicion
that the directness of the Epic style would be better preserved by the exci-
sion of 158-64 altogether. We thus get rid of the fivfitopeg 1100 uéfiovteg
(164), so that it is the whole body (rroivrotg 156) which is compared to the
herd of wolves, as it should be.’ So also, but more decidedly, Wilamowitz
(1916), 125. The excision re-unites 0‘i 06 in 156 with its verb f)u50v'r0 in 166.
I have bracketed 165 in addition. Patroclus’ preparations have been
described separately (130-54), and he then, apart from 165, disappears
until 219. The Myrmidons are pictured rallying about their usual leader,
Achilles (155, 166, 168, 198ff.). Patroclus’ appearance beside Achilles in
165-6 is awkward. If 165 is removed together with 158-64, we get a per-
fect seamless join:
0'i 0%: 7\6i<01 dig
ocboivoi, roioiv re rrepi <1>pe0‘1v éiorreroc ci7u<fi,
0vr" év 0’ 01001 roiow dprj'1'0g‘i01'0tr’ ’A)(1Ms15<;,
C)"-'13"8 -ptivwv ‘irnroug re Kori dvépotc 0’t0m01u3r0<g.
8‘:
""i
H 180. The words X00031 1<0t7\.r] have since antiquity been construed
together, ‘beautiful in the dance’. I suspect that Xoptiit goes rather with
Etude, as a shorthand expression meaning ‘gave birth to (as a result of
being seen) in the dance’, as explained more fully in the following lines.
Cf. 150-1, on the horses Xanthos and Balios ‘whom Podarge bore to
Zephyrus as she grazed in a meadow beside the stream of Ocean’, mean-
ing that Zephyrus saw her and made love to her there (cf. Y 224a = Hes.
Th. 279).
II. Notes on Individual Passages 237
II 352--7. This simile may originally have been composed to follow 296,
where the reinforcement of the Myrmidon force by the mass of the
Danaans is narrated with surprisingly little emphasis. The further simile in
297ff. would follow well after 357. The passage may have been disturbed
by the decision to introduce a section about individual feats of the leaders
(306-51: Patroclus, Menelaus, Meges, the sons of Nestor, the LocrianAjax,
Peneleos, Meriones and Idomeneus). Telamonian Ajax is reserved for
358-63, where he serves to keep Hector occupied and out of Patroclus’
way; but by the same token 358--63 would stand better immediately after
351, to explain the obvious omission from the series of fiyeuoveg. The pro»
cess of incorporating 306-51 + 358-63 may have resulted in the detach»-
ment of 352-7 from 296 and its attachment to 351.
H 382. err) 6’ "E1<'rop1 Kéidteto Buuoc, ‘his heart called out an order at
Hector’, is unintelligible. The reading of p60, 1<c1<7\u1'0, is equally so, and
probably has no more significance than countless other wild errors in this
source; two better papyri agree with the vulgate. It may nevertheless be
that the second e of Kéxitcto is faulty, and that the poet wrote 1<é|<M'ro, ‘his
heart was set towards Hector’. I cannot produce an exact parallel for the
expression, but it seems more credible than the alternatives.
II 428. dvrzuhoxeihai is the form given by the older and better
manuscripts in all places where the word occurs: here, X 302, [Hes.] Scat.
405, Barr. 294, Ar. Eq. 197 and 204. V/ackernagel (1955), I 628, argued for
-Xfiitou, so that the reference is to the eagles’ talons, not their beaks. This
seems tautologous beside votuwtévuxot, but it is certainly favoured by the
application of the epithet to crabs in the Batmchomyomac/via, and prima
facie by the Aristophanic interpretation of the oracular word: iiti
(XY1<D7\(I1Q Tutu; Xcpotv 0tpTr0tCwv tbepet. However, this is perhaps not con-
clusive given the loose way in which the rest of the ‘oracle’ is interpreted
(for instance, the iipoixtov signifies a sausage, because both are long).
Against -xsikott from ;(e"17\og, Wackernagel argues that )(ei7\0c; is very rare
in the sense of a bird’s beak. It does occur, however, at Eur. Ion 1199,
Call. fr. 194. 82, Epitap/oium Biorzis 15, Mnasalcas epigr. 15. 4 Gow~Page
(/1.1’. 9. 333), and Opp. Hal. 3. 247.
The formation of the compound is certainly anomalous if it is based on
)(z-;‘i?\0<;. To explain it one may conjecture that the original formula was sin»-
gular, Wauwtévuxoc dw<u7\o;(e17u"’|c, with a correctly-formed compound
from the stem Xstkco-, and that when a poet wanted to turn it into the
plural, since ~)(s17\éec_; would not fit the verse, he resorted to the device of
treating it as if it were an a-stem in -ng and making the plural in -ou. There
was a minor problem with the accent, as -Xethai would seem much less
II. Notes on Individual Passages 239
natural than —)(t-;'i7\cr.t (as if from ’*‘-)(ei?tr|g). Perhaps the rhapsodes simply
shifted the accent in accord with this feeling. Or perhaps dv1<u7\o)(e1?\r|g
inherited a recessive accent from the Aeolic period of the epic.
H 507. The Myrmidons caught hold of Sarpedon’s snorting horses,
ieuévouc c|>ofJéc060t1, ilm-3’: Jtiftov dpuozt’ (itvo'u<Twv (so Zenodotus’ text, also
/0 C R). The second hemistich is a repetition from 371, where it referred
to horses that had broken their chariot-poles in trying to cross the ditch
and left the chariots behind them. Sarpedon’s horses, however, presumab-
ly remain yoked to his chariot and have no way of leaving it behind. Aris-
tarchus’ reading Xitrrev, which is also that of the vulgate, is quite obscure;
he explained it as an aorist passive, = €7te1<1>6ev, but even if such a form
were possible, we should have gained nothing for the sense. His recourse
to such an artificial exegesis, however, suggests that Ptittev was not his own
conjecture but what he regarded as the paradosis.
What frightened the horses? Presumably the fact that there was nobo-
dy in the chariot to control them. The problematic hemistich, with its
reference to a chariot and c’Jivot1<1'z-:c;, must have been intended to express
this somehow. I have cited in my apparatus two old conjectures which
strain in the right direction, though neither can be regarded as the final
solution. If ]ortin’s éivotlctcg were right, the plural would have to be taken
as loosely generalizing, or as referring to Sarpedon and his charioteer
(though none was mentioned when he leapt down from the vehicle at 426).
Bayfield’s dtvoikttop suffers from the drawback that the word does not
otherwise occur in the Homeric poems, though dtvomtoptog does (0 397),
and it is striking that at Hymn. Ap. 234 the horses that run with empty
chariots in the grove of Poseidon at Onchestus are described with the line
Keiv’ Eixcot icpotéouotv dvartopinv dotévteg. I have also pondered the
possibility that dvohctwv is to be retained and that Airrev is the corrupt
word, concealing some other verb that meant ‘rattled empty’ or something
of the sort, but I have failed to think of anything plausible.
H 548. KOtT0l Kpfifiev or Kort’ éixpnfiev. No doubt the original phrase was
Kort’ éinpnflev, ‘down from the summit’, and Kottot |<pfi6ev, understood as
‘down from (or over) the head’, arose from it by misdivision; cf. Leumann
(1950), 56-8, my note on Hes. T/0. 574 (West [1966]), and janko here. But
the latter is clearly what was intended in some epic passages, and I think
it was so here, since the Trojans had heads but not t’i1<p0t1.
II 554. Msvorrtdoew or -(£6010. If we prefer the older form of the geni-
tive, we must assume Attic correption of the first syllable of H0ttp0K7\fiog.
According to Janko, ‘the MSS show that (Homer) chose this vulgar inno-
vation instead’ (of using the —eto genitive). They show nothing of the sort,
240 II. Notes on Individual Passages
"9 Neither does the imitation of the phrase at Nonnus Dion. 2. 39 f. ifitpepov éixpou I fiiéveg.
242 II. Notes on Individual Passages
7° 5fipo<; has this sense, in antithesis to the leaders, at B 188/198, A 328, M 213, 447, P 577.
7' -5:-301 for ii‘-51-zéot as in 1" 364 Beoufaéot.
72 Did the author read poixqt 5m, as written by lo, or was this variant generated by the para-
phrase?
II. Notes on Individual Passages 243
2
Aristarchus apparently read iiooov instead of 50001. He too was unable
it
i to account for the ts and pronounced it to be rreptttocg, K016’ ’Opnpu<fiv
ouvfifietotv (sch. 368d). He construed ilrti with iiooov, as it were ta’ iioov
Tfig poixng ’€o‘r0tootv oi tiiptotot, ‘over as much of the fighting as was occu-
I pied by the stand of the diptotot.’
.5
The phrase iiooot dptotot occurs at line-end in A 211, A 691, M 13, each
time with i-iootv understood, meaning ‘all the diptotot’. I take it to be cor-
rect here: 50001 diptotot I Eototootv o’cu<I>’t Mev0mo15m itottottefivnditt is the
i
subject of fiépt Katéxovto, in antithesis to 370 0’1 5’ 5i7\7\0t Tpu3t=;c; I<T7\. I
i
have left uoixnc; étti 6’ within obeli. If one has to guess at what may be con-
cealed in it, one may most naturally think of a dative noun coupled with
E
fiépt by the 6’, such as ouixhnt 6’; cf. 649 ou’>ti1<cx 5’ fiépa uev o1<é5otocv
;:
Kori 0’tTto3oev éuixhnv. This has at least some letters in common with udxng
$3
-i
i, .
or uoixnt. But it would be a very surprising corruption.
$
P 385. The original root of ’t5pt6c; was "‘swidr6s-, and a dative *’15pt51'
r
i. prior *’t5p6ot with shortened stem vowel, and that in turn implies a dative
*’t5p6'1', contracting to "’15|:)o’i. That the contraction is to be assumed is
shown by 745, where the dative must be scanned as a spondee.
P 695-6. ducbaoinz see West (2001).
In 696 (and \P 397, 'r 472) Bodtepfi 56 oi iioxeto cbtovri is transmitted
unanimously. But at 5 705, where the same formula occurs, some pre-
Didymean texts had §oKt-3T0, which was taken to mean éyéveto. Didymus
ii.
rightly rejected this as a form of eiui. Modern philologists, however,” have
Z53
recognized in €o1<c'ro a reflex of "'se-s/e-eto, an isolated survival of an IE.
i-
1
'€ root sek attested in Sanskrit and Lithuanian and meaning ‘dry up, fail’.
I
IE
This gives more point to 80t7\epfi, which is not applied to the voice except
I-.
i
in this phrase. iioxsto is certainly the dzfficilior lectio, unlikely to have ori-
i.‘
i ginated as a mere error when Eoxeto provided so easy and obvious a sense.
if
2;.
P 727. Payne Knight was right to eject the vcip, which does not suit the
E1 sense or the structure of the simile. uév Te is just right. This makes room
for the spondaic form éiwg in place of a monosyllabic ‘€wc_: which does not
1:
t.-
otherwise occur in the Iliad (though it does in the Odyssey).
I
P.
'i
i‘. .
Z 26-49. This passage looks as if it has been embellished by several inter-
I.
E.
polations. First, 26-7, condemned by Diintzer, who objected: ‘Dafi er sich
i.
I‘.
auf die Erde geworfen, mufi schon 23 ff. angenommen werden.’ This is not
73 Schulze (1933), 368 sq.; Bechtel (1914), 142; Schwyzer (1939), 652 n. 5.
II"
I-.:
§..
E
ii 1. '-
Bi.i
244 II. Notes on Individual Passages
in itself a cogent point. Achilles did not need to lie on the ground in order
to pick up dust; nor could there be any objection to the poet’s saying
explicitly something that he has implied, while adding detail to it.
However, the spoiling of the hair by tearing may be thought superfluous
when it has already been spoiled with dust and ash, and the verb
filoxvve is repeated from 24. It may also be felt that the phrase iceito uévotg
uevakwori, magnificently used of the slain Kebriones at H 776, and pro-
bably (primarily) of the slain Achilles in the Aet/oiopis (cf. to 40), is not so
suitable here and has been appropriated by a lesser poet.
The more telling argument is that 26-7 conflict with 23-5 and with what
follows. Achilles scoops up some dust and ashes and pours them over his
head; his face is dirtied (and of course his hair), and a layer of ash settles
on his vs-:K'r0ipeo<; Xlttiiv. We picture him sitting or standing. The outototi
run out and surround him, wailing and beating their breasts. Then comes
32,. ’Avti>\o)(oc; 6’ étéptofiev 65150510, where érépwfiev is naturally taken to
mean ‘opposite Achilles, facing him’; cf. A 247, X 79. He is holding Achil-
les’ hands: if Achilles were grovelling on the ground, Antilochus would
have to be in an awkward crouching posture, and in any case the detail
does not fit well with what we read in 27, that Achilles is using his hands
to tear his hair. When Thetis arrives (70), she stands by Achilles and holds
his head. Thus the rolling on the ground is out of accord with the rest of
the scene.
Later on, it is true, Achilles is represented as lying on the ground as he
gives vent to his grief: 178, 461, T 4.74 But this seems not to have been the
idea in the poet’s mind when he was composing 22--137.
The next suspect line is 34. It is natural enough that Antilochus should
clasp Achilles’ hands as they lament (33). But 34 introduces an unexpec-
ted, sensationalistic motivation: ‘for he was afraid he might cut his throat’.
It also introduces an awkward switch of subjects to and fro. i5 as in 33
changes the subject from Antilochus to Achilles, and he continues as the
subject in 35, while the subject of oeiote in 34 is Antilochus. We can say,
as Nicanor did, that ii 5’ eoteve Kvodtktuov Kfip in 33 is parenthetic, equi-
valent to ‘while he groaned’. But then we should expect a new pronoun to
signal the change of subject in 35.
Finally there is the catalogue of Nereids in 39-49, athetized by Zen-
odotus and Aristarchus ting 'Ho16i5e1ov i-fxwv )((Xp(X1<Tfip(1. The scholium
explains further that Homer’s practice (unlike Hesiod’s) is to speak of
7’ One of these lines, X 461, is a curiously inorganic one appended at the end of Thetis’ ap~
peal to Hephaestus, and I suspect that it too is interpolated, as Diintzer thought.
ii
,.
E.<.
i.
2.
ii
l
Jr.
