L'antiquité classique

Vergil's Circe : Source for a Sorceress

Veerle Stoffelen

Résumé
La sorcière Circé, connue surtout par l'Odyssée, apparaît trois fois au chant VII de l'Énéide. La première fois, Virgile emprunte
de nombreux traits à Homère, mais laisse Énée dépasser l'île de Circé. Le héros adresse ainsi un dernier adieu à son passé
mythologique avant de commencer sa tâche historique en Italie. Dans les épisodes suivants, la sorcière devient une figure de
plus en plus virgilienne. Au total, elle est un produit de la liberté poétique et du choix personnel de l'auteur.

Citer ce document / Cite this document :

Stoffelen Veerle. Vergil's Circe : Source for a Sorceress. In: L'antiquité classique, Tome 63, 1994. pp. 121-135;

doi : https://doi.org/10.3406/antiq.1994.1186

https://www.persee.fr/doc/antiq_0770-2817_1994_num_63_1_1186

Fichier pdf généré le 11/09/2019

 Il

Vergil's Circe : Sources for a Sorceress1

The name of Circe is best known from Homer' Odyssey : in the

tenth and the twelfth book (Od., X, 135-574; XII, 1-135) the author
tells us how his hero stays on the sorceress' island for a long, long time
until his friends remind him of his duty to return home. Less well
remembered though is that Vergil mentions Circe three times in the
course of his Roman epic, the Aeneid. She appears a first time when
Aeneas sails by her island between his departure from Cumae and his
landing at Latium, the place where he finally finishes his wanderings
(Aen., VII, 10-24). A little bit further in the same book (v. 187-191) the
sorceress is mentioned as the one who changed the old king Picus into a
bird, and she appears a third time in the same book as the one who gave
king Latinus the very special horses he now offers to Aeneas (v.
280-283). Both the second and the third passage belong to the section in
which Vergil tells about the embassy Aeneas sends to lang Latinus.
A closer look at these passages will teach us that Homer's Circe
was only a starting-point for Vergil. In the first passage his Circe is still
largely the same character she was when Odysseus met her, though
Vergil deliberately changes some details in her portrayal. For neither die
second nor the third passage we can find direct parallels in Homer; in his
epic the sorceress was not involved in this kind of activity. From that
point on Circe starts to live a more Vergilian Ufe2.
Among the secondary literature devoted to Circe, two articles need
to be mentioned in connection with Vergil's use of this figure : Charles
SEGAL, Circean Temptations : Homer, Vergil, Ovid, in T.A.Ph.A., 99
(1968), p. 419-442 and Emmanuel S. HATZANTONIS, Le amare fortune
di Circe nella letteratura latina, in Latomus, 30 (1971), p. 3-22. Segal
starts with a profound analysis of the Circe-figure in the Odyssey,
before discussing Vergil's and Ovid's adaptation of the Homeric
material. He pays much attention to the first passage (Aen., VII, 10-24),
but hardly discusses Circe's further appearances in the Aeneid.
Hatzantonis briefly compares the Vergilian sorceress in Aen:, VU, 10-24

1 At the beginning of this article I would like to thank Prof. H.C. Gotoff of the
University of Cincinnati and Prof. Dr. W. Evenepoel from the Katholieke Universiteit
Leuven for their readiness to correct earlier drafts of this paper and for their useful advice.
Of course they should not be held responsible for any errors or any views expressed in this
article.
2 HATZANTONIS (Le amare fortune di Circe nella letteratura latina, 1971, p. 11)
distinguishes «la Circe modellata sulla maga di Omero» and «una Circe nuova».
122 V. STOFFELEN

with her Homeric predecessor and looks for possible sources for the
sorceress of the next two passages (Aen., VII, 189-191, 280-283), but
does not really explain the reasons for Vergil's specific treatment of the
Circe-figure.
This article will concentrate on three questions concerning Vergil's
use of the Circe-figure : «What were his sources for the three passages
in which she appears? How does he use them? And why does he use
them in this particular way?»

PASSAGE I : Aen., VII, 10-243.

10 próxima Circaeae raduntur litora terrae,

dives inaccessos ubi Solis filia lucos
adsiduo resonat cantu, tectisque superbis
urit odoratam nocturna in lumina cedrum
arguto tenuis percurrens pectine telas.
15 hinc exaudiri gemitus iraeque leonum
vincla recusantum et sera sub node rudentum,
saetigeri sues atque in praesepibus ursi
saevire acformae magnorum ululare luporum,
quos hominum ex facie dea saeva potentibus herbis
20 induerat Circe in vultus ac tergaferarum.
quae ne monstra pii paterentur talia Troes
delati inportus neu litora dira subirent
Neptunus ventis implevit vela secundis,
atque fugam dedit et praeter vada férvida vexit.

This first passage about Circe contains three main parts : a) Circe
and her activities (v. 10-14), b) a description of her animals (v. 15-20),
and c) a more narrative part in which Vergil tells his audience how
Neptune helps the Trojans to sail by the island of the sorceress (v.
21-24).
When the ancient Roman audience heard or read this passage, their
thoughts must have turned immediately towards Homer's Odyssey and
the hero's visit to the sorceress' island. Therefore a closer look at
Homer's Circe might help us learn more about Vergil's use of this
source. In the Homeric Circe-story, descriptive and narrative elements
are interwoven, so that it is more a matter-of-fact account than the
Vergilian, which contains almost only descriptive features. In Homer,
Circe was a character that could talk and act, while Vergil's Circe never

3 This and the following passages are cited frpm Mynors' Vergil-edition (P. Vergili
Moronis opera recognovit brevique adnotatione critica instruxit R.A.B. Mynors, Oxford,
1986).
 VERGIL'S CIRCE : SOURCES FOR A SORCERESS 123

becomes a real character4, but remains a part of the scenery against

which the author develops the story of Aeneas' wanderings. With the
elements he borrows, the details he adds, and some poetic techniques
Vergil creates a very condensed, descriptive vignette, which is
characterized by its highly ornate style and evocative quality. Let us
illustrate this with a few examples5.
Vergil refers to Circe's dwelling-place as the Circaea terra (v. 10),
while Homer always calls her place an island (Cf. Od., Χ, 135; XI, 70;
ΧΠ, 3). In the prophecy of Helenus Vergil too describes her place as an
island {Aeaeaeque ínsula Circae, Aen., Ill, 386) but in the catalogue of
the Italian troops he speaks again about the Circaeumque iugum {Aen.,
VII, 799)6.
The sorceress herself is called the daughter of the Sun (v. 11).
Circe's descent from the Sun must have been an old and very common
element in her legend7. Homer says that both she and her brother Aietes
are children of the Sun {Od., X, 138) and Hesiod calls her the Ήελίου
θυγάτη ρ Ύπε ρ ιον ίδαο {Theo g ., 1011).
The woods which the Vergilian Circe fills with her songs were
already there when Odysseus arrived on her island : he and his
companions see smoke rising up from her house δια δρυμα πυκνά και
ϋλην {Od., Χ, 150, 197) and her island is ύληεσσαν {Od., X, 308). In
the Aeneid though the woods on her island are inaccessus; this adjective,
a neologism here, remains rather unusual throughout Latin literature;
Vergil himself uses it only once more in the description of Cacus' cave
{Aen., VIII, 195). This little detail helps Vergil to create a special
atmosphere : Circe's home is not a nice, gentle place where everyone is
welcome, it is a lucus inaccessus.
The next three lines are dedicated to a description of Circe's
activities : her singing, the burning of fragrant cedar and her weaving.
Homer mentions a few times that the sorceress is singing with a

4 Circe can never become the character she was in Homer's Odyssey because the hero
never meets her, never talks to her; or, if we exchange our prospective for the author's, we
see that it works just the other way around : because Vergil chooses not to let his hero
visit her, talk to her, she can never become the character she was in the Odyssey.
5 For a short survey of the parallels and differences between Homer's and Vergil's
Circe, cf. Hatzantonis, Le amare fortune di Circe nella letter atura latina, 1971, p. 9-11.
6 Many commentators have tried to find out whether Circe's place was an island or
not. Servius reflects ancient attention to this problem (Ad Aen., VII, 10). Quoting Varro he
supposes that her place had previously been an island, later on joined to the mainland (Ad
Aen., Ill, 386). Also Pliny the Elder dealt with this problem in his Nat. Hist., Ill, 57
(Cercei, quondam insula immenso quidem mari circumdata, ut creditur Hornero, et nunc
planitie). Cf. Hatzantonis, Le amare fortune di Circe nella letteratura latina, 1971, p.
1; for further references, cf. Franz BÖMER, P. Ovidius Naso. Metamorphosen, Buch XIV-
XV, 1986, p. 90.
7 Cf. Bömer, ad Ovid., Met., XIV, 10.
124 V. STOFFELEN

beautiful voice (cf. Od., Χ, 221, 227, 254)8. Vergil refers to this
singing of hers without real verbal imitation, changing the Homeric
καλός into the more neutral adsiduo. He had referred already to her
songs in his Eclogues {Eel., 8, 70 : carminibus Circe socios mutavit
Ulixi). There the rather unexpected innovation of Circe using incantation
to change Odysseus' companions into animals is particularly applicable
to the poem : Vergil is talking about the power of incantation.
The burning of cedar is not a typically Circean activity, but is
rather an adaptation of Homer's description of Calypso's cave9.
Although Homer tells us that smoke rises up from Circe's house (Od.,
X, 149-150, 196-197) there is no indication of the burning of cedar.
Calypso's island on the other hand smells because of the burnt cedar
(Od., V, 59-61). Vergil uses this idea but changes the verbal
construction :

- δαιομένων is turned into the active verb writ

- όδμή and όδώδει are both replaced by the adjective odoratam.

The fact that Circe burns cedar in the moonlight fits into the setting
Vergil creates (it is night when Aeneas and his friends pass Circe's
place, cf. v.8-9), but contributes also to the magic sphere he evokes.
Line 14 about the weaving is the only line about which Servius
says in his commentary : est autem Homeri, qui ait ιστόν έποιχομένη.
Indeed weaving seems to be one of the most typical activities of Circe;
cf. Od., Χ, 222 :

Ιστόν έποιχομένης μέγαν, αμβροτον10.