5.
i
P.
collectivities of goddesses, such as the Muses or Eileithyiai, without giving
them individual names. This is a true and pertinent observation. Internal
é
z
i
evidence suggests that the names of the Muses were actually invented by
l2
i
Hesiod.” As for the Nereids, one or two such as Thetis and Amphitrite
must have existed as individual figures before Hesiod, but his list of fifty
S
names must surely be largely his own creation. The list in the Iliad has
sI
I,5
3.
ii
many of the same names, and two or three whole lines are virtually the
,.
§
2
(the catalogue of rivers).7° This does not, in itself, speak against the auth-
R
enticity of Z 39-49. The objection to the passage is its departure from the
"i
E
Iliad poet’s normal manner, its disproportionate and irrelevant suspension
5i.
2
i
of the progress of the narrative. A catalogue of troops as a preliminary to
a battle is one thing; but these Nereids are going to do nothing but follow
T
¢.
ii"
along with Thetis as a background chorus. The interpolation is of some-
2..
55-,. what the same kind as I" 144, where Helen’s two anonymous attendants
iE‘ are provided with names.
i
F: The absence of the passage from the Argolic text, reported by Callis-
R
tratus, is not as powerful a confirmation as it might appear, since it might
E
5.
i
i
represent either casual omission by a rhapsode or an accidental semi dz/z
E
if
méme an méme (38 ~ 49).
E
E.
2. E 48. The tradition is strongly for ’Au0i6u10t, and the rhyme with
2
>.. ’Q peifivtot in the same line might be thought to favour this reading. On the
t
I
other hand it might be the source of corruption. ’Qpei6ut0t is a meaning-
\
i ful name (if not very suitable for a Nereid), ‘she who rushes over the
mountain’,77 whereas ’Auci9u10t is obscure, unless it were for "“Auti6v10t,
‘she who rushes alongside’. ’Au0'i8&;t0t, on the other hand, is perspicuous:
‘Sand-sprite’, corresponding to Hesiod’s \P0tu0i6n (T/9. 260), who appears
in Pindar as \P(x|.1ci6c10t.78
i
al
i 75 See West (1966), 32.
7“ See Wilamowitz (1916), 57 n. 1; E. Bethe, Homer. Dic/attmg and Sage II (Leipzig-Berlin
1922), 304; West (1995), 208.
i.
77 The first syllable shows metrical lengthening, for ’Opei6u10t.
78 Nem. 3. 64. The alternation between -slot and -n is common in these names; see my notes
on Hes. Tb. 244, 250, 913-14 (West [1%<a]).
i
i
‘E
E
246 II. Notes on Individual Passages
I
i
Z 142. Zenodotus’ oivopefioou is the superior reading, and would have
been adopted more widely but for the prejudice against this ancient
source. The imperatival infinitive is common in Homer, and especially
when an imperative has preceded," as here; cf. Schwyzer-Debrunner
(1950), 381; Chantraine (1953), 316, ‘l’infinitif sert volontiers a développer
5 une idée qui a d’abord été exprimée a l’impératif’, quoting A 322 f., B 8—10,
F 459, etc. Here it is rhythmically preferable to the elided dtvopei5oat(s),
and it is dzfficilior lectio, as the trend would be to assimilate the form to
the preceding 5131's.
Z 192. The paradosis is éE7\7\ov 6’ 013 rev oioa, ref) éiv zdxutét 're15;(ea 515w,
with variants ofaoé rev and oi5 Bnv in the first half and T013 éiv in the
second. The forms Ten, Tsii cannot be original (for T80 and "réo or Ti-5’), and
the syntax is disconcerting. It looks as if it is trying to say ‘I do not know
anyone else whose armour I could wear’, with the ‘anyone else’ attracted
into the genitive of the following ‘whose’. But while the attraction of case
xi\=~. ;-\.c>-~.w'aA.4- ~\. is tolerable, such a sentence surely requires a relative pronoun as the
connective, not an interrogative. The weakly attested T013 would give us a
relative, but the weight of the evidence is for T613, making it an indirect
question. If it were simply 01314 o’i5ot, T1VO€_I (oiv) 1's15)(e0t otito, all would be
straightforward (though the dv does not seem very much at home in a
deliberative question). The Tivog might be amplified with éi}\>\ot>: our oioa.
1;
Tivoc, (Eiv) éi>.7\ov reiixeot 61503, or perhaps with inversion, t’iN\ou 6’ ot’>r<
oioot Tivoc (div) Teifixeot 51500. The syntactic confusion arises when to c’i7\7\0u
is appended an indefinite pronoun: tii?\?\oL> 56 Ttvocg o1’>1< oioot rivoc (oiv)
il
.2 Teéxeot 6130.).
The variant 611v for T81) might seem to offer an escape from the diffi-
culty. But apart from the weakness of its attestation, the particle is not
appropriate. Bnv is often used after oii, but whether in negative or positive
propositions, it always carries the notion ‘I imagine’ or ‘I venture to
.3 assert’. ‘I imagine I do not know’ is nonsense. This reading presumably
originated as a Byzantine conjecture.
The division oiit’ 613 (Z W) might lead one to think of o1’>1< sf). 613 oioot
i
:5 is a common enough phrase, only we never find 013K 613 oioot, ‘I do not
ii know well’. Again the sense is unsatisfactory.
i
We are left with oii rev, or rather (Payne Knight, Pick) oii Teo. There is
l
evidently a contamination of two constructions, ‘I do not know anyone
else whose armour I could wear’ (with a relative connective) and ‘I do not
ii
.__, know who else’s armour I could wear’ (indirect question). If we accept this
:}j§
5.‘.
-ii->
explanation, it only remains to tidy up the linguistic form of the subordi-
.1
nate clause. Retaining the modal particle, we will have to assume either a
..E-Ei
i
‘E
248 II. Notes on Individual Passages
‘r
|
.|
1;
.i
either to confirm its presence here or to account for a postulated corrup-
tion. The variant v‘i1, though not attested before the twelfth century, may
offer an escape; it is found corrupted to viei at B 791 and perhaps (D 34,
X 302. As there are examples of dative -1 being elided in Homer, one pos-
sibility is 1>‘1’ £11651, as given by some later manuscripts, which would prob-
ably have been written anciently with scriptio plena for clarity, vii cuter.
1
No less attractive, however, is Nauck’s vii 1101, with the dative pronoun
serving as the equivalent of a possessive adjective; cf. A 505 Tiunodv 1101
viév, [5 50 untépt not nvnotfipeg évréxpotov our é6e7\o15on1, 6 771 E5 0'1
(bow/og vii T€Tu1<T011. Cf. Headlam on Herondas 1. 61; Wackernagel (1924),
77f.
Z 461. See above on 26-49 (p. 244 n. 74).
Z 505. fiep0<1>u5vwv is strongly supported as the only ancient reading; but
it is unintelligible. The ancient explanations, ‘reaching the limits of the air
-s
with their voices’, ‘vocalizing by (striking) the air’, ‘making proclamation
<v15
32
Z in the early morning’, are all ridiculous, and no better ones have been
devised in modern times. The weakly attested variant iepoclxévtov, found in
Photius and the Sada, is very attractive, not because of the sacrosanctity
fij
‘Z
of the herald’s calling, but because in some of its Homeric uses the Greek
:?
1' i
‘teptig appears to have the sense ‘strong, vigorous, full of impetus’, corre-
i sponding to Vedic isird-.83 In two passages of the Rgveda (9. 84. 4; 10.
I ii
ii"
.'€
ii 98. 3), as Schulze noted,“ this word is applied to the voice. ‘1ep6<1)wvog,
therefore, might have been an ancient epithet of the herald. It is not
3
‘ 1
12
=r
32 A 268 6Es'i(0u) éoiivou; P 89 cio[3éoT<.o1' 0136’. (See above on these verses.) The special case
of ’Evuot?\i@t_tj,v6pe1q>6vTn1, which derives historically from a formula with regular prosody,
would only be relevant here if poets had taken it as licensing them to indulge in other horren-
dous crases — a dangerous assumption. The particles nfi. 51']. I]. izrrei are found in crasis
with some frequency. Otherwise I can cite nothing from the Iliad more alarming than A 277
l'l|'|7\ei‘6_[1_éBe?\’, where the vowel—sequence, ne, is one that commonly contracts. The Odyssey
provides a couple more: 01 226 ei7\ot1rivg__f1e voipog, p 375 giiipivvmte oo[3o3T01. ea 247 0:314
Eiyxvg 015 Trpocoin. For a full survey see La Roche (1869), 281-3.
83 Schulze (1892), 211f.; Durante (1976), 94 f.
. 55
._,
250 II. Notes on Individual Passages
*5 Her lines 578-»81 Usher = E 603»-6 with the extra clause, and 582-5 = 2 593-6. But 577 =
6 15, so it is clear that she had the parallel Iliad and Odyssey passages both in mind. The Odys—
sey, therefore, may have been her only source for the extra clause.
Z
i
<.
This was also the text known to Athenaeus (181a-d), who, however, claims
i
that Aristarchus had removed, after Teprrouevot, the words uetiit 56 o‘q>1v
éuémreto 6e'io<; dtotooc I rbopuiluov, which occur in the similar passage
5 15--19: l
I?
i 82 g 0'1 uizv Botivvvto Kort)’ 1'>wepe¢ec uévot otfnuor
yeiroveg first Eton Mevekoiov rvoaainoro
‘i
'rep1r6uev01' permit 66 o<i>1v énémeto Beioc oioroéc
-'.z€
J
oopuiltov, oottb 6%. Kufitotntfipe Kort’ 0n3to15c_:
=i
.$
3
II
uokrrfic éfioipxovtsc éoivevov Karo: péoooug.
Wolf ‘restored’ the extra clause to the text of the Iliad, and his line-num-
bering has remained standard, giving the insertion an air of authority
which it does not merit.
'_>1
.2‘
Athenaeus does not say that anyone claimed to have found manuscript
i
<1
support for the longer text. The only basis for the story, evidently, was
.1.
1:
i
comparison of the Iliad with the Odyssey passage. When we examine the
context in Athenaeus, we see that it is part of an argument that the lines
in the Odyssey are interpolated from the Iliad.
The argument runs as follows (180b—1d):
Homer assigned delight in singing and dancing to Penelope’s suitors and to the Phae—
acians, but not to men such as Nestor or Menelaus. 0'1 Trepi ’Ap'ioto:p)<ov, however,
.{ Z
1‘.
not understanding that Megapenthes’ wedding was over and that Menelaus and Helen
were concluding the celebrations on their own, added the lines 5 15-19, transferring
L
ii}1| ;
them from the 'O7rP\oTroti0t together with the false reading éfioipxovtegz of) votp
-11:
éfidpxovteg oi icufitotntfipsg, 6000 éioipxovrorg r06 031606 rrcivtwc <bp)(ot'5vto. (The
IE
=1}; implication, taken up by Wolf, is that the transmitted éfiripxovteg is to be emended
1;2::
to éfioipxovtog.) The dancing and tumbling are entirely appropriate to the Cretan
iE chorus in E (passages are quoted in support of this contention). But not content with
interpolating the lines into Menelaus’ symposium, where they do not belong (being
:1
if
1
1
alien to Laconian Trouosiot and to the king’s owdnpooévn), Aristarchus removed the
i
!
singer from the Cretan chorus, thus depriving us of the means to correct the corrupt
.4 ilfiripxovteg, as reference to the singer is no longer possible. That the Odyssey pass—
1l"
11
ii age is an interpolation is further evident from the fact that in what follows, Tele-
J.
2
;§ machus and Pisistratus display no awareness of any musical entertainment, there is
just feasting and talk. 4
This vigorous reasoning is presumably not Athenaeus’ own but taken over
I from a post-Aristarchean Homeric scholar. I believe that M. Miillergi’ was
right to identify this scholar as Seleucus, from whose writings Athenaeus
-Z
i
often quotes, and who is cited a few pages later (188f) for criticism of an
Aristarchean reading in the same episode of the Odyssey (6 74). The argu-
i i.
lé
'.'i::
bate.87 If this were the original text, it would be difficult to understand the
logic of 79-82.
Aristarchus understood Agamemnon to remain seated because he was
suffering from the pain of his wound (Aristonicus at 77; Porphyry p. 110.
8 Sodano). In the Massaliotica and Chia, although he no longer sat, a refer-
ence to his pain was supplied, as if to compensate: ufivtv oivototevcixwv Kori
13¢’ ‘é7u<e0g dhyeot rroioxwv.
T 107. The tradition has followed Aristarchus and Herodian in accent-
ing tpeuotfioelg as a verb rather than tpeuotng eig ‘you are a liar’.88 But a
verb wevoréw or -din) is otherwise unknown, and as Wackernagel pointed
out,89 we do not want a future but a present, as Zeus’ false assertion has
already been made. It is artificial to render it ‘du wirst zum Liigner wer-
den, als Liigner erscheinen’ (Faesi), ‘thou shalt prove a liar’ (Leaf). Prob-
ably the accompanying éittfifioelc; was responsible for the reading of the
letter-sequence -HCEIC as another future ending. Herodian calls the divi-
sion weiiotng eic; “diT01rov”, but it is hard to see why. For the use of the
agent noun in relation to a single utterance cf. Soph. Ant. 1194 f., Ti voip oe
uodtfioioooru’ (iv div éc iiorepov | weiiotou q>avo15|n-:6’; and also Eur. Or.
1608 f., (Menelaus) dsroupe Buvottpoc, ¢-oiovavov. (Or.) weuofig edpvg. (Men.)
ciaaét Ktevéig pov Bvyorrép’; (Or.) 013 ipeuofig ET’ ei.
T 223. According to the bT scholia, dtuntog is accented proparoxytone
when it means ‘harvest-time’, oxytone when it means the harvested grain.
This doctrine is found in various other grammatical sources, and appears
to have been purveyed by I-Ierodian.9° Other sources, however, distinguish
the accents in exactly the opposite way (schD here; [Ammon.] Diff 38;
Tzetzes on Hes. Op. 384). If we accept the first version, we should write
dtpntog 5’ olivtotog, as the sense is clearly ‘the gain’, not (as the scholiast
takes it) ‘the time taken to fell many warriors’. ,
T 242. be may be defended on the ground that the expression is short
for éiuot uev uiifiog iinv. éiuoi at €pvov tetéheoto.