This feature adds a touch of peaceful domesticity to a scene with a

fundamentally ominous mood (weaving being a common activity of
many Greek and Roman matrons)11. As for the construction, this line

8 Cf. Od., V, 61 about Calypso's singing.

9 Cf. CONINGTON-NETTLESfflP, ad Aen., VII, 11 : «it is her <Calypso's> cave that is
full of the scent of burning cedar, an incident which Virg. has transferred to Circe». For the
blending of these two female figures, cf. also Georg Ν. Knauer, Die Aeneis und Homer,
Göttingen, 19792, p. 138; Erich Kaiser, Odyssee-Szenen als Topoi, in Museum
Helveticum, 21 (1964), p. 198.
10 Cf. also Od., X, 226, 254.
11 Cf. Eduard Fraenkel, Some aspects of the structure of Aeneid VII, in J.R.S., 35
(1945), p.2 : «Here Circe (...) makes a personal appearance, combining with the
dangerous charms of the Homeric sorceress some of the domestic virtues of a Roman
gentlewoman». One could argue that in both the Odyssey and the Aeneid this domestic
behaviour forms part of the enchantment : the women use apparently innocent weaving
and singing to try to lure the men in the sorceress' trap, successfully in the Odyssey, but -
thanks to Neptune's intervention - without succès in the Aeneid.
 VERGIL'S CIRCE : SOURCES FOR A SORCERESS 125

may be closer to the description of Calypso's weaving : Ιστόν

έποιχομένη χρυσείη κερκίδ' ϋφαινεν (Od., V, 62). Vergil imitated this
line a first time in his Georgics to describe the weaving of a farmer's
wife12 (thus finding his inspiration for a homey weaving scene in a
Homeric episode full of myth and magic) :

arguto coniunx percurrens pectine telas (Georg., I, 294).

By the time he wrote his Aeneid this line must have become one of
the many Latin lines Vergil had in the back of his head and used when
appropriate. In the description of Circe he refers thus to one of his own
lines13, again to describe a female character at the loom, but this time in a
context which is again much closer to the Homeric original.

The section about Circe's animals must have had a challenging

appeal to Vergil's talent, since he obviously enjoyed creating a
masterpiece of poetic word- and sound-play, leaving behind the bare
facts of the Homeric version. Vergil mentions four different animals :
lions (v. 15-16), pigs (v. 17), bears (v. 17) and wolves (v. 18). Homer
only talks about wolves, lions and pigs (Od., X, 212, 218, 239-240,
433). In the Aeneid the lions roar, bellow in the middle of the night
(gemitus, v. 15; rudere, ν. 16); in the Odyssey Circe's animals are more
friendly : they wag their tails when Odysseus' companions reach the
house (Od., X, 214-215). Already Heyne noted this change in the mood
of the description : «Discessit autem ab Hornero, qui mutatas has feras
humanam mansuetudinem servantes exhibuerat»14.
The «ae-sound» in lines 17-19 (and the «s-sound» in v. 17)
creates a special effect. This sound-play may have influenced Vergil's
choice of some words : saetigeri (though it is always used for pigs, cf.
Aen., XI, 198; XII, 170); praesepibus and ursi (while Homer never
mentions any bears in connection with Circe); and formae magnorum
ululare luporum15. Dea saeva must be an adaptation of the Homeric δεινή
θεός (cf. Od., X, 136; XI, 8; XII, 150), but at the same time it
contributes to the atmosphere Vergil creates (Circe is a dea saeva) and
the effect he reaches by the repeated use of the «ae-sound».

12 Cf. Conington-Nettleship, ad Aen., VII, 14.

13 Richard F. Thomas (Vergil's Georgics and the art of reference, in H.S.C.P., 90
[1986], p. 182-185) gives other examples of this technique in Vergil's oeuvre.
14 Ad Aen., VII, 15-18; cf. also Conington-Nettleship, ad Aen., VII, 15; Kenneth J.
Reckford, Latent tragedy in Aeneid VII, 1-285, in AJ .Ph., 82 (1961), p. 255.
15 In the second half of this line the reader's ear is struck by the intended repetition
of the «u» and the «o»-sound {magnor- ululare luporum). The combination of these two
vowels underlines the ominous mood of the scene.
126 V. STOFFELEN

The last two lines about her animals (v. 19-20) call our special
attention because of their structure :

quos hominum ex facie . . .

...in vultus ac terga ferarum

forms a chiasmus16 with a variatio in it (facie ++ vultus ac terga). This

figure of speech puts hominum and ferarum in strong contrast (and
represents in this way the metamorphosis of the homines into ferae).
This chiasmus surrounds the middle part of the sentence, which
contains the main verb (induerat), preceded by the potentibus herbis, the
herbs Circe uses for this act17; this central part is surrounded by the
subject, which is split into two parts : dea saeva and Circe.
From line 21 on, Vergil's style becomes more narrative; he tells
how Neptune sends favorable winds to the Trojans, so that they can
avoid Circe and her animals18 and arrive, finally, in Latium. Neptune
was also helpful towards the Trojans in book I after the storm raised by
Aeolus on Juno's request, and in book V, when he promises Venus to
lead them safely to the harbour of the Avernus19. His help is thus
consistent with his former attitude. But the intervention of Neptune here
is an entirely Vergilian creation : Homer did not need any divine help to
let Odysseus and his companions pass Circe's island, because he let
them stay there for quite a while. For Vergil the god is a convenient deus
ex machina, for unlike Homer, he will not let his hero get involved in
magic phantasies (cf. infra).
These examples show how Vergil borrowed the basic features of
Homer's Circe-description, but adapted them to the version he himself
wanted to write. Segal supposes that Vergil had a precedent for this
(and negative) portrayal of Circe, namely Apollonius Rhodius'
Argonautica20. Before we accept or reject this thesis, a closer
investigation
Apollonius'
of thisCirce
Hellenistic
has a very
Circe-passage
specific function
is necessary.
: Zeus forces Jason
and Medea to visit her so that she can purify them from the murder of
Apsyrtus (Cf. IV, 559-561, 585-588). Her function is specifically

17 Also
16 Cf. Segal,
HomerCircean
tells ustemptations
how Circe: Homer,
changes Vergil,
Odysseus'
Ovid,companions
1968, p. 432.into pigs by
giving them φάρμακα {Od., X, 213, 236).
18 The monstra in v. 21 may be a reference to Homer's πέλωρα {Od., X, 219).
19 Cf. Agathe THORNTON, The living Universe. Gods and men in Virgil's Aeneid,
Leiden, 1976, p. 106.
20 Circean temptations : Homer, Vergil, Ovid, 1968, p. 429.
 VERGIL'S CIRCE : SOURCES FOR A SORCERESS 127

related to the situation of Jason and Medea, and is therefore very

different
Apollonius'
from her Circe-episode
role in Homer's
contains
Odyssey21
many
or narrative
Vergil's Aeneid.
elements and a
lot of action and dialogue. The only descriptive features are very
common ones, like the poison (αθρόα φάρμακα, ν. 666-667), the fact
that she is the sister of Aietes (v. 684), and her animals (θήρες, ν. 672),
which are described in a very special way for which we cannot find any
parallel, either in Homer or in Vergil.
The only thing that may have influenced Vergil is the atmosphere
of horror and black magic Apollonius creates in this episode (especially
in the description of her dream, v. 665-669). It seems to me that at this
point the similarity between Apollonius and Vergil ends. We can wonder
whether Vergil, whose vignette contains no verbal reference to the Circe-
passage of Apollonius, used the Argonautica as a source. Vergil was
capable of using his own fantasy and creativity to create an atmosphere.
Now that we have tried to find out which elements Vergil borrows
from Homer and how he uses them, a few questions still remain to be
answered : Why did Vergil choose to let Aeneas sail by Circe's island
instead of visiting her, as Odysseus did? And if that was his choice, why
then did he decide to spend 15 lines on a description of a place Aeneas
didn't even visit? And if he decided to insert this episode in his epic,
why then here, at the beginning of book VII?
It is important to keep in mind that Vergil had the freedom to
choose whether his hero would visit Circe or not. When he wrote his
Aeneid there were very few things he absolutely had to introduce in his
epic; Aeneas' departure from Troy together with some companions and
the foundation of a new home in Latium formed only a large frame
which the author could fill up however he wanted. Where Aeneas went
and where he didn't, whom he met and whom he didn't was all Vergil's
own and deliberate choice. Though every Roman knew that Odysseus
had stayed at the sorceress' place for about a year, still this did not mean
that Aeneas had to do the same. So the decision to let Aeneas sail by was
all Vergil's. Worth mentioning is the fact that from the first line on,
Vergil wants his audience to know that, unlike Odysseus, Aeneas and
his companions will sail by Circe's island - as though he wants to warn
his readers and destroy false expectations they might cherish because of
Homer's version of the story. The verb raduntur makes immediately
clear that we cannot expect a long stay on the island, as was the case in
Homer's Odyssey11.

21 In Homer's story she plays the main role in one of the various «retarding» scenes
with a fairy-tale flavour (as other figures such as Calypso do).
22 Cf. Ronald BASTO, The grazing of Circe's shore : a note on Aeneid 7,10, in
Classical World, 76 (1982-83), p. 43. He interprets the verb raduntur as an indication to
the reader that the Odyssean half of Aeneas' wanderings is now finished; cf. infra, n. 32.
128 V. STOFFELEN