T 365-8. Didymus records an athetesis of these four lines, ve7\o'iov yup
To fipvxciofial Tov ’A)(17t7\é0t. He adds that according to Dionysius of Sidon
Aristarchus originally athetized them, but later changed his mind and
decided that ‘this sort of thing’ was poetic. The lines certainly look like an
insertion, for 369 ff. are the natural continuation of 364. But can the inser-
tion be ascribed to the original poet?
My feeling is that Aristarchus’ initial reaction was right. The close ver-
bal parallel with the ps.-Hesiodic Scutum, 164 tdiv Kori oodvttov pizv
icon/0t)(i1 Tré?tev, puts the author in bad company, and it may be suspected
that he was a rhapsode of the same period (early sixth century) who saw
fit to add what he thought was a dramatic touch to the account of Achil-
les’ arming. Having diverted from xoptiooeto to the hero’s gnashing teeth,
flashing eyes, and impassioned heart, he has to steer himself back to the
armour (368) in order to be in position for the following lines.
T 384. The neglect of the digamma in '60 together with the hiatus indi-
cates that something is amiss with the vulgate’s iio 01131013. With scriptio
plemz, the text becomes be so avtov, which is better interpreted as st-; ii’
0u3to13 (Heyne). A longer tonic form of the pronoun, eéo (< "‘sewe'syo), is
equally possible (van Leeuwen; Schwyzer [1939], 609; cf. Leaf on N 495);
it will be parallel to ts (T 171, Q 134, and probably E 162, P 551) and éoi
or izoi (N 495), both used before cases of (XOTOQ,.
T 30a--c. The three lines quoted in schT are a gloss on 30 {mi-3p utipov.
The scholiast says that they were read by some in place of line 30, but this
cannot be right: they must either have been substituted for 28--30 or (more
likely) appended after 30.
Y 85. See above on K 451.
Y 125-31. The poet has brought the gods to the battlefield, all eager to
support their own side. Apollo has emboldened Aeneas to fight Achilles.
The poet cannot now have the pro-Achaean deities take no notice of the
impending encounter. But he wants to describe it as a straightforward trial
of arms, without the complication of divine interference, up to the point
where it becomes necessary for Aeneas to be rescued. So he resorts to the
clumsy manoeuvre narrated in 112-55 (a section which Payne Knight and
Diintzer wanted to eliminate altogether). Hera sees the two heroes approa-
ching one another and proposes to the other main pro-Achaean deities,
Poseidon and Athene, that they should take action. (Hermes and Hephaes-
tus, assigned to this party in 34-7, are here ignored.) Poseidon responds
with the counter-proposal (for which very weak reasons are given) that
they should on the contrary withdraw and leave the mortals to get on with
it, so long as the pro-Trojan gods take no initiative. They do withdraw,
and the pro-Trojan gods do likewise; 155 is the poet’s bashful acknowl-
edgment that this represents a complete suspension of the theomachy
which Zeus set in motion in 4ff.
II. Notes on Individual Passages 255
basically saying ‘Why so many words?’ There are various possible exci-
sions that would improve the passage; I have cited four in my apparatus:
Christ’s of 244-50, Heyne’s of 246-55, Bekker’s of 248-50, and Diintzer’s
of 250 alone. But the question requires a brief discussion.
The genealogy in 213-41 looks like an addition to the original speech,
hardly anticipated in 203-9. But it may be an addition by the first poet,
after he received the impulse to honour the Aeneiadai as the latter~day
royalty of Troy (cf. 179-83, 302-8); it serves to legitimate the dynasty by
explaining its relationship to the house of Priam. The point of 242-3, in
which dtpetrj means ‘social status’, is that the fortunes of Aeneas’ line may
go up or down in future: royal power (which Achilles has accused Aeneas
of angling for) is not ruled out, but it all depends on the will of Zeus.
There is no further point of substance to be made in the speech, and
Aeneas should now say something like ‘But why make long speeches at
each other? We need to fight.’ Lines 256-8 provide a conclusion of the
kind required. What is wanted between 243 and 256, however, seems to be
provided in duplicate by 244 ff. dc7\7\’ dive |.1m<€;T1 Totfita and 251 ff. d7\7\6t
tin Eptoag These two passages thus seem to be in competition, and
Ludwig Friedl%inder’s opinion that 244--9 and 251-5 represent alternative
recensions (with 250 to be discarded) deserves to be recalled.“ At any rate,
I should favour the deletion of one or other of them as a minimum mea-
sure. Heyne’s slightly longer athetesis of 246-55 would also leave a per-
fectly satisfactory text, and it would rid us of the especially weak lines
248-50 - weak because too general in sense, with no special point in the
present situation.
Y 259. Didymus (in A) reports Aristarchus as having read swat ‘with-
out the e’, and refers to N 407 where a shield is described as olvtotfi. The
manuscript gives otvwt the accent of oetvtfit, and the assumption seems to
be that Aristarchus regarded it as an adjective equivalent to otvmtog. Pos-
sibly, however, he intended the noun oivtoi, taking ooixei to be in apposi-
tion. oivog is used of a round, concave goblet, and the analogy between
shield and drinking-vessel had inspired Timotheos’ (i>10‘Otr| "Apetog.92
Aristarchus’ text also diverged from our tradition in giving o0i1<e'i isihoto’,
not ooixet fi'?t0t0'£-;(v). We know that he had a general preference for ‘Ionic’
unaugmented verbal forms, but it may be doubted whether he indulged
this preference to the extent of altering ooixel fjhaoev into ooixei §7t0to’,
i“1“_
9‘ I should take ‘alternative recension’ in this case to mean that different rhapsodes’ copies
were found to have different versions, which were both incorporated in a subsequent copy.
92 PMG 797; cf. Arist. Poet. 1457b22; Rlaet. 1407a16, 1412b35.
II. Notes on Individual Passages 257
producing a hiatus and what might have seemed to him an irrational leng-
thening ofdative -1. In favour of his reading are the frequency in the Iliad
both of ooiicei scanned U U -93 and of unaugmented §7\otoe.94
T 282. For ocbfi 0t7\uo'io1(v) Bentley conjectured 6<1>60r7\uo'i1v, thinking that
a genitive was called for, as in E 696, ‘P 765, al. But the dative is also cor-
rect, cf. C 235, 0 19, X 463.
T 333. Most editors print the reading of A and other manuscripts,
oivticr Hnaeiwvog imepfiiiuoto uoixeofion. But given the existence of a well
supported variant (already in two papyri), that is likely to be an intrusive
recollection of 88.
<1) 95. Aristonicus records a remarkable Zenodotean variant here. See
West (2001).
<I> 111. The ancient variant 5:-:i7\nr; clearly has no merit; but how did it
originate? I suppose as a consequence of the false punctuation mentioned
by Nicanor, by which €oot-Jtott was attached to the previous line instead of
starting a new sentence. There was then a temptation to subordinate ii’ firing
ii beihn ii uéoov fiuotp, making it ‘death and fate will come upon me too,
either (at) morn or eve or midday’. If so, we should expect or-:i?\ng to be
paired with fiofig, even if metre precluded any alteration of uéoov fiuotp.
In fact we find in the Silaylline Omcles, at 2. 182, Tijfiel vdtp (v.l. voip T’)
fioi3c; ii oeihng ii uéoov finap. But it is strange that no other trace of fioiig
remains in the tradition, when 5ei7tn<; is so widely attested.
The contracted form oeihn occurs only here in Homer. Otherwise we
find oeiehocg (<1) 232) or oeiehov iiuotp. Resolution of the contraction would
rid us of the hiatus and is very plausible. oetéhn (Wackernagel [1916], 166,
following Fick’s Aeolic oetéka) is the lightest change, though the parallel
of line 232, the only other place in Homer where the word occurs as a sub-
stantive, would favour Brandreth’s oeieaog.
(I) 122-3. The triple accusative is highly unusual, but intelligible: it com-
bines the ideas ‘lick you off, lick your wound off, lick the blood off’. The
ancient variant u§Tet7\fig, ‘lick away the blood of(f) your wound’, is a lec-
tio facilior. However, it is also possible that o’ was intended as a dative,
and was a replacement for the more correct but less perspicuous T(ot). Cf.
above on A 170.
93 Before double consonant, O 272, N 130; before 1.1-, A 572; before a single stop, 9 267, (D
241, reflecting Mycenaean "'sakel9ei < "‘t'waleesei.
9‘ Six instances, including two of elided Elmo’; many more of €7t0zooe(v).
258 II. Notes on Individual Passages
<I> 126--35. The passage does not read very coherently, with the strange
asyndeton in 126, the repetition of iX915g in 127 after i)(6t501v in 122, and
in the next line the abrupt switch to a plural address to the Trojans (or is
it the sons of Priam? cf. 105). Where Bptéiokwv stands, we might have
expected Evfiot st. Presumably the participle is rather emphatic: ‘with a leap
through the waves the fish will dart under where the dark surface is
disturbed, to feast on Lycaon’s fat’.
Aristophanes’ dig KS is definitely superior to fig Ks in 127, even though
it looks from Porphyry (37. 10 Sodano) as if he may have conjectured it,
appealing to the lack of distinction between O and Q in the old orthogra-
phy: (3 {Sis fiékriorog ’Ap1oTo¢>o1vng Kdreivo to év Toig, Hapontompioig
Aeyépevov, “6pt61o1<wv rig Kurd Kfiua ué7\0uvotv q>p'i;(’ 131roO\15Zs1 ixfiég.
OZ K6 ¢~civno1 Avxoiovog dpvéra onuév”, 6:-:i1<vu01v dag fipotprnuévov
1'>Tro7teiTto1'r0 éK tfig Trodtoutig ypauuotttlcfiq 013 vétp Xpfi To “OE K8
<1>o’wn01v” du<o13ew dag éipfipov t‘>rro1'ou<r11<6v, pc'i>\7\ov 6% oivfi 1’-zmppfiuorrog
atotpe1?\fi<1>6ou 1'06 dig, ii pc'i7\?\ov oifivosopov oc’1r1t65n' 6n7\0f>rcxt yétp “iva
<1><ivn1”-
It is not safe to infer from this passage that Aristophanes read
15Trtx7\15Ee1 with Philitas and Callistratus, especially as the report of Aristo-
phanes’ opinion may have been transmitted through Callistratus, his pupil.
We cannot accept this reading, as the only exegesis offered for it, namely
that the fishes’ diet of fat will protect them against chill, is ludicrous. Paral-
lels confirm that pékouvcxv <1>p'i1<0t refers to the ruffling of the sea surface,
as caused by the wind or by a creature in the water.95 imoiifiet suits the
required sense weil. But the fact that oiiooto elsewhere always has a long
alpha in Homer (100 instances) is troubling. Unless the variants conceal
some different verb that no one has yet succeeded in divining, the
anomalous scansion, together with the awkward connection between this
and the preceding sentence, makes us suspect an interpolation. It is diffi-
cult to see these two lines as the work of the same poet who composed
(for example) A, Z, or X. But to remove them alone would still leave a very
unsatisfactory transition from 125 to 128. V. Koch’s athetesis of 126-9
hardly makes things any better. Payne Knight condemned the whole
plural address 128-35. Aristophanes had already athetized 130-5 as an
insertion by someone who thought that Scamander’s anger in 136 needed
clearer motivation. Perhaps 126-35 is all secondary, and the river’s wrath
95 H 63 f. éxstiotto wtévtov i-Em dppifi tlB7\CiV81 E5:-'; re rrévtog int’ ouitfig; \P 692 dag 6’ 60’ into
q>pn<E>g Bopéw 0'wom0’t7\7\em1 i)(B1Sg; 6 402 (of Proteus) p6?\0£iVl]l q>pu<‘1 i<0O\u¢>6ei<;, all quoted by
Porphyry.
I
f
I
was provoked by Achilles’ confident statement that it would transport
Lycaon’s body to the sea.
<I> 146. Already in I-Ierodian’s time there was uncertainty on whether to
write ‘C3cx'i1<"r0tuévwv as one word or two. He observed (on 301) that the col-
location is similar to X 72 éipni Krauévwl, and that the meaning is simply
that given by the. two elements taken separately. He might have added that
there is not (and could not be) a verb "'80£i1<Teiv<.0 or "‘dpn'ii<teivw.
By the same reckoning we ought perhaps to write 66 Kriuevog rather
than euxriuevog. There is no verb "'et’>ic'riCw. But the participle Kriuevog
does not occur by itself, whereas Krtipevog does, as do 606i and dipni. So
there is some ground for supposing that the poets thought of iliiktiilevoc;
as a single lexical unit.
1
<I> 252. The scholia (bT, h, G), the Etymologicum Genuinum, and Eusta-
9
thius attribute the reading uékotvog TOD to Aristarchus, and then quote the
refutation, which is (in the wording of bT) dtvvoei 5% 611 6 Trolntfig To
“Ton” éy1<7u'ru<ov ouoénw Aéyet (timtét ii "rev M-Ever ii Teo 01(X7\6?\U|JéVUJQ,
Eust.). This is a typically Aristarchean observation, and dyvoei at iiti a
-.» .- _i. _- .
typically Aristarchean expression in criticizing a reading of one of his
predecessors, usually Zenodotus. But as Zenodotus’ text was unaccented,
i
he hardly comes into question here, unless he discussed the text in one of
his critical works. It may be conjectured that in the scholiasts’ source
I ’Apiototp)(og was an error for ’Ap1o1'0q>0’wng; there are other places where
l the names are confused.% If so, Aristophanes knew MEAANOCTOY as the
transmitted reading; ‘black—-boned’ was a nonsensical interpretation, and he
was laudably reluctant to read T00 as the article.”
ii
“i
1
<1) 307. The eccentric variant of p12, Tl‘0tvT0l96lv g;§Q[, has generally been
supplemented to make a participle: efiopuefiv or efielfiofiv Gerhard, efiotiptov
van Leeuwen, ilvpduevoq, Allen (S. West [1967], 154). But in the five other
places in Homer where Troivtofiev is followed by efi/EEK, it is a genitive
phrase: er i<eu0u<I>v N 28, er: uekéwv H 110, ‘P 689, ex i<MouI>v W 112, £35
dtvptfiv p 171. The best supplement will be sfi olpéoov. All the mountain
l
E
-1 9“ Porphyry’s wording is ambiguous (p. 275. 9 Schr.): oiire, dig rbnotv ’Apior0tp)(o<;, éyi<?\i—
vovreg Eiv irpoqmpoipeflot, “pé7\0tv6<; 'tou”' émtnoetiwv yup eine “1015 Bnprrrfiporg”, em 6 pékag
totoiirog. Is he citing Aristarchus for the reading or for the rejection of it?