But the initial question remains : Why? In Homer, Circe had three
important functions :
1) She was one of the creatures (human or not) which Odysseus
met on his «10-year-trip» to Ithaca.
2) She is the one who leads Odysseus to the Underworld.
3) After Odysseus' visit to Hades, she gives him advice about his
further voyage.
By the time Aeneas reaches Circe's island, these three functions
are already taken over by other figures :
1) Aeneas stayed for a long time at Dido's house in Carthage, but
of course this does not mean that he could not stay at someone else's
place afterwards (for let us not forget that Odysseus enjoyed staying
both with Circe and Calypso for a long time). In this case though we
have to look for another reason, which might bring us very close to the
main explanation for the way in which Vergil treats this whole
Circe-episode. His Aeneid is a poem about the legendary-historic
foundation of Rome, put in a Homeric setting; its character is more
historical and less mythological than that of Homer's Odyssey. Less
mythological does not mean that mythological creatures cannot play a
role in the Aeneid, but rather that the gap between the hero and the
mythological world is bigger in the Aeneid than in the Odyssey. The
character Vergil chose to give to his epic made it less plausible that its
hero, the founder of Rome, would stay for a whole year at the house of
a semi-divine fairy-tale figure or share the bed of a nymph for seven
years23. People he meets on his trip are usually legendary, quasi-
historical and human. And if Aeneas is kept at a woman's house, she is
a human, «historical» figure, a queen of Carthage, and their love-affair
is not a mere fairy-tale, but a forerunner of the troubles between Rome
and Carthage in the second century B.C.24.
It is interesting in this case to see how Vergil uses a similar
procedure for the Cyclops-episode in book III25 : when Aeneas and his
friends reach Sicily they meet Achaemenides, a figure invented by
Vergil26. He was a companion of Odysseus, left behind in Polyphemus'
cave. He tells the Trojans how the cyclops ate two of his friends and
how Odysseus made him drunk and pierced his eye. Just after he
finishes his story, Polyphemus appears with his sheep, blind and

23 Cf. WILLIAMS, ad Aen, ΠΙ, 588 f. : «We may well conjecture that as the Aeneid
progressed Aeneas became to Virgil a less legendary figure, more Roman, more historical,
less Odyssean, and that it became more necessary for his adventures to belong to the real
world».
24 Cf. SEGAL, Circean temptations : Homer, Vergil, Ovid, 1968, p. 429-430.
25 Cf. Franz J. Worstbrock, Elemente einer Poetik der Aeneis, Münster, 1963,
p. 37.
26 Cf. R. Heinze, Virgils epische Technik, 19655, p. 112.
 VERGIL'S CIRCE : SOURCES FOR A SORCERESS 129

holding a pine-trunk in his hands to guide his steps. The Trojans flee,
taking Achaemenides with them (Aen., Ill, 588-683). In both the
Cyclops and the Circe passage, Vergil omits (or avoids) a confrontation
between the hero and the mythological creature; he only describes it as
seen through the eyes of his hero. Aeneas only sees (or hears) it, and
can escape its dangerous power, while Odysseus was really confronted
with it27.
2) Aeneas is led to the Underworld by the Sibyl of Cumae. His
final destination is Latium, Vergil has located the Underworld in Cumae,
and Circe's island is somewhere in between (cf. infra). The geographical
position of these places makes it rather difficult to give Circe again this
guiding function, unless Vergil were to let Aeneas sail up to the North,
before being sent back on his steps by the sorceress. This would be an
unnecessary detour, especially since there was another figure Vergil
could easily use for this role : the legendary Sibyl at Cumae, who had
much closer connections with Rome than the sorceress Circe, and who
got thereby a place in the foundation legend of Rome28. When we
compare Vergil's version of the Aeneas-legend with that of other
contemporary writers, it becomes all the more clear that the introduction
of the Sibyl was a deliberate choice of Vergil. Livy for example writes
ab Sicilia classe ad Laurentum agrum tenuisse (1,1,4), without
mentioning a visit of Aeneas to the Sibyl in Cumae.
3) In the Aeneid, it is Helenus (Priamus' son, who ruled over a
part of Pyrrhus' kingdom by the time Aeneas visits him) who foretells
Aeneas the future dangers on his trip {Aen., Ill, 374-462). Circe
couldn't fulfil this function anymore in the Aeneid, since her island was
the last place the hero met during the Odyssean part of his adventure.
This is again a consequence of the location of her dwelling-place (cf.
infra).
We come thus to the conclusion that the three functions Circe had
in Homer's Odyssey are all taken over by other figures in Vergil's
Aeneid. The more historical character of the Roman poem especially
explains, I think, why Aeneas passes Circe's island instead of visiting it.
Worth remembering in this context is that Vergil tells us immediately in
line 10 that Aeneas and his companions sail by her island.
This conclusion raises the question why Vergil after all decided to
spend 15 lines on a description of Circe, her island and her animals (if
she had no
Trojans' trip).
further
The answer
role in to
histhis
poem),
last question
and whyis,here
I think,
(at the
quite
endsimple.
of the

27 Remarkable is also the fact that Polyphemus' blindness in the Aeneid is a

consequence of what happened in the Odyssey.
28 It is remarkable that the Sibyl starts to appear quite frequently in Augustan
literature (cf. for example Verg., Eel., 4), a fact which might find an explanation in the
role Apollo plays in the Augustan propaganda.
130 V. STOFFELEN

If Vergil wanted to introduce a description of Circe's island, it had to be

a part of the Odyssean half of the Aeneid. The exact location of Circe's
dwelling-place in the Odyssey is much disputed29, but in post-Homeric
tradition the sorceress became more and more associated with the West
Coast of Italy30. A promontory, somewhat South of the mouth of the
Tiber, was supposed to be her «island». By the time Cicero wrote his De
Natura Deorum, the inhabitants of this place even worshipped Circe as a
goddess31. Since Vergil preferred to maintain this geographical position,
he had to introduce the description of Circe's home in that part of the
voyage in which the Trojans pass the West Coast of Italy. As a result
Circe's island became the last place Aeneas met before arriving in
Latium, the place where he leaves the Odyssean world behind and enters
the Iliad part of his task.
Finally, it remains to be asked what effect Vergil wanted to reach
with this little intermezzo in a poem about the legendary-historic
foundation of Rome, put in a Homeric setting. The listener is first aware
of the handsome articulation of the description. But more important is
that he wants to call attention to the difference in character between
Homer's Odyssey and his Aeneid. Right before our passage Aeneas has
visited the underworld; this interlude, so magically entered and
mysteriously left, was strange enough for our «historic» hero. The next
major step of his adventure will be Latium, the place where his «Italian
history» is about to begin. By passing the Homeric-flavoured
Circe-island without visiting it, Aeneas gets a last view of his past, a
past he leaves behind forever32. He is not a second Odysseus who visits
the magic island of the sorceress; he can only hear her and her wild
animals at a distance. He's now the Roman Aeneas who will have to
fight for a new home in Italy. In this way the passing becomes a last
farewell to his mythological wanderings and creates the distance between
the Homeric setting and the historic hero Vergil needed at this point in
his epic.

29 Cf. Bernhard Paetz, Kirke und Odysseus. Überlieferung und Deutung von Homer
bis Calderón, Berlin, 1970, p. 13; Armin und Hans-Helmut Wolf, Die wirkliche Reise des
Odysseus. Zur Rekonstruktion des homerischen Weltbildes, München-Wien, 1983, p. 11.
30 Hesiod (Theog., 1011-1013) calls her the mother of Latinus, which indicates a
close connection with Italy. Cf. Hatzantonis, Le amare fortune di Circe nella letteratura
latina, 1971, p. 3-4; Albin LESKY,A«a, in Wiener Studien, 63 (1948), p. 52; Escher,
Aiaia, in R.E., I, 1 (München, 1988 [= 1893]), col. 920-921.
31 ClC, Nat. Deor., Ill, 19 : Quamquam Circen quoque coloni nostri Circeienses
religiose colunt.
32 Cf. Basto, The grazing of Circe's shore : a note on Aeneid 7, 10 (cf. supra, n.
22), p. 43 : «The literal grazing of Circe's shore by Aeneas and his men is an indication
to the reader that, symbolically, the Odyssean half of the Aeneid has been finished and is
now being bypassed.» Cf. also SEGAL, Circean temptations : Homer, Vergil, Ovid, 1968,
p. 431.
 VERGIL'S CIRCE : SOURCES FOR A SORCERESS 131

PASSAGE II : Aen., VII, 187-191.

Ipse Quirinali lituo parvaque sedebat

succinctus trabea laevaque ancile gerebat
Picus, equum domitor, quern capta cupidine coniunx
1 90 aurea percussion virga versumque venenis
fecit avem Circe sparsitque coloribus alas.

When Aeneas understands that he has reached his final destination,

he sends an embassy to king Latinus. Aeneas' companions are
welcomed by the king in his palace. His palace is, as Vergil describes it,
decorated with wooden images of gods and heroes, war trophies and a
Latinus'
statue of Picus, grandfather, whom Circe had changed into a
wood-pecker. The reference to this metamorphosis is introduced rather
unexpectedly. Moreover, Vergil only refers to the story, without really
taking (or using) the opportunity to elaborate on it in greater detail33 :
after only three lines, he turns his camera to Latinus, who welcomes the
Teucri. The description of the palace is over and the narrative goes on.
For Aeneas and his companions the Odyssean part of the trip is
over, they now find themselves in an Italian environment. This change
in scenery also effects our Circe-figure : she is still a sorceress,
changing people into animals, but the specific metamorphosis mentioned
in these lines has no Homeric precedent. The Homeric sorceress of the
previous passage is now involved in an old Italian story34.
In his commentary on these lines Servius gives us the basic
outlines of the Picus story35 : Fabula autem talis est... he says, and tells
then how Pomona loved Picus and married him, how Circe fell in love
with Picus, but was spurned by him and changed him therefore into a
bird, the Picus Martius, because he himself used to be an augur and had
a. picus in his house, which foretold him the future. Unfortunately
Servius does not mention any specific source Vergil might have used for

33
34
35 Cf. Hatzantonis,
Ibid.
For ancient references
Le amare
to Picus'
fortune
shapes
di Circe
and metamorphosis,
nella letteratura cf.
latina,
Carter's
1971,article
p. 12. in