97 Porphyry l.c. continues: d0tJ\6c to dipflpov 7\u1rs"1 rrpocmeipevov. This objection to T06 is
then answered by quoting some other Homeric examples of the article (A 11, A 614, E 460).
i
Again this looks like Aristarchean argumentation. Aristarchus commented on the force of the
article at A 11, Tl 358, T 147, and was well aware that the article is not entirely absent from
Homer. Cf. Lehrs (1882), 355f.
260 II. Notes on Individual Passages
streams that feed Scamander contribute to his fury and flood; cf. M 19,
II 392, and the terms in which he appeals to Simoeis in 312.
<1» 323. TU|.1[30)(OfiC2 see West (2001).
(D 345-8. Aristophanes’ reading in 347, corruptly reported in T as
efievotiveiv, was evidently a form of éfiotvaivw, whether itfiotvaivnl (Heyne)
or efiotufivnl (Spitzner). This raises the question whether he did not read
the same verb in 345 and 348, as the requisite form, éfinvoivfin, differs from
the vulgate e’<§np0iv0n only by one letter, and that not a very dissimilar one
(P: Y). Enpaivw does not occur elsewhere in Homer; oriaotivw is found at
1 321 (otiaavfiév), and the compound is current in Ionic prose.
(I) 384. There seems much to be said for Doederlein’s Xwouévw Tl'8p2 the
combatants’ wrath had been aroused, but she persuaded them to stop. Cf.
E 260, where Zeus’ respect for Night deflects him from his pursuit of
Hypnos, ii 5’ errcxéoorro Xwéuevog Ttep. The participle agreeing with the
unexpressed object might easily have been changed to agree with the
expressed subject.
(I) 397. Trcxvotptov: see West (2001).
<I> 455. The manuscript tradition offers the choice between two verbs,
duto-7\éIrto and (intro-1<67rTw. The Berlin papyrus p449 (first century BC)
adds a third, dirt-07\6TrTto, and Eustathius a fourth: 1246. 37 éihkot at
dmokouoéuev, 1’]’r01 1<o7to[iu5oe1v' Aoiioov yoip, cboroi, JT(Xp0( Kvrrpioic; to
i<o7\0[36v. Cf. Hesych. 7t 1271, 7\oi300v' i<67\oupov, |<o>to[36v, refipowouévov,
and or 6466 drrro7\ouoéuev0u' i<o7\o[3o$oe1v, which presumably also relates to
this passage. This last variant I suspect of being a learned conjecture based
on knowledge of the Cypriot adjective, from which it is not clear how a
genuine verb form dnrokovoéuev could be derived.
Of the other forms on offer, dt7toKo1pépev ‘chop off’ (the vulgate reading)
looks the most natural. In the Odyssean references to cutting off some-
one’s (nose and) ears the verbs are duo roiuvw (0 86, X 475--6) and ditto
ciuoiw (<1) 300-1). dtrokénw is a choicer verb, but properly means ‘peel
off’; map‘: voip {501 t Xakrog €?\e1pev I ¢i5>t7t0'i rs icon‘: cj>7\ot6v is said in A 236f.
of the speaker’s staff. Aristarchus, we gather from Aristonicus’ note, had
dtTro?te1|J€uev before him and commented that its use for dtrrolcowetv was
catachrestic. The reading of the papyrus, 0<Tto7\ot|1euev, might be dismissed
as a conflation of —I<0t[J— and -?te1|:—, were it not that there is a verb
67\61rrw, used by Alexandrian and later poets, especially of pulling out hair,
by Nicander also of stripping out the central pith from the giant fennel
(T/oer. 595). This seems as suitable to our context as dnokewéusv, and it
would be a tenable hypothesis that the latter was corrupted from
dflohowéuev. No sure conclusion emerges.
II. Notes on Individual Passages 261
%Hw@
98 Or at any rate not, I think, before Aesch. Ag. 206, fiotpefiot pisv Kfip to pi] mfléofiotl.
99 See West (1997), 223 f.
262 II. Notes on Individual Passages
reason he gave the name Ileus to the child she bore him. Perhaps the iden-
tification of this Ileus with the father of Ajax is secondary. In any case we
have an Ileus associated with Troy, and ’I7\fi'iov is the correct adjectival
form. ,
Besides ’I7\fi’i0v, Hesychius has the entry ei7\fi'i0v' ev r‘]7\iw1 Bepuotvfiév.
This no doubt represents another ancient attempt to deal with the unfa-
miliar toponym.
(D 570. Aristarchus’ reason for athetizing the line seems to have been the
apparent contradiction between the observation that Zeus was granting
Achilles success and the argument that Achilles was only mortal and that
Agenor had a chance of killing him if he faced up to him. This is not con-
clusive; the sense may be, ‘he is not invulnerable, he only has one life, peo-
ple do not say he is anything but mortal; his continuing success is only
due to Zeus’ favour (which may be withdrawn at any time).’
(I) 575. So far as I see, van Leeuwen is the only editor who has adopted
the pre-Aristarchean (probably Zenodotean) variant Kuvukotyuov, or (as I
prefer to accent it in obedience to the usual rule for nouns with a modify-
ing prefix) Kvvuhavpov. Aristarchus objected to it on the ground that
137\0tvt16g ‘barking’ is something specific to dogs, so that there is no point
in a compound ‘dog-barking’. But it was observed that the word occurred
in Stesichorus, who was inferred to have read it in his Homer. Whether or
not this was Stesichorus’ source for it, his usage establishes that it was a
real word. It is the oldest attested reading in our passage, and the dzffici-
Zior lectio. It also gives a better rhythm, as érrei Kev 137\0tw.10v dtxotiom
involves a breach of Hermann’s Bridge.
(IJ 596. The articulation 060’ Er’ Ectoev is preferable to 01356 T’ iioroev. The
case is similar to A 437 and ‘P 730; see the notes there.
X 81. Where a speech was originally introduced by a non-formulaic.refer-
ence to utterance, such as 00t5pe1'0 in 79, a more banal speech-introductory
line is frequently interpolated, as the external evidence shows; cf. above,
pp. 12f. This makes 81 potentially suspect; and its repetition of 6o’u<pu )(é0v—
001 from 79 makes it quite probable that Bekker was right to condemn it.
X 141. The two earliest papyri, 12 and 271 (both Ptolemaic), give Se T
vTr0t100t, implying psilotic 1’J‘Tr0u00t. This is plausible for what is regarded as
an Aeolic form; see G. Hinrichs, De Homericae elomtionis vestigiis Aeo-
licis (Diss. Leipzig, 1875), 61; H. ]acobsohn, Phil. 67 (1908), 492; Wathelet
(1970), 284 with n. 33. It might be thought that a psilotic form could not
have maintained itself beside 07501, 13310, i5Trep0e, etc. Yet we find 1’5u|Jt-:6;
besides uueig, so the possibility must be allowed. On the other hand the
writing of T for 0 may be merely casual.
II. Notes on Individual Passages 263
X 142. The spelling ertoutooel in p673 (first century BC; not a very
carefully written copy) provides what is to my knowledge the first attes-
tation of the form postulated by philologists as the basis of (710000. See
Wackernagel (1955), I 587; Chantraine (1968-80), 39; West (1998b), xxx.
X 287. To the left of this verse there is a note in A, ev (’i?\7tt01 “Kori 0¢>1v”.
This does not make sense as a variant for Kori KEV. It may be conjectured
that ocbtv was a variant for oqnot in 288, and that the scribe of A (or a pre-
decessor), finding ocbtv in the margin, mistook its reference.
X 336. d.'1'1<1.6g is the form offered by the tradition in place of the regu-
lar (but unmetrical) dtetkéwg. The T scholium reports that some scanned it
as a disyllable (0ti1<6f><; or dtitdig), while others said that it (the second syl-
lable) was shortened, i.e. d'iKt5c_3. The latter would have to be explained as
a unique zero-grade form of the root, F1K for Fem, which is highly unlike-
ly, or perhaps (as Wackernagel declared) impossible.""’ We must accept that
on represents the contraction of 0t(/-_)e1, to be written d1 (fit) — the normal
Attic form. Such a contraction, though not otherwise found with this
word in Homer, is paralleled by d0?\0ci>6po1><; (I 124, al.) < dFe07\0-, dxovre
(E 366, al.) < d/—_é1<ov1'e, éitoouou (Hymn. 6. 2) < d/—'eioo11ou. In the ter-
mination, however, —tl’)g should be resolved to -étog; cf. e 98 and T 269
vnueptétog, 6 419 and 459 dotepciaéwcg.
X 371. This may be an interpolation of the type where a word is added
to the previous sentence to make something more explicit and then the rest
of the verse is filled out with inessential verbiage (cf. above, p. 12). Cer-
tainly 370 could stand well enough without the addition of "E1<T0pog. The
remainder pointlessly anticipates the information to be given in 375, and
it is awkwardly phrased: ‘and none stood by him unwoundingly’ is
a strange way of saying ‘without inflicting a wound’. It looks as if
ourrioaoxe napaordg in 375 has been twisted back to make 0130’ éipa 0'1
Tu; dvournrei ye rrorpectn.
X 374. I have preferred the imperfect evérrpnflev to the better supported
aorist, because, although Hector did set fire to a ship, that was an abor-
tive beginning of a plan to destroy the whole fleet, which could not be car-
ried through. The Achaeans now look back to what he was when he was
posing the greatest threat, rather than to what he was when he completed
his highest achievement.
X 415. eEovouou<7\fi0nv is written in W and G as efi 0vou0u<?tfi6nv, and
this is supported by 0 278 ex 5’ 0v0110u<7\fi0nv. The efi can be taken
with the following ovouoiltov (and in 0 278 with ovduotlieg), cf. I" 166
efiovoufivntg. eEov0u0u<?u‘|0r|v would be perfectly good as an expression of
tivoua e1<1<(x?\éwv, only e1<i<tx7\é<n in Homer means ‘call someoneout’ (of a
house, etc.), not ‘call aloud’ (Q 582, K 471, 0.) 1, cf. [5 400).
X 432. dt1ro0vfit0l<w occurs three times in the Odyssey, but not other-
wise in the Iliad, where the normal compound is the Ionic 1<otT0t0vfi1o1<t0
(14 times). Cobet called for oe13 i<otTotTe0vntI5Tog here, and as one
manuscript is reported as giving this -~ the dzfflcilior lectio — there is some
temptation to take it seriously. I also draw attention to van Leeuwen’s pro-
posal to detach the_ prefix as a preposition governing the pronoun, ‘How
am I to live apart from you now that you are dead?’ For this strong sense
of drré cf. B 292, A 292.10‘ But emphasis on apartness is less relevant to
Hecuba, for whom Hector was important as the city’s champion (432-5)
rather than as a companion, than it would have been to Andromache; and
the genitive absolute is characteristic.
X 436. This verse, repeated from P 478 and 672, does not appear to have
been suspected previously, but seems very weak and pointless here where
the fact of Hector’s death is the whole basis of the speech.
X 451. The reading 0t131'fig reported by Allen from the twelfth-century
Vaticanus gr. 903 is of some interest. It might be dismissed a mere mechan-
ical echo of 0t’100in<; eicvpfig earlier in the line, though it is doubtful
whether this factor would have sufficed to disrupt the obvious concord of
poi 0113'rfit. The double dative of the paradosis is quite acceptable as a
oxfiuot K010’ i57\ov 1<ot’1 uépog; see examples in Kiihner-Gerth, I 289f. and
430. However, otfitfig is possibly more idiomatic, besides being the diffz'ci-
liar Zectio. We might compare the combination of dative enclitic pronoun
with genitive extension in II 531 iittt oi tiix’ 1"j1<o1Jo'e pevotg 0:-30g e13—
Eauévoro; Aesch. Pers. 633 6’ diet 1101 uorxotpittxg ioofioriutov [30ro17\eiJ<;
fidpfiapa outbnvfi iévrog T0: Tl'O£V(X107\‘ oriorvfi 515o0p00r [3o’1vu011'o1;‘°2
X 454. oii ytitp drr’ oiiottog ein 611613 errocg, ‘may such news stay far from
my ears.’ We might have expected an enclitic pol, as in the parallel expres-
sion at E 272, oii yup 51"] 1101 dtr’ o1’50t'r0<; dnfie vévorro. The form ein-:13 (for
which we should require euéo or euei’) gives an unwanted emphasis, and
van Leeuwen’s change to the adjective euoii is welcome; o1’50tToc; euoii = pot
oiiottog, there being in the possessive adjective no distinction of emphatic
'°' Scribes sometimes write ditto in such cases, but there is no warrant for this when the noun
follows.
‘°2 Cf. Wackernagel (1926-8), II 77.
II. Notes on Individual Passages 265
and unemphatic forms. It was easy for someone to read it as the Attic
genitive form of the pronoun and to Ionicize it as euefi. Cf. above on
M 283.
X 485-99. Andromache’s lament is one of the emotional high points of
the Iliad. We are in the presence of great poetry here, but also of a pas-
sage irreconcilable with the view that our text is the product of a straight-
forward process of dictation by an oral poet who never altered or added
to his primary effusions. Aristarchus athetized lines 487-99 as being
d010i0e'r01, ‘not well arranged’. He thought it unrealistic that Astyanax
should be represented as having to go begging for his drinks while his
mother, grandfather, and various uncles were still alive, or that a prince of
the royal blood should be in danger of having his arable lands stolen.
These are not telling arguments; but more to the point is the observation
that what is described (at any rate from 490 to 499) is the plight of orphans
generally, not specifically of Hector’s son. The introduction of Astyanax’s
name at 500 effects an abrupt reversion from the general to the particular.
That forced transition was surely never made by an oral poet: it results
from the expansion of a written text. Aristarchus proposed to make 500
follow on 486:
oiite 013 rotittot
Eooeou, "Enron. Eivetap. emit Bdvec, oiire ooi o13to<;.
’AoT1>dvot‘<§, Be; rrpiv pi-JV KTA.