W.H. RosCHER, Lexicon der Griechischen und Römischen Mythologie, ΙΠ, 2 (Hildesheim,
1965 [=Leipzig, 1902-1909]), col. 2494-2496 and G. Rohde's in R.E., XX, 1 (Stuttgart,
1941), col. 1214-1218. - Comparison between the various references to Picus'
metamorphosis in Latin literature shows that different versions of this legend existed. For
Vergil Circe was Picus' coniunx; according to later authors he was married to Canens
(OVID., Met., XIV, 312 ff.) or Pomona (SERV,. Ad Aen., VII, 190) and repelled Circe's
advances (in this way a more acceptable explanation could be given to her deed; cf.
Carter, col. 2496; ROHDE, col. 1218). Attempts have been made to make Vergil's
version correspond with the other by interpreting coniunx as non quae er at sed quae esse
cupiebat (SERV., Ad Aen., VII, 190). Fordyce finds this solution unacceptable, since no
parallel for this use of the word can be found (Ad Aen., VII, 189 ff.).
132 V. STOFFELEN

these lines, so that it is difficult for us to find out where he found the
inspiration for his version of the Picus-story. Various solutions are
possible :
1) The story of Picus' transformation by Circe was already written
down by an earlier writer,
2) it was a common legend, first put on paper by Vergil's hand, or
3) the basic elements of the story existed but were substantially
expanded by Vergil.
At this point, maybe some lost works of the first century B.C.
could have solved our problem, if only they had survived. Indeed,
Nonius Marcellus, a grammarian of the fourth century A.D., tells in his
De conpendiosa doctrina, 518 M how Aemilius Macer mentioned Picus
in the first book of his Ornithogonia36 :

PICUMNUS et avis est Marti dicata, quam picum vel picam

vocant, et deus qui sacris Romanis adhibetur.
Aemilius Macer in Ornithogoniae (Theogoniae codd., corr.
Bentinus) lib. I :
et nunc agrestis inter picumnus habetur.

This Aemilius Macer was a didactic poet, born in Verona, who

died in Asia in 16 B.C.; he wrote poems on birds, serpents and herbs;
his Ornithogonia must have contained a lot of Italian legends.
Furthermore, both Nonius Marcellus and Servius seem to know
that also Varro wrote about the Picus Martius in one of his works :

Varro de vita populi Romani lib III :

P. Aelius Paetus cum esset praetor urbanus et sedens in sella curuli
ius diceret populo, picus Martius advolaverit atque in capite eius adsedit
(NONIUS MARCELLUS, De conpendiosa doctrina, 518 M) and Varro
Pilumnum et Pitumnum infantium deos esse ait <...> nom Stercutii
Picus, Ρ ici Faunus, Fauni Latinus est filius. sed in his nominibus
abutitur poeta (SERVIUS, AdAen., X, 76).

Of course, these few testimonies do not tell us whether or not

Aemilius Macer or Varro mentioned Picus' metamorphosis by Circe, but
at least they show that some recent literature about the Picus-legend was
available by the time Vergil wrote his Aeneid. Therefore it seems
reasonable to me to assume that at least one of these contemporary

36 The manuscripts read Theogonia, but this is an error in the tradition, cf. C.
Buechner - W. Morel, Fragmenta poetarum latinorum epicorum et lyricorum praeter
Ennium et Lucilium, 19822, p. 138; Hellfried Dahlmann, Über Aemilius Macer, Mainz,
1981, p. 6-7.
 VERGIL'S CIRCE : SOURCES FOR A SORCERESS 133

writers was an important source for Vergil's sudden reference to Picus'

transformation into a bird in the seventh book of his AeneicP7.
The fragmentary character in which these works are preserved
makes it impossible for us to figure out how Vergil used his sources and
why he used them in this specific way. The only thing we can say is that
he refers only very briefly to a story that was perhaps told in greater
detail by some earlier poet.
This fragmentary character makes it hard even to evaluate the
originality of Circe's intervention in the metamorphosis of Picus. None
of the fragments mentions her in connection with Picus, but this does
not necessarily mean that her intervention was invented by Vergil.
For our over-all interpretation of Vergil's Circe-figure, two
elements are important to keep in mind about her appearance here in
Aen., VII, 189-191. First of all, the Homeric sorceress, whose island
the Trojans passed in the beginning of book VII, is now the main actress
in an Italian legend. She is still responsible for a transformation of a
human being into an animal, but this time the person involved is Picus, a
legendary king of Italy. This is an important step in the development of
Vergil's Circe : although she never leaves her Homeric predecessor
behind, she gradually becomes a more Vergilian and a more Italian
Circe. Secondly, she plays again a role which is not necessary to the
narrative of the Aeneid. This was already the case in the previous
passage about the passing of her island, and we will find it again in
Aen., VII, 280-283, the last passage to be discussed in this article.

PASSAGE III : Aen., VII, 280-283.

280. absenti Aeneae currum geminosque iugalis

semine ab aetherio spirantis naribus ignem
illorum de gente patri quos daedala Circe
supposita de matre nothosfurata creavit.