This is an improvement on the sequence 499/500, but still not quite natu-
ral. It seems to me more likely that 500 ff. originally continued 484:
Q I
In that case we should suppose that the poet composed an alternative con-
tinuation, consisting initially of 482-4 + 500-5, and interposed it. 505
Tf07\7\0( Tr0i0r|o1 invites elaboration, and 487-99 (or just 490-9) would stand
well after it. In other words:
First draft: 477-81 + 508-14.
Second draft: 477-84 + 500-5 + 487-99 (then what?)
Final conflation: 477-s4, >=-485-6, 487-99, 500-5, ==~5os--7, sos-14.
("‘ = bridging couplets)
ii’ 58. The variant fifi Keiovtec; in the la scholia is probably no more than
a learned recollection of Didymus on A 606. 4
‘P 74. Notions of the other world are notoriously liable to exhibit incon-
sistencies and contradictions. Yet it appears a particularly flagrant contra-
diction to represent Patroclus as unable to pass through the portals of
Hades, or to cross the River (an alternative boundary-marker), and then
in the next breath to say that he is wandering div’ e1’>p1>Tt0?ti:<; "A'i00<; 065,
‘over’ or ‘throughout’ the House of Hades, the very place he is excluded
from. This makes me suspect that 74 is the addition of a different poet.
\I/ 88. The ancient variant dud)’ doTp0tv0‘t7\n1o1v epioootcg, attributed by
Didymus to (xi 7t7tei01>g Ttbv Kort’ 0iv0p(x, differs in two respects from the
vulgate: in having a different verb, and in giving a feminine form of the
noun. As to the latter, Didymus comments that it is ‘more Ionic’;'°4 he
quotes Anacreon PMG 398, 0’10Tp0q(o17~.011 0’ i-ipwtog e’1o1v | uotviou T8 Kori
I<U0Ol}.101. The masculine form is guaranteed at E 466, where the word has
its primary anatomical sense. It also appears in another Anacreon fragment
(PMG 388. 2 Zuhivovg doTp0tv0‘t7\o1>g), where the reference is to ear-rings.
This suggests that the feminine form may have been confined to the game.
In this connection it recurs at Herondas 3. 7 and Leonidas epigr. 45. 3
Gow-Page (AP. 6. 309).‘°5 Where Herodotus refers to the game (1. 94. 3)
the manuscripts (or at least the editions) give T0311 dotpavoihwv, but pos-
sibly dto1'potvot7\ét0v should be restored. In Attic the masculine was used of
the game, as in other connections, and Attic tradition would account for
the elimination of the feminine in the Herodotus passage and in the
Homeric vulgate, if duo’ dtotptxvddtntotv was the original reading. It may,
on the other hand, have been a ‘new Ionic’ intrusion. At any rate it can
claim to be an old variant.
i--i-_-»---»»-s»---.-------»
1°‘ Cf. Ael. Dion. 01 190 (from Pamphilus), where the Homeric variant is again noted.
‘°5 Cf. Headlam-Knox and Cunningham on Herondas l.c.
i
?
|
i
5
1
II’ 104. Didymus records an objection to this line which on the face of
it seems well founded. Patroclus’ ghost has spoken to Achilles t’-Zjltbpdvtog
Kai o1>veTt6g. Why then should Achilles conclude that the souls of the dead
are devoid of opéveg? évoéoetorou 013v at Tfig ’O61>ooeio1<; 0 ori;(og' exei
‘i vétp rug tuuxérg e’i6w7\0r oi<1u36n cbpovfioewc duéroxa iméfieto. This cannot
5 mean literally ‘the line has been interpolated from the Odyssey’, because it
does not occur in that poem. But it does correspond to the doctrine of the
Nekyia according to which only Teiresias’ qnpévecg are €uTte60t, while all the
other souls are mere flitting shadows, incapable of rational speech until
i
fortified by a draught of blood. The ancient critic argued that this Odys-
i sean doctrine has been inappropriately imported here. He is not assu-
mingiof’ that the Iliad and Odyssey were by different poets, for ‘Homer’
has to be understood as the subject of 1‘JTté0eTo: in the Odyssey Homer
adopted a different presupposition about the condition of the dead.
Having reported the objection, Didymus records a way of meeting it:
i ‘Alternatively, by tbpéveg here he means not the rational faculty but a phy-
1 sical organ, as in other places (1 301, II 481), and this stands for the body,
1 pars pro toto. Thus Aristophanes the grammarian.’ Was this Aristophanes’
i
answer to his own difficulty, or to a predecessor’s? More likely to his own,
I think, given the detail with which the original objection is stated. In any
'5
case the suggestion that Achilles means ‘they have no physical bodies’ -
3
appropriate as that would be to the situation - is wholly unconvincing. To
ii
911
7¢
express that idea he would have had to say something like ii 00‘: Ti eon Kori
r eiv ‘A1600 66uo1o1v | e’i'6w>tov' ooipxeg st rod ootéot our Em rrduitorv.
1
.2
§
There follows in A the note of Aristonicus: fi 61Tt7\fi 66, iitt trig TLOV
drdqnwv 1p1>)(0t<; "Ounpog En owtfiotioocg tijv tbpévnotv 131ro'ri0e'rou. This
gives us Aristarchus’ justification of the line, which differs from Aristo-
‘..__
‘i
‘a
the problem, which is why Achilles should conclude, from his encounter
with the ghost, that the dead are devoid of tbpéveg. His words anyway
ignore the fact that Patroclus is not, strictly speaking, in Hades; he takes
his experience to be evidence of the condition of the dead (without regard
i to the question of burial), and by ‘in Hades’ he means no more_ than ‘dead’.
‘°“ As implied by Erbse’s paraphrase, ‘versus factus sec. animarum cognitiones, quales poeta
Odysseae sibi finxit, in Iliade pro spurio ducendus est’ (V 385).
‘°7 Both are reproduced (less precisely, as usual) in the bT scholium, introduced respective-
ly by rtvég (= Aristophanes) and i<d.7\?uov 66 (= Aristarchus).
268 II. Notes on Individual Passages
~m
‘°* Cf. Pl. Phd. 63c (quoted by Richardson), sf5e7\:rrig situ sivai T1 1"o'i.<; retsheotnxéor.
‘°" Whether or not this loose expression goes back to Didymus itself, it presumably stands
for two (or at most three) of the individual city texts.
II. Notes on Individual Passages 269
Tc 'Ir6?ug T6 used of another world’s-end people, the Cimmerians (7\ 18), and
Hesiod uses it of the Black Men of Africa (1<uo’wco1 éivopeg, Op. 527); cf.
also the ofipog ’Oveipwv of to 12.
‘P 226. Many nineteenth-century critics objected to ‘Ewo"¢»6pog as an un-
Ionic form; they wanted somehow to restore "'Ijoo¢6p0<; or ’i'fit0o<i>6pog.
However, there are plentiful Homeric parallels for cw < n(F)0, and indeed
for its treatment as one syllable.“° The prosody remains surprising —
Hesiod scans ‘Ewotbdpog as four syllables, Th. 381 -~ though Pindar pro-
vides a sort of parallel with his trisyllabic ’A0o<l>6pog (if that is the correct
spelling)m at [st/0. 3/4. 42. One might simply delete the connecting par-
ticle and read fiuog ‘Ewocbdpog (so Porphyry), but everywhere else in
Homeric narrative fiuog is combined with Z56. Christ proposed changing
the conjunction to 80T6; he pointed to v 93, where €0T€ is picked up in the
main clause by Tfiuoc, and supposed that it was altered to fiuog here be-
cause of the correlative Tfiuog in 228. It is possible that such things
happened in the course of the transmission, but it is a bold hypothesis in
the absence of guaranteed cases.
The temporal clause in that Odyssey passage reads
eiit’ dotfip unepéoxe cbotoivrotrog. iic re uoihtota
iipxetou dvvémwv q>o'iog fiofig fipiveveincg,
and it prompted Brandreth to a different conjecture in ‘P 226:
fipog 5’ dotfip eioi cbowg épéwv érft vofiotv,
6v re uérot KpOKOTf8Tf?\OQ imetp éiha Kiovottou fitég.
This is extremely attractive. In X 26, similarly, Sirius is introduced not by
its name but by dtotépot, its identity being then specified in a relative
clause, 5g poi T’ orrtépng eiow, dpilnkot 66 oi ouivoti (and then confirmed
in another, Eiv ts Ktiv’ Dpiwvog éirizdtnotv K0t7\éOUO’lV). And at X 317f. we
have
oiog 5’ dotfip eiot pet’ dotpoioi vuxtog duo7q/<51
"Eon'epog. 8g 1<0’t7\>t1ot0c; év ot’>pavo31‘iota'r0u dotfip.
where oioc 5’ dtotfip siot is a precise formal parallel for uog5’ dtotfip aim.
It could of course be argued on the other hand that the naming of
"EoTrcpog there protects the naming of 'Ewo<1>6pog in W 226.
\P 244. "Aiot KM-Zfifiwpatz see West (2001), which perhaps justifies the
removal of the obelos.
-—--------u-----------------w-—u
~ww
“Z In F 51 the nominative and accusative were ancient variants, but the accusative ‘in appo»
sition to the sentence’ is clearly right.
‘*3 So the best manuscripts; others ueuvéto.
“‘ V.l. pe|Jvt5(1)ue90t.
‘*5 Cf. Thumb--Scherer (1959), 263; West (1974), 89.
II. Notes on Individual Passages 271
‘*6 Allen, Homeri Opera V 150, included it as no. 22 in his collection of ‘versus heroici
I-Iomero adscripti qui neque in Iliade Odyssea Hymnis neque in Cycli fragmentis inveniuntur’,
and it is ‘Homerus’ fr. 2 in M. Davies’s Epicorum Gmecorum Fmgmenta (p. 105). Clem. Strom.
1. 4. 1, "Ounpoc; 51$ Kori téxtova oo¢ov 1<ot7\e’i, need not allude to the same hemistich but may
be accounted for from O 411f.
II. Notes on Individual Passages 273
(though oocbin does); ijpape réxrwv occurs only here and in A 110, Kai T5:
pi-iv dtoicfioag 1<ep0t0E6o<; ijpotps T61<'rwv, which cannot be the source of the
quotation.
‘P 721. ‘But when the Achaeans began to get bored’, or ‘but when they
began to bore the Achaeans’? The first sounds more natural in English, and
(i(Vl(§‘(§(.0 is more commonly intransitive in Homer (Z 300, <19 270, 5 460, 598,
X 87; transitive only at 1' 323). 5 460 provides a particularly close parallel.
We should therefore prefer the nominative - already known to Aristarchus,
though rejected by him, and given by three papyri - to the accusative.
‘P 730. On the choice between 01556 T’ and 0135’ 6T’ see in general Den-
niston (1954), 531. In the footnote on that page he mentions the present
passage as one where 611 ‘is not very suitable’. But it is perfectly idioma-
tic, expressing the fact that Odysseus was unable to advance further the
course of action on which he was embarked. ‘He shifted him slightly off
the ground, but did not go on to lift him up.’ Cf. above on A 437, (I) 596;
C. Ruijgh, Amour de “T6 épique” (Amsterdam 1971), 707.
‘P 744. ‘Phoenicians brought it over the sea, landed in harbour, and gave
it to Thoas as a gift; and ]ason’s son Euneos gave it to Patroclus in ex-
change for Priam’s son Lycaon.’ It is curious that the location of the
harbour and of Thoas’ kingdom is not specified. I suspect that a line may
have disappeared. The natural sequence would be
<I>oivu<e<; 5’ éivov éZv5peg 6n’ fiepoe156ot atévrov
<Afiuvov ix; rjvott-)6r|v . . . >
otfioav 5’ 6v Puuéveoot, ®6otvri 56 5d3pov 65¢-motv.
An alternative would be
<I>oivu<ec; 5’ éivov div5peg err’ fiepoe156ot rrdvrov,
otfiootv 5’ 6v Afiuvwi, [5oto17\fi'i 5i: 5tiipov 65w1<0w.
In this case the king would be the one named in 747, Euneos, not his
father Thoas. The desire to replace the unspecific lictoiitfii by a definite
name would have been the cause of the rewriting.
For orfioow 5’ 6v Afiuvwi cf. E 258 = p 427 otfioe 5’ év Aiytirrrwi
rro'r0tuo31 véotg, T 188 otfioe 5’ z’-av ’Auv1o651, 561 re orréog E’l7\616Ui1’]Q. On
the other hand the transmitted text of 745 can be defended by |.l 305
otijootuev 6v Fuuévi y?t0z<i>1>p<i5t e1’Jepv6a vfiot.
‘P 792. 6p15fio0t06ot1 must be taken as the paradosis; the other variants
are all corrupted from it, or are attempts to make it more metrical or intel-
ligible. The testimonies of Philoxenus, Hesychius, and p13 confirm it as
the reading generally current in antiquity, though Didymus found
éplcrjootoflou in ‘some of the commentaries’ - perhaps an etymologizing
gloss rather than a genuine variant.
274 II. Notes on Individual Passages
The form is baffling. The 6p15— root is appropriate, but what is the
verb? How can the long iota be justified? And why is it in the aorist? We
expect a present, as in N 325 Trooi 5’ 05 ‘)TOJ<; 6o1’tv épilleiv (also of Achil-
les’ unchallenged fleetness of foot), or <1) 184f. ;(o0\e1r6v T01 6p1o'0z-:v6og
Kpoviwvoc; | T£‘(Xt()’lV 6p1C6uev0u. To be more precise, we expect 6piCe1v or
épiléuevou or épileofiou, or 6p15aivc1v or 6p150m/éuevou; or, from other
roots, dvT1¢>spiCe1v or IOOQJQDICEIV. None of these, to be sure, would fill
the same metrical slot as is filled by the problematic form in the received
text, U — — — U. Perhaps the poet, with N 325 in mind, invented a form in
order to extend 6piCe1v to the length he required. But it is still hard to see
how he arrived at 6pT5rj0cx0'0ou.
The alternative is to suppose that it is an early corruption from some-
thing else. Naber’s Troooiv 5npio‘0to60u deserves mention, though it still
leaves us with an aorist. It would have to be assumed that the syllables 511
pl somehow became transposed and the pt then rationalized as 6p1-; not a
very plausible story. If the aorist were acceptable, one might do as well
with irdoo’ 6pt5<ot omiootofiou (cf. at 292 = r 11 i-fpiv otfioavteg 6v 1‘>u'iv);
when this was written with scriptio plena, Irooot ep150t, it might have
appeared that there were too many syllables after 6p15-.
Others have thought that the latter part of the line too was out of order:
’A)(17\7\e‘i is an abnormal contracted form,” and the instruction in schT to
write it with an eta, ’A)(1?\7\fi1, though not implausible as the result of con-
traction,“ has been taken to point to a variant ei pi] ’A)(17\fi'i, which in fact
occurs in T and R — something editors have failed to notice. As pi] ’A- can-
not (as some have thought) be compressed into one short syllable, this
means pushing the ei back into the fourth biceps and altering ’A)(ouo'i<;.
Now, there is an independent reason for considering the replacement of
’A)(ouoi<; by 6i?t7\o1<;: as Wackernagel pointed out,1“’ ei ufi ‘except’ is regu-
larly preceded by a form of 6'i7\7\og, as in P 477, Z 193, p 326, p 383. ’A)(0u0i
and oi d7\.7\o1 are variants at <1) 4 and ‘P 540.
If the second half of the line once read éi7\7\o1g, ei ufi ’A)(17\fi'i, then
6p15fioao6ou is a syllable too long. Hence conjectures such as Brandreth’s
6ptC6uev0u, Lange’s 6piCeo6ou, and Menrad’s 6pt5otiveo00u (the active
6pt50uv6uevou would have been better). The possibility that the text of cer-
tain lines suffered such extensive remodelling in the earliest period of
"7 Paralleled in the Iliad only by E 115 Hopfiei and Q 61 lIr|7\e'i (where elided IIr]7\fi’('i) is
possible). V
"8 Cf. the old Attic Tr67\n1 (from 'n'67\n‘i), later pronounced and written Tl'6?\E;l.
1"‘ Rh. Mas. 44 (1889), 631 = (1979), 1s2o.
H. Notes on Individual Passages 275
2
transmission cannot be ruled out. On the other hand the uncertainty is
obviouslytoo great for such conjectures to be adopted.
‘P 864. The absence of the line from two papyri and part of the medie-
val tradition is a strong indication that it is not authentic. Leaf says it can-
not be dispensed with. Certainly 0135’ firraiknoev dvotrttt seems very
abrupt on its own. But éivomti for ‘Apollo’ is abnormally elliptical
anyway. The explanation is, I think, clear. The poet did not originally men-
tion that Teucer failed to make a vow to the archer god, only that Merio-
nes, drawing his bow at a venture, made one. 863, as we have it, is second-
ary to 872; the poet felt it sufficient to say éivoucrt because he had named
Apollo in the parallel passage, even though that was to stand later. For the
same reason he did not think it necessary to spell out the content of the
vow that Teucer failed to make.
Once the afterthought was incorporated, the text read
Tefixpog 6?»: Trpdfiroc Kkfipwi 7\0i)(sv' ou31'i1<a 5’ i6v
fiicev é1ru<porréw<;' 0136’ fineiknosv éivomn.
<’5pv160<; pi-zv éiucxp're' pévnpe voip 0'1 T6 Y’ ’Arr67t>twv'
I
otfitétp E‘) ufipivflov [idle 1<1'?\.
But what had it read before? It is not difficult to guess:
s Tsfmpoc; 6% rrptiiroc 1<7\fipo.>i 7\0'o(av' ouirirot 6’ iév
fixsv énu<porrétoc;' Kai fa’ 6pv160c pi-JV éiuaptev,
5
f
ouiréip ii ufipwfiov fioiks 1<1"7\.
‘P 870-1. Variants of this couplet are recorded from the Massaliotica and
Antimachus. In the Massaliotic version, Meriones was not said to seize the
bow from Teucer’s hands, but sirnply to set an arrow to the bow, for he
i
II
i.
already had it (the bow?) in his hands, dag iifiuvev. Here it is not clear what
is the point of the explanation év ytiip Xepoiv §)(ev Tr0i)\ou. Antimachus’ text
is given differently in A and T: éfisihero Téfiov I )(epoiv' (A), or éfieipuoti
1 Teztilcpov | T6Zov' Xspoi 6’ 6'i0‘Tov A is no doubt right as regards the
r.
verb in 870, as T’s éfieipvoe will represent assimilation to the vulgate, but
SI
Ii for the rest, we must accept T’s version; words have simply been omitted
E;
1.
in A. In Antimachus’ text, then, Meriones ‘took the bow from Teucer; the
arrow he already had in his hands’. This does not differ in substance from
2.
5
the vulgate, only in its wording, which seems inferior.
$
5 The vulgate text is entirely satisfactory, except for the obscurity of the
l final phrase, Gog ’i6uvev. Elsewhere ifiiivoa is used in the active of a deity
l
Ii
r<
3
f.
directing a hero’s arrow or other missile towards a target (A 132, E 290,
P 632), but in the middle for aiming one’s own missile (Z 3, X 8). This
2 distinction of voice appears not to be applicable here, as there is no ques-
I tion of Meriones’ influencing the flight of Teucer’s arrow. The reference
l ,
i
must be either to Teucer’s shot or to Meriones’ taking aim; but the latter l
is difficult, as he is not said to get his sights on the bird until 874. Fur-
ther, €)(ev rroihou, dag ’i6uvev should mean ‘he had had the arrow allthe time,
just as he had aimed (or had been aiming) it’. But he could not usefully
aim it before he got the bow. Commentators are accordingly driven to
suppose that the meaning is ‘while Teucer was shooting’ or ‘ever since Teu-
cer shot his’. So sch 870—1a2 (bT), éiorov yoip. éij E5100 érsivog ’i6vvsv érri
Tov (Ii<071'6V, Kai oi‘3r0<; sixty; sch 871 (T), év iiowt érsivoc éréfievev, Kori
0i‘5T0c; 1<0tTs'i)(e Tov c’>'ior6v. Leaf admits that ‘the change of subject in ’i6vvev
i
is very harsh, and ting does not mean while. The last difficulty might be met
by reading iitocg with synizesis, and Voss’ dag ’t615vo1 to a certain extent
makes things easier (cf. iitbp’ ’1615vo1, that he mig/at steer, 6 255).’ Voss’s con-
jecture, adopted by Nauck and Doederlein, would mean that Meriones had
been holding his arrow ready so that he could take aim (that is, as soon as
the bow was available). This does not seem very natural. I
I
‘P 879. A and T disagree over whether the reading of the Massaliotica .'l
was Moiofinv or Moiofin. The former is Zectio dzfficilior, and has a parallel r
r
in the third person plural uldvfinv at A 146, for which cf. Chantraine -l
(1942), 471 f. It is unlikely, therefore, to be a casual corruption of the read- 1
ri and nearly all of the medieval tradition. This is, however, impossible: a
character in the narrative cannot be referred to with 556. The demonstra-
tive is Tdv, and the oé following it must be the sentence-connecting par-
ticle. We must either read Tov at T’ with 9 and \W, or simply Tov 5%.
as T6 is hard to justify here. The otheralternative leaves us with a hia-
tus, but one of a sort found in several other places before contracted forms
of éoito, where restoration of the uncontracted form (or a metrical equiva-
lent) eliminates the hiatus. Here Grashof’s Tov 5’ é0‘t0to1<ev would achieve
.4
this. But perhaps we should think rather of siotorcsv, as found in E 802 (at
verse-end), A 125, T 408, a form in which the initial §;— has been length-
II. Notes on Individual Passages 277
ened to compensate for the loss of quantity resulting from the contraction
of ’*-~0iso1<:t-:v to -0to1<ev. In those instances the poet did not have the op-
tion of leaving a hiatus; here, given the ambiguities of early script, we can-
not be sure whether he sang T<’)v 5?; iiotottz-iv or Tov 5’ 8’fOLOI<6V, as either
might have been written T0v5se0toKev.
Q 54. This verse was apparently already known to Aeschylus, who
adapted it (in the context of 50--3) in his "Elctopog 7u5'rp0t (fr. 266). He
arranged the ideas in a more logical sequence: Hector is admonished,
‘maltreat a corpse if you like, it feels no pain, but we, the gods, observe it,
and our véusotg is aroused’. In the original, however, 54 does not follow
very naturally after 53. It looks like an addition by someone who mistook
the sense of 52, 013 uév oi T6 vs 1<0‘0\7uov 01551-5 1" éiustvov, as if it meant
literally that dragging Hector brought Achilles no advantage.
The bT scholium records an athetesis of 53, without identifying the
scholar responsible (probably not Aristarchus). That line too was in
Aeschylus’ text. But it is prosodically abnormal (veusoonfléwuév oi), and
if it was in fact spurious, and absent from the text known to the author of
54, his interpolation becomes somewhat easier to account for.
Q 74. Herodian reports that certain critics had interpreted Bewv as Géwv,
‘someone run and fetch Thetis’. This has better parallels in the Iliad than
commonly noticed (K 54; M 343). But it is less appropriate to the gods
and to the actual situation than to the urgent contexts of K and M, and
not at all appropriate to the summoning of Thetis, who lives at the bot-
tom of the sea and is not be reached by ‘running’. We must therefore agree
with Herodian’s decision for Gt-Zdw, genitive dependent on Ttg.
Q 134. It is surprising that editors have allowed enclitic 001 to stand.
The only Homeric form is Tot. There is no case here for the emphatic form
of the pronoun; it is excluded by the word order, and by the parallel 0'1 in
1 13.
Q 150 2 179. All editors, so far as I see, print the inoffensive-looking
fi5i3. With this reading the herald’s duties are stated as (a) driving a mule-
cart (purpose unspecified) and (b) bringing Hector’s body back to Troy.
The variant ii K8, which has respectable attestation in both passages, gives
a tighter construction: the herald will drive the mule-cart which will con-
vey Hector’s body. That this is the true reading is made virtually certain
by the parallel of Z; 36-8, 600i’ div’ érrérpuvov rrotrépot 1<7\u'r5v M561 irpé
riutdvovg Kori diuoticxv é<1>orr7\ioou, fi' rev divnoiv I Ctfiorpdi T8 icon
Tr€Tr7\ou<; Kori pfivsa 01Y()L7‘t68VT(X. Van Thiel signals the parallel in his appa-
ratus, and indeed emphasizes it by means of an exclamation mark, but fails
to act on it. ’
278 II. Notes on Individual Passages
12° And so already Bergk, PLG‘ (1882) on Theognis 1190: ‘manifesta est correctio’.
*2‘ So also Rzach; Schulze (1892), 421. Bergk loc. cit. had already conjectured oiivnot)’.
122 This is not in fact among the attested forms of the verb, for which see W. Veitch, Greek
Verbs Irregiiiar and Defective (4th ed., Oxford 1879), 492; Kiihner-Blass II 502.
II. Notes on Individual Passages 279
Q 570. KOLI unshortened in hiatus is very rare and usually suspect. Ano-
ther example is O 290 izppiioaro Kai éooiwosv, cf. X 372 épuoafro Kai
éooitoocv. Bekker conjectured fi5€ in both places, and this can be suppor-
ted by reference to K 44 épiiooerat first oat-3051, K 286 1<a1<<I>v éiduioouai
fi5i: oat6ow.‘2" Similarly, at Q 641 oitov Traooiunv Kai a’i6o1ra oivov
Brandreth’s fi5’ a’i00Tra can be supported from 11 295 ij pol o‘i1'ov i=§5t01<z-3v
é'i7ug fi5’ a’i00rra o"1vov.‘25 What about Kai ixérnv here? 0135’ iitérnv would
do just as well while avoiding the metrical fault.
Q 616. We are being told something that is special about Sipylos: it is
where certain nymphs have their suvai. These must be a specific company
of local nymphs, and ’A)(e7\- must be a definite, identifiable river or spring
in the region. If the vulgate ’A)(s7\u5'iov is right, then, it is not to be taken
as a generic word for river water, but as the name of a particular water-
course on Mt. Sipylos. Rivers of this name are known from various parts
of Greece and from the Troad, though not from Lydia. Of course there
might have been one in Lydia too. On the other hand, there is positive evi-
dence for an Acheles or Akeles in precisely the right area. The scholia refer
to it — ’A;(é7tng yap rroraubc dab Zrrr67\ou béet sic; rfiv Zuvpvaiwv yfiv,
A — and quote Panyassis (fr. 17 Matthews) for its nymphs, v15u(1>a1
’A)(z-;7\r‘|T15eg. They record a variant reading ’A)(t-;7trTiov (A 1' schD) or
’A)(t;7\r’i 010v Hellanicus and others knew a Lydian polis Akeles, named
from a son of Heracles and Malis, a slave of Omphale. Its ethnic, accord-
ing to Herodian, was ’A1<e7\fio1o<;.‘26 In schT the Acheles is named after a
son of Heracles and Omphale herself. Evidently Acheles and Akeles are
one and the same.
It seems likely that our poet took the local name and extended it into
—fio1og on the basis of the ethnic; cf. his river Tttaprjotog at B 751. Some
later rhapsodes or scribes, being unfamiliar with the name, assimilated it
to the well known ’A)(e7tt6'iot_',. Probably the original text was ’A1<t-flnjoitov,
as ’A)(t-;7\r'|— occurs only in the scholia on this line, where ’A)(s7\t6- had alrea-
dy established itself as the common reading. The divergence between
—rjo1ov (T) and -rjiov (AD) is to be resolved in favour of the former, as the
latter is again likely to reflect the influence of the -oiiov variant.
Q 662-3. The point of these two lines should be, ‘you must allow us
time to gather wood from the mountains for the pyre’. In 778ff. the Tro-
jans do indeed spend nine days gathering wood. But in 664 Priam’s pro-
posal is to spend the nine days lamenting Hector évi uevoipottg — hardly the
way to describe a period of constant expeditions to Mt. Ida. It is 664-7,
évvfiuap Tfit 5%; 5vto5e1<0’iTr|1 that give the direct answer to Achilles’
question Trooofipap ...; 662-3 are, I believe, genuine lines, but not part of
the passage as originally composed. They were inserted later by the poet,
after he had related the domestic laments for Hector (723—76) and found
himself with nineidays still to fill. He decided to use again the theme of
woodcutting expeditions which he had used in the preparations for
Patroclus’ funeral (\P 114ff.), and to make them occupy the nine days. He
then made the adjustment in Priam’s programmatic speech (660ff.), but
without covering his traces.