Here again, as in the previous passage, Vergil refers suddenly to

an act for which the sorceress Circe is responsible, though she was not
involved in that kind of activity in Homer's Odyssey. If Homer was not
Vergil's inspiration for this particular passage, which source then did he
use (if indeed he used one)? In this case, Servius' commentary on these
lines gives us some very important information : patri quos daedala
Circe : et hocfingit eamfecisse, tractwn autem est de Hornero, qui tales

37 Cf. BöMER, ad Ovid., Met., XIV, 308 ff. : «und es gibt als sicher das sowohl
Vergil als auch Ovid auf Macer zurück gehen».; cf. also Dahlmann, Über Aemilius Macer,
1981, p. 7.
134 V. STOFFELEN

equos habuisse inducit Anchisen (AdAen., VII, 282). Servius mentions

two facts that are related to our problem :

- et hoc fingit tells that Circe's responsibility for this TOO was
Vergil's invention (the verbfingere is very important here!).
- the story of the horses itself (without the presence of Circe) is
borrowed from Homer's account about the horses of Anchises.
This story occurs in the fifth book of the Iliad (//., V, 268-269)38,
in the αριστεία of Diomedes. In the midst of the battle, Sthenelos
advises Diomedes
Aeneas' horses andto says
flee for
: Aeneas; in his answer Diomedes talks about

της γενεής έ'κλεψεν αναξ ανδρών Άγχίσης

λάθρη Λαομέδοντος ύποσχών θήλεας ίππους.

The situation in both passages is the same : someone

(Anchises-Circe) steals horses (εκλεψεν λάθρη -furata) of a special,
divine race (της γενεής - ab aetherio semine), mates them with a mortal
mare (ύποσχών θήλεας 'ίππους - de matre suppositä) and creates thus a
mixed stock (nothos). We can even point out some verbal parallels39 :

εκλεψεν λάθρη -furata

ύποσχών θήλ!εας ίππους - de suppositä matre.

The different context though makes the allusion rather unusual, for
there is no further similarity between the battle scene in the Iliad and the
visit of the Trojan embassy to Latinus. The only parallel between the two
passages (which can perhaps explain why Vergil borrows this story of
Homer and inserts it in a totally different context) is that Latinus gives
two of his divine horses to Aeneas, like Anchises gave two of the six
«mixed» horses to his son.
If we assume that Vergil chose to give a special character to the
horses which Latinus offers to Aeneas, it is easy to imagine that he
looked for a useful precedent in the Iliad or the Odyssey. What was
more appropriate than the passage in which Diomedes tells how
Anchises bred semi-divine horses and gave two of them to his son? We
can say thus that this allusion to //., V, 268-269 invites the audience to
remember how Aeneas got the same kind of horses from his father in
Troy. This adds, I think, a new dimension to the passage and stresses

38 Cf. Hatzantonis, Le amare fortune di Circe nella letter atura latina, 1971, p. 11-
12; Knauer, Die Aeneis und Homer, 19792, p. 400; CONINGTON-NETTLESHIP, ad ken., VII,
281
39 Cf. Conington-Nettleship, ad ken., VII, 283.
 VERGIL'S CIRCE : SOURCES FOR A SORCERESS 135

the high value and the extraordinary character of the horses Latinus
sends to Aeneas. But of course, it was impossible for Vergil to maintain
Anchises as the author of the action (for how could Anchises have bred
horses Latinus owns in Latium?). And so Vergil brings Circe back on
the stage as a deus ex machina, a suitable author of a supernatural deed
(she was known as a sorceress, able to do all kinds of strange things
with animals). Maybe her role in the transformation of Picus inspired
Vergil to choose this solution. For that passage though Vergil may have
had his sources, while the introduction of Circe here is entirely due to
his own imagination and creativity - as Servius pointed out : et hoc
fingit eamfecisse. Instead of looking for some other divine creature, he
preferred to introduce Circe again.
Various attempts have been made to explain why Circe appears
here for the third time in book VII, without being mentioned in any other
book of the Aeneid40. Vergil's choice to introduce Circe here for the
third time in only 300 lines has, I think, to do with the break between the
Odyssean and the Italian world in the beginning of book VE. We have
seen how in the opening lines of this book Aeneas leaves behind his
Odyssean wanderings and enters the Italian scenery. In order to smooth
the rupture between these two worlds Vergil seems to have chosen to
preserve some Homeric colour in this new environment. A Homeric
sorceress, playing a role in Italian legends, is in this case a perfect
solution for an author who wants to make this change in landscape less
abrupt. Beside that, she also gives him the opportunity to link Italian
tradition and Greek cultural heritage.

We can conclude, I think, that, although Circe does not play an

important role in Vergil's Aeneid, she is nevertheless a very
and interesting example of his creative and original use of
sources. In the first passage, he changes her character and function, but
the allusion to the Homeric model is still visible. In the two following
episodes she is still a sorceress, but Vergil makes her responsible for
acts for which no parallel can be found in Homer's work; the Homeric
Circe enters the Italian scene. In the Vergilian Aeneid, Circe has become
a Vergilian Circe, product of the author's poetic freedom and choice.

Gerard Van Laethemlaan 16, Veerle STOFFELEN

B-2650 Edegem.

40 Cf. e.g. Mario A. Di Cesare, The altar and the city, New York-London, 1974,
p. 138 (Circe as part of a pattern of female fiends in book VII); Segal, Circean
Temptations : Homer, Vergil, Ovid, 1968, p. 430-431 (Circe foreshadowing the violence
and brutish unreason of this book).