Q 667. See above on K 451.
Q 724. Editors generally prefer av5po<i>6vo1o, probably because it is the
reading of the medieval vulgate and thus traditional. There is little in the
argument that it is ‘surely better, in view of what [Andromache] will say
of Hektor’s prowess at 736-9’ (Richardson). On the other side, there are
now three papyri supporting ‘u'rTro50iuo1o, as well as the early parody in
Matron’s ’Arru<5v 5e'irrvov, Supp. Heii. 534. 31, 1'piv7\nc;’11rrro50i|.1o1o KOI[)T|
para )(epo’tv Exovta. This weight of ancient attestation leads me to prefer
‘rrt:rr050iu01o.
Q 736. It is remarkable that both of the available papyri here should side
with that part of the medieval tradition which has the unmetrical reading
Xwdpevocg, (I51 Ttvt 5fi JTOU. It must have its origin in a misreading of
Xtoépevog as a dactyl; in p14 it has in fact been altered into Xwuevog.
Q 759. dtvavoioi [3é7\e0o1 or (itvavoig fiekéeoot? The evidence of the tra-
dition in such cases is that the first pattern is preferred, the long dative of
the adjective and the ‘correct’ form of the es-stern dative, -so-oi. Phrases
with énéeoot, however, form an exception to the general rule, so that we
find Ll81M)(IO1C_; éitésoot etc. rather than -otoi(v) Errcoot (which some older
scholars introduced).'27 The reason is evidently that after the disappearance
of the digamma in ii‘-0101 Férrcooi, rhapsodes chose to avoid hiatus by
recourse to the Aeolic secondary form éité-soot rather than by recourse to
the Ionic movable nu.
Q 763-6. Helen’s extraordinary statement that it is now the twentieth
year since she left Sparta has caused suspicion to fall on lines 765-6, as it
seemed to presuppose the story in the Cyprian that the Achaeans had lost
ten years before their arrival at Troy through having invaded Mysia by
1
“'8 F. G. Welcker, Der epische Cyclas II (Bonn 1849), 265. Cf. Leaf, who writes that the story
‘is as old as the Kypria, at least in the germ, for the Chrestomathy of Proklos tells of the
abortive expedition, but not of the ten years lost’. The ten years are specified in Apollod. Epit.
3. 18, which is a parallel source to Proclus. It remains a real possibility, however, that the year-
count did not feature in the Cyclic poem but was supplied by later mythography precisely in
order to account for Helen’s ‘twenty years’.
‘Z9 K. Reinhardt, Die Ilia: und ihr Dichter (Giittingen 1961), 488-90.
I” Ibid., 487. It has often been suspected that the figure of twenty years results from
mechanical borrowing of the lines 1' 222-3. Indeed the passages cannot be independent of one
another, and it is in the Odyssey that the twenty-year absence from home belongs.
II. Notes on Individual Passages 283
U1 Cf. also the Tabulae Iliacae and other artistic evidence summarized by E. Bethe, Homer.
Dichtzmg and Sage (Leipzig-Berlin 21929), II 167 = Der zroisclae Epen/ereis (Darmstadt 1966),
19.
284 II. Notes on Individual Passages
bridge passages linking the Iliad and Aethiopis. The two fragments must
be regarded as variants of a single text. But how are their divergent second
lines to be reconciled?
First, it seems highly probable that Cronert was right in emending
otpnplnl in the papyrus to ’O1'pfipn<;, since ‘daughter’ calls for a genitive,
and Otrere is well attested as the name of Penthesileia’s mothenm The
only way of avoiding the emendation would be to combine our two sour-
ces in a hypothetical original such as
fi7\6e 5’ ’Au0tCt6v
”Apnoc <1<portepo13> p.l6YOt7\.I‘|TOp0Q dv5poq>6vo1o A
orpnpfi Bvvciifnpi eficioficg lI6V960i7\610t.
The assumption would be that the writer of the papyrus simply omitted
the second of the three lines, and that the source of the scholium omitted
the third, while taking from it the essential word Guycitnp and putting this
in the place of icporrepoii (or whatever stood after ”Apnog). On this hypo-
thesis Penthesileia’s mother would not have been named; possibly the
epithet 6'rpr|pr‘| might have given rise to Otrere as the mother’s name in
other sources. But the name is so well established in the mythographic tra-
dition that we must assume it to have been used in the Aet/aiopis, which
necessarily preceded the composition of these bridge verses. The author of
the verses then could not have been unfamiliar with it; all the more likely
that otpnplnl does represent the name.
It remains a likely assumption that a three—line original stands behind
the two—line quotations in the scholium and the papyrus, as the two
second lines have nothing in common but the word Buvoitnp and can
scarcely be variants of one another. Penthesileia herself must have been
named, and probably both of her parents. We could simply juxtapose the
concurrent versions, as above, but reading ’O'rpfipnc and finding some
replacement for the second Bvvoitnpz
1’"]7t66 5’ ’Au0tCt6v
”Apn0c; Bvvdrmp ueyakfiropog civ5po<}>6vo1o
<?3'in¢; r’> ’Otpfipnc, eiietofig Hsvfisoiaeia.
i_
132 Apollod. l.c.; Lyc. 997 with SCl'1./P3I‘3Pl‘1l‘.; I-Iyg. Fab. 30. 10, 112. 4; sch“ F 189; Serv.
auct. Aen. 1. 491; Tzetz. Postbom. 8, 57, 127. Otrera is listed in the catalogue of Amazons at
Hyg. Fab. 163, and in Fab. 223. 1, 225. 2 she is identified as a consort of Ares and as the ori-
ginal builder of the temple of Diana at Ephesus. Cf. also Ap. Rhod. 2. 387; sch Ap. Rhod. 2.
1031:-1.
II. Notes on Individual Passages 285
ALLEN, T. W., Homer. The Origins and the Transmission, Oxford 1924.
~—, Horneri Ilias, Oxford 1931. References by page are to the first volume (Prolego-
mena).
APTHORP, M. ]., The Manuscript Evidence for Interpolation in Homer, Heidelberg
1980.
BECHTEL, P., Lexilogas zn Homer, Halle 1914.
BEKKER, I., Homerische Blatter, I, Bonn 1863; II, Bonn 1872.
BERGK, T., Griechische Literatargeschichte I, Berlin 1872.
BETHE, E., Homer, Dichtang and Sage I—-III, Berlin-~Leipzig 1914-27.
BOLLING, G. M., The External Evidence for Interpolation in Homer, Oxford 1925.
-, Ilias Atheniensiam, Baltimore 1950.
BURKERT, W., ‘Die Leistung eines Kreophylos’, Mas. Help. 29 (1972), 74-85.
-—, ‘Kynaithos, Polycrates, and the Homeric Hymn to Apollo’, in G. W. Bowersock
et al. (ed.), Arktoaros. Hellenic Studies presented to Bernard M. W. Knox, Ber-
lin—-New York 1979, 53——62.
-—, ‘The Making of Homer in the Sixth Century B.C.: Rhapsodes versus Stesichoros’,
in Papers on the Amasis Painter and his World, Malibu 1987, 43-62.
—, Kleine Schriflen I: Homerica, Giittingen 2001.
CAUER, P., Grandfragen der I-Iomerkritik, 3. Aufl., Leipzig 1921-~3.
CHANDLER, H. W., A Practical Introduction to Greek Accentaation, 2nd ed., Oxford
I881.
CHANTRAINE, P., Grammaire homériqae I, Paris 1942; II, Paris 1953.
—-, Dictionnaire étymologiqne de la langne grecqae, Paris 1968-80.
COBET, C. G., Miscellanea critica, Leiden 1876.
CRESPO, E., Elementos antigaos y modernos en la prosodia homérica (Minos Suppl. 7),
Salamanca 1977.
DENNISTON, ]. D., The Greek Particles, 2nd ed., Oxford 1954.
DUNTZER, H., De Zenodoti stadiis Homericis, Gottingae 1848.
-, Homerische Ahhandlnngen, Leipzig 1872.
DURANTE, M., Salla preistoria della tradizione poetica greca I, Roma 1971; II, Roma 1976.
EDWARDS, M. W., The Iliad: A Commentary, V: Books 17-20, Cambridge 1991.
ERBSE, H., ‘Uber Aristarchs Iliasausgaben’, Hermes 87 (1959), 275-303.
-—-, Scholia Graeca in Homeri Iliadern I—VII, Berlin 1969—88.
FICK, A., Die homerische Odyssee in der arspriinglichen sprachform wiederhergestellt,
Gtittingen 1883.
—, Die homerische Ilias nach ihrer entstehz/mg hetrachtet and in der nrspriinglichen
sprachform twederhergestellt, Gottingen 1886.
FRIIS ]OHANSEN, K., The Iliad in Early Greek Art, Copenhagen 1967.
GERSTINGER, H., Archie fiir Bihliographie 1 (1926).
HAINSWORTH, (1.) B., The Iliad: A Commentary, III: Books 9—12, Cambridge 1993.
HASLAM, M. W., ‘Homeric Papyri and the Transmission of the Text’, in I. MORRIS
and B. POWELL (ed.), A New Companion to Homer, Leiden-—New York-Koln 1997,
55--100.
HOEKSTRA, A., Homeric Modifications of Formalaic Prototypes, Amsterdam 1965.
Bibliography 287
IOANNIDOU, G., Catalogue of the Greek and Latin Literary Papyri in Berlin, Berlin
1996.
JANKO, R., Homer, Hesiod, and the Hymns, Cambridge 1982.
-, The Iliad: A Commentary, IV: Books 13:-16, Cambridge 1992.
KENYON, F. G., Classical Texts from Papyri in the British Maseam, London 1891.
KIRK, G. S., The Iliad: A Commentary, I: Books 1-4, Cambridge 1985.
-, The Iliad: A Commentary, II: Books 5-8, Cambridge 1990.
LAC!-IMANN, K., Betrachtangen zar Ilias, Berlin’ 1847.
LAMEERE, W., Apercas de paléographie homériqae, Brussels 1960.
LA ROCHE, ]., Die homerische Textkritik im Altertham, Leipzig 1866.
-, Homerische Untersuchangen I~—II, Leipzig 1869, 1893.
LEAF, W., The Iliad, edited with apparatus’ criticus, prolegomena, notes, and appen-
dices, 2nd ed., London 1900-2.
LEHRS, K., De Aristarchi stadiis homericis, ed. tertia, Lipsiae 1882.
LEUMANN, M., Homerische Worter, Basel 1950.
LUDWICH, A., Aristarchs homerische Textkritik nach den Fragmenten des Didymos
dargestellt and heartheilt, Leipzig 1884~—5.
MAEHLER, H., ‘Fragmente antiker Homer-Handschriften aus Agypten’, in Festschrzft
zam 1S0ja'hrigen Bestehen des Berliner /-Igyptischen Museums (Mitteilangen ans der
iigyptischen Sammlang, 8), Berlin 1975, 363~—94.
MANFREDI, M. (and others), Papiri dell’Iliade, Firenze 2000.
MEISTER, K., Die homerische Kanstsprache, Leipzig 1921.
MERKELBACH, R., ‘Die pisistratische Redaktion der homerischen Gedichte’, Rheini—
sches Museum 95 (1952), 23—47 = Philologica. Aasgewahlte Kleine Schriften (Stutt-
gart—Leipzig 1997), 1—~23.
MONRO, D. B., A Grammar of the Homeric Dialect, 2nd ed., Oxford 1891.
—, Homer’s Odyssey, Books XIII~XXI V, Oxford 1901.
MONTANARI, P., Stadi di filologia omerica antica I, Pisa 1979; II, Pisa 1995.
— ‘Zenodotus, Aristarchus and the Ekdosis of Homer’, in G. W. Most (ed.), Editing
Texts / Texte edieren (Aporemata 2), Gottingen 1998.
NABER, S. A., Qaaestiones Homericae, Amstelodami 1877.
NAGY, G., review of M. L. WEST, Homeri Ilias (volumen prius), Bryn Mater Classi-
cal Review 2000-09~12. -
NAUCK, A., Aristophanis Byzantii grammatici Alexandrinifragmenta, Halis Saxonum
1848.
NICKAU, K., ‘Zenodotos von Ephesos’, RE Xa (1972), 23—45.
-, Untersachangen zar textkritischen Methode des Zenodotos von Ephesos,
Berlin——New York 1977.
NITZSCH, G. W., Beitrage zar Geschichte der epischen Poesie der Griechen, Leipzig
1862.
PALMER, L. R., ‘The Language of Homer’, in A. ].’B. WACE and F. H. Sruesnvos (ed.),
A Companion to Homer, London 1962, 75-178.
PASQUALI, G., Storia della tradizione, Firenze 1952.
PFEIFFER, R., History of Classical Scholarship. (I) From the Beginnings to the End of
the Hellenistic Age, Oxford 1968.
POWELL, B. B., Homer and the Origin of the Greek Alphahet, Cambridge 1991.
288 Bibliography
PRIEST, N. E., Homeric Papyri in the Michigan Collection, Diss. Michigan 1975.
RENGAKOS, A., Der Homertext und die hellenistischen Dichter (Hermes Einzelschrifl
ten, 64), Stuttgart 1993. ‘-
RICHARDSON, N. ]., The Iliad: A Commentary, VI: Books 21--24, Cambridge 1993.
RISCH, E., Worthildung der homerischen Sprache, 2. Aufl., Berlin-New York 1974.
SCHADEWALDT, W., Iliasstudien, Leipzig 1938 (Abh. Sachs. Akad. 43(6)).
SCHMIDT, M., Didymi Chalcenteri Grammatici Alexandrini Fragmenta, Lipsiae 1854.
SCHULZE, W., Quaestiones epicae, Giitersloh 1892.
-, Kleine Schriflen, Giittingen 1933.
SCHWARTZ, E., Adversaria, Progr. Gottingen 1908.
-, in ’Av'ri6topov, Festschrift Wackernagel, Gottingen 1923. -
SCHWENDNER, G. W., Literary and non~literary Papyri from the University of Michi-
gan Collection, Diss. Michigan 1988.
SCHWYZER, E., Griechische Grammatik, 1. Band, Miinchen 1939; 2. Band, vervoll-
standigt und herausgegeben von A. DEBRUNNER, Miinchen 1950.
SEIDER, R., Paliiographie der griechischen Papyri I-II, Stuttgart 1968-70.
THEODORIDIS, CI-IR., Die Fragmente des Grammatikers Philoxenos, Berlin-New York
1976.
THREATTE, L., The Grammar of Attic Inscriptions I-II, Berlin-New York 1980-96.
THUMB, A., Handhuch der griechischen Dialekte, 2. Teil, 2. erweiterte Auflage von
A. SCHERER, Heidelberg 1959.
TURNER, E. G., Greek Manuscripts of the Ancient World, Oxford 1971; 2nd ed. re» l
vised by P. ]. Parsons, 1980.
VAN DER VALK, M., Researches on the Text and Scholia of the Iliad, Leiden 1963-4.
VAN THIEL, H., ‘Zenodot, Aristarch und andere’, ZPE 90 (1992), 1-32.
—, ‘Der Homertext in Alexandria’, ZPE 115 (1997), 13-36.
VILLOISON, ]. B. G. D’ANSSE DE, Homeri Ilias ad veteris codicis Veneti fidem recensi-
ta, Venetiis 1788. '
VON DER MUHLL, P., Kritisches Hypomnema zur Ilias, Basel 1952.
WACKERNAGEL, ]., Sprachliche Untersuchungen zu Homer, Gfiittingen 1916.
—, Vorlesungen iiher Syntax I-II, Basel 1920-4; 2. Aufl. 1926-8.
-, Kleine Schriften I-II, Gtittingen 1955; III, Gtittingen 1979.
WATHELET, P., Les traits éoliens dans la langue de l’ép0pée grecque, Roma 1970.
WEST, M. L., Hesiod. Theogony, Oxford 1966.
-, Textual Criticism and Editorial Technique, Stuttgart 1973. ’
-, Studies in Greek Elegy and Iamhus, Berlin-New York 1974.
—, Hesiod. Works and Days, Oxford 1978.
-, Greek Metre, Oxford 1982.
-, The Hesiodic Catalogue of Women, Oxford 1985.
—, Ancient Greek Music, Oxford 1992.
--, ‘The Date of the Iliad’, Museum Helveticum 52 (1995), 203-19.
-, The East Face of Helicon, Oxford 1997. "
-, ‘The Textual Criticism and Editing of Homer’, in G. W. Most (ed.), Editing
Texts/ Texte edieren (Aporemata 2), Gottingen 1998, 94-110.
-, Homeri Ilias. Volumen prius: Rhapsodiae I-XII. Volumen alterum: Rhapsodiae
XIII-XXIV, Stutgardiae--Lipsiae-Monachii 1998-2000.
Bibliography 289
-—, ‘Frfihe Interpolationen in der Ilias’, Nachrichten der Aleademie der Wissenschafteu
in Géittingen (Phil.-hist. KL), 1999(4), 183-91.
--, ‘The Invention of Homer’, Classical Quarterly 49 (1999), 364--82.
-, ‘Some Homeric Words’, forthcoming in Glotta 76 (2001).
WEST, S. R., The Ptolemaic Papyri of Homer (Papyrologica Col0m'ens:'a, 3),
K6111-—Opladen 1967.
——, ‘Chalcenteric Negligence’, Classical Quarterly 20 (1970), 288-96.
-—, in A. Heubeck, S. West, ]. B. Hainsworth, A Commentary on Homer’s Odyssey, I:
Introduction and Books I-VIII, Oxford 1988.
WILAMOWITZ-MOELLENDORFF, U. VON, Homerische Untersuchungen, Berlin 1884.
-, Die Ilias und Homer, Berlin 1916.
WILSON, N. G., Scholars of Byzantium, London 1983.
WOLF, F. A., Prolegomena to Homer (1795), translated with introduction and notes
by A. Grafton, G. W. Most, and ]. E. G. Zetzel, Princeton 1985.
-, Vorlesungen ulher die vier ersten Gesange von Homers Ilias I-~II, hrsg. L. Usteri,
Bern 1830--1.
WYATT, W. F., Metrical Lengthening in Homer, Roma 1969.
INDEXES
1. GREEK WORDS
2. PASSAGES DISCUSSED
ILIAD AND ODYSSEY
A 3-5: 45, 173 B 491-2: 12 n. 28, 177-8
A 11: 173 B 516: 178
A 33, 34, 37, 40, 42: 87 n. 9 B 525-6: 12 n. 25, 178-9
A 47: 26 B 527: 179
A 48: 87 n. 9 B 535: 12 n. 27, 179
A 64: 162 B 546-58: 12 n. 25, 179-80
A 66--7: 22 B 557-8: 14 n. 32, 179-80
A 69: 173-4 B 572: 12 n. 25,181
A 97: 261 B 585: 181
A 170: 174 B 590: 176
A 185, 207: 174 B 674: 12 n. 28, 41 n. 35
A 244: 87 n. 9 B 703, 708-9: 13 n. 31, 181-2
A 344; 182, 174 B 742-4: 12 n. 27
A 265: 186 B 749, 765, 7972 182
A 428: 169 n. 19 B 811: 183
A 510: 87 n. 9 B 849: 87 11. 9
A 531-3: 174 B 855ah: 140
A 609-11: 9 B 865: 183
B 4: 45 F 10: 23
B 12: 87 n. 9 T 13: 184-5
B 15: 24-5, 87 n. 9, 175 F 18-20: 12 n. 28
B 60-70: 41 F 41: 12 n. 29,185
B 111-18: 26 F 51: 270 n. 112
B 111: 175 F 78: 13 n. 31
B 144: 45 I‘ 100: 197-8
B 168: 13 n. 31 F 144: 12 n. 25, 185-6
B 206: 13 n. 31, 175 F 147: 186
B 216: 22 F 162, 165, 168, 170, 174: 87 n 9
B 232-3: 22 F 188; 186-7
B 258: 175 r 212, 316; 187
B 264: 164-5 4 1" 354, 355; 87 11. 9
B 291, 356: 176 r 389; 13 11. so
B 399, 435, 448, 466: 177 r 417; 87 11. 9
294 Indexes
HOMERIC SCI-IOLIA
A 3: 76 I 159: 74
A 97:55 I212: 49
A 381: 70 2221:): 37 n. 19
A 423-4: 54, 59, 70-1 383: 49
A 559: 174 401b: 74
-4:.
I 416: 60 n. 43
B 128: 61 653: 49
.,-_
T 26a: 55 n. 23 ‘P 824-5: 60 n. 43
T 81-2: 74 ‘P 870-1a‘: 71
T 365-8: 64
Q 6--9a: 60 n. 43
<I> 126-7c: 76 Q 71-3: 53-4
(I) 130-5a‘: 64 Q 110b': 80
(D 363a: 74 Q 614-17a: 60 n. 43
OTHER AUTHORS
Aeschylus Eudocia
Suppl. 800-1; 173 Homerocentones 578-81: 250 n. 85
fr. 266 Radt: 277 Euripides
Alcaeus Erechth. 60 Austin: 20
fr. 401B Voigt: 8 Hec. 1077: 173 n. 1
Alcman Ion 505: 173 n. 1
PMGF 26. 1: 250 Galen
Alexis in Hipp. Epial. 111 2. 4; 68-9
fr. 140 K.-A.: 21 Herodotus
Anacreon 1. 94. 3: 266
PMG 408: 43 n. 41 [Hesiod]
Anecdotum Osanni: 73 Scut. 22-6: 179
Apollonius Rhodius —-- 405: 238-9
1. 358: 200 fr. 204. 44-51 M.—W.: 180
3. 1196: 205 I1. 30 ‘Homerus’
4. 1629-30: 211 fr. 2 Davies: 272
Apollonius Sophista Hymn to Apollo 166-76: 17
119. 222 176 Ibycus
Aristophanes PMGF S151. 23 ff; 178
Eq. 197,204; 238-9 Mimnermus
Athenaeus fr. 14. 1-2 West: 194
1801)-Id: 251--2 Nonnus '1
Stesichorus Xenophon
PMGF 255: 262 Anah. 7. 5. 14: 21 n. 58
Suda Mem. 1. 6. 14: 20
s.v. 1112301100: 48 -- 4.2.1, 8-10: 20
Timotheus Symp. 3. 6: 19
Pers. 138: 173 n. 1
3. (}IiI41iI{l\I5IIhJI)I§)(
abridgment 41 225 n. 54, 232, 234, 253-8, 259, 280,
accentuation 38, 46, 73, 78, 80, 82, 197, 262, 267, 278, 279
223-4, 253, 278, 279 Aristonicus 46, 47, 49-50, 54, 56, 60, 65,
acephalous lines 205 75, 82
Adrastus 181 Aristophanes of Byzantium 34, 36-7, 51,
Aeolic forms 164-5, 221, 224, 262 57-60, 68, 77, 184, 197, 258--60, 267-8
Aethiopis 12 n. 26, 283-4 Aristotle 24, 25, 72
Agathocles of Cyzicus 76 article 173, 208, 209, 259
Aineiadai 7, 256 Asclepiades of Myrlea 75
Ainianes 182 Athena Ilias 7-8
Ajax 180 Athenian interpolations 12, 179-80, 220
Alcibiades 20 Athenocles of Cyzicus 76-7
Alcidarnas 24 Athens 17-23, 31-2
Alexander the Great 25, 72 atheteses 38-9, 60, 64, 202-3, 210, 218,
- of Cotyaea 153, 191, 227 230, 232, 244-5, 255, 258, 262, 265,
Alexandrian library 34-6, 68-9, 72-3 267, 277; critical ripostes 78, 81 n. 116
allegories 75, 76, 137 Atticisms 31-2, 176, 205
Allen, T. W. 88, 158, 181, 167 authorial revisions 3, 199, 212, 218-~19,
alphabets 4, 21-3, 223, 227, 258 238, 256, 265-6, 272, 275, 281
alternative recensions 256 Autochthon 83
Ammonius 56, 62-5, 77, 79--80
Anacreon 18 Babylonian poems, transmission 15 n. 35
Anaximander (the younger) 24 banalization 40, 57
Anecdotum Osanni 73, 139 Bathyllus 48
Antimachus’ text 52-4, 68, 85, 275 book-collecting 20; books dedicated in
antisigma 204 temples 6; book trade 21
Antisthenes 24 ‘book’-division 3, 18--19, 165
Apellicon 73
Apion 48 Callistratus 36, 56-8, 60-1, 69-70, 72, 73,
Apollodorus of Athens 65, 80 78-9, 258
Apollonius of Rhodes 76, 173, 205 Catalogue of Ships, omitted in some
Aretades 80 manuscripts 140-3, 148
Arideikes 83 centos 136-7, 166-7
Aristarchus 34, 36-8, 51-5, 57-67, 69-72 Chaeris 77, 79, 81
74-5, 76, 79, 81-3, 144, 146, 160, 167, Chamaeleon 75
185, 197, 200, 202-4, 207, 210, 218, Choeroboscus 223
302 Indexes
dual forms used for plural 41-2, 77 220-1, 225-7, 230-4, 236-7, 243-5,
Diintzer, H. 34--5 248, 256, 258, 262-4, 266, 271, 275, I
277,279,282-3
1
Edda 7 11- 3 Ioanniltios 142, 154
‘editions’ (61<560e10) 50-2, 56-73, 83-4 Ionian textual tradition 32, 43-5, 68, 235
emendation, pre-Alexandrian 26-8, 42; iota, 1035 of in long diphthongs 30; in
by Zenodotus? 42-3 correption 234-5
Epapllmdittls 75 Ixion, see Demetrius
- -._, »—-
I
1
Indexes 303
l
Kreophylos, Kreophyleioi 16 Pisistratus 9, 17
Pius 75, 83
lacunas? 196, 273, 283 Plato’s quotations 80
Lehrs, K. 46 plus—verses 11-14, 33, 40, 173-4, 201-2,
J
F
Lesbian interest in.Troad 7-8 214,227-8
literacy 20 political interpolation 12-14, 178-9
Ludwich, A. 36, 158, 160-1, 165-7 Polycrates 16, 18
Lycurgus 9 Posidonius 0 d1v01vv0501"n0 ’Ap10t61p)(00
1 Mahabharata 11
55
Prodicus 25 n. 71
Menestheus 180 Protagoras 24, 25 n. 71
metrical faults 162, 208, 211 (bis); cf hiatus psflosb 32,44,282
Metrodorus 24-5 Ptolemy of Ascalon 82, 210
1 modernization 28-31, 221 - Epithetes 56, 78, 225 n. 54
Mythographus Homericus 130-1, 133-5, -- Pindarion 81-2
137 - son of Aristonicus 47
punctuation 4
Nagy, G. 159 I1. 2, 160 11. 5
Nemesion’s Tetralogy 65 qoppa 4, 29
neo-Ionic forms 32, 34, 43-4 quotations 165-6; in papyrus texts 136-8;
Neoptolemus 12 Plato s quotations 80
Neoteles 80
‘Nestor’s cup’ 3-4, 210, 246
Ramayana 11
-._>N-*1“--
Nicanor 46, 73
recitations, partial 8-9
Nicias 83
rhapsodes 5-6, 8-10, 15, 18-20, 25-6,
Nonnus, transmission of 191
40-5, 231
Norse texts 7 n. 8
rhetorical expansion 12-13, 204, 254
nu movable 163, 188
l Rhianus 56-8, 68
ring composition 198, 226
‘oral’ variants 14-15, 40, 57, 159 n. 2
Roland, Song of 11
orality and writing 3
orthography 4, 21-3, 29-32, 154, 164
Otranto 143 Sacred War, First 12
l
scholia minora 129-36; D scholia 139,
Panathenaea 8, 9, 17-19, 68 140, 143; h scholia 145 n. 10
papyri14,86-138,168 schooling 19
Parmeniscus 81 schoolmasters 20-1
parodies 166 scriptio plena 38, 249, 254
particles inserted 162-3, 192, 200, 203 Seleucus 47-9, 60, 70, 72, 75, 82, 208,
patrons 5-6 251--2
Pergamene scholarship 9, 11, 58, 63, 75 Shah-Nameh 11
phantom substituted for person 193 Sicyon 12, 181
Pheia 12, 200 simplification of word order 255
Philemon 58-9 sophists 24-6, 43 n. 41, 75
Philitas 76 Sosigenes 58, 68
304 Indexes