THE STRUCTURE OF PROPERTIUS BOOK 3

E. COURTNEY

IT IS NOW WELL ESTABLISHED that Books 1 and 4 of Propertius are

composed on a symmetrical scheme in which the poems are arranged in
groups and pairs on a principle of approximate numerical equality-
or even, as I hold,1 exact equality. Mr A. Woolley (BICS 14 [1967] 80)
has tried to establish a similar scheme for Book 3, but his attempt en-
counters two difficulties. First, there is no approach to numerical corres-
pondence; second, the links between individual poems and groups are
nowhere near so plain as in Books 1 and 4, and in many cases those
discerned by him are so tenuous that I fear they must be termed illusory.
It is my contention that Book 3 is not in fact assembled on the same
principles as 1 and 4, and I wish to suggest why this should be so.
The reason, in my view, is this. In Book 1 the situation is static.
The poet's concern is to portray under its different aspects his love-affair
with Cynthia as it was at one period; the book opens with them in love,
it ends with them in love. Book 4 is of course of a different character,
which automatically entails that there is no attempt to depict a situation
either developing or static. In this respect Book 3 is quite different;
it is intended to lead up to the final break between Propertius and
Cynthia in poems 24-25. The method of composing in symmetrical
groups of poems, involving continual reference backward, is clearly of
its very nature unsuitable for building up to a gradual climax; and I now
turn to indicating how Propertius did this.
In this book he first expresses a wish for escape from love in poem 17,
the hymn to Bacchus; the god is a symbol of wine itself, which can drown
the pain of love (cf. Tibullus 1.2.1-4; this poem is much imitated by
Propertius both here and elsewhere). Love is a vitium (6), a malum (10),
a servitium superbum (41). The next poem, on the death of Marcellus,
has nothing to do with Propertius and his love. This is the first instance
of what will appear several times, namely that Propertius does his best
to keep his readers in suspense by giving hints which arouse unsatisfied
curiosity; they now think that 17 indicates no more than yet another
temporary quarrel. Poem 19 is addressed to an unnamed woman, who
keeps remarking that men are lustful; Propertius replies that women are
even more so. The reader naturally supposes that Cynthia is addressed;
this impression has to be revised when he has read the next two poems,
20a and 20b. Propertius has employed deliberate mystification as one
of his means of maintaining suspense.
1Phoenix 22 (1968) 250, where I neglected to remark that Propertius draws attention
to the numerical symmetry in Book 4 with the famous line (4.2.57) sex superant versus.
48
PHOENIX,Vol. 24 (1970) 1.
 PROPERTIUS BOOK 3 49

The manuscripts present the next thirty lines as one poem, no. 20;
this cannot be right. Lines 1-10 were separated from 11-30 by Scaliger,
and the poems may be read as Propertius wrote them in the editions of
Postgate and Butler-Barber. Poem 20a is addressed to a woman, again
unnamed, on her desertion by a lover; Propertius invites her to attach
herself to him and promises fidelity. We are meant to connect this with the
eight-line exemplum of Scylla which concluded poem 19; that poem turns
out to have been a warning to the girl not to abandon herself unreservedly
to her Minos, and we now see the point of vos, innuptae, felicius urite
taedas in the final (according to Housman's transposition) couplet of 19.
In reading 20a we are still left under the impression that it is addressed
to Cynthia, and Propertius deliberately confirms this impression by the
line (7) est tibi forma potens, sunt castae Palladis artes; compare 2.5.28,
Cynthia forma potens, 1.2.30, omnia . . . quae . . . Minerva probet. But
doubt creeps in with the next verse, splendidaque a doctofama refulget avo.
Cynthia's real name was Hostia, but everything which we have heard of
her so far strongly suggests that she was the usual type of courtesan,
the freedwoman of some Hostius (this question is further discussed in an
appendix); her nomen (2.20.19) is not inconsistent with this. Many
editors tell us that 3.20.8 refers to the poet Hostius. Even in a compli-
ment would Propertius, with his literary creed, be likely to call him a
doctus poeta or refer to his splendida fama? (Though imitated by Virgil,
he is not even mentioned by any literary source until Macrobius.)
Doubts are reinforced by the next line (9) fortunata domus (the sense
of which is made quite plain by avo). It is incredible that this could refer
to Cynthia (in spite of her alleged propinqui 2.6.7;2 her domus in 2.8.14
means her household); on the supposition that it did, Markland felt
himself constrained to alter to nimis. And in nostros curre, puella, toros
(10), if addressed to Cynthia, surely ought to be recurre; it would be more
than strange to give no hint that she has previously frequented Proper-
tius' bed. For curre Dr Shackleton Bailey, Propertiana 307, compares
[Tib.] 3.9.24, where in fact we have recurre!
Here ends this little poem of ten lines, the same length as 1.21 and 22
(in view of the uncertainties about Book 2 I do not quote any parallels
from it); though brief, it is perfectly rounded, and line 10 makes an
entirely satisfactory conclusion. Poem 20b is related to 20a as 1.8b is to
8a. It begins thus:
13 nox mihi prima venit;primae data (Itali; date codd.) temporanoctis.
14 longius in primo, Luna, moraretoro.
11 tu quoque,qui aestivosspatiosius exigis ignes,
12 Phoebe,moraturaecontrahelucis iter.
2These alleged propinqui have the ius osculi (see RE s.v.; Walbank Commentaryon
Polybius 1.671); but that is far from proving her to be a Roman matron, since we have
no information to show exactly what this practice meant in the time of Propertius.
50 PHOENIX

(The transposition, corrected by Scaliger, was caused by the homoeo-

teleuton toros, 10, toro, 14). The situation has changed from 20a; he
no longer addresses the girl, and he is accepted now, not pleading.
Line 13 is just as clearly a beginning as 10 was a conclusion.
Here I pause to examine some views put forward recently by Professor
Gordon Williams in his book Tradition and Originality in Roman Poetry.
In his discussion of this poem (413) it is surprising to find him repeating
the old objection to Scaliger's conjecture that the sun ought to be named
before the moon. In this context it is absolutely natural that the moon
should come first, and this was firmly stated by Leo (GGA [1898] 745 =
Ausg. kl. Schr. 2.204). The index to Ausg. kl. Schr. (445), referring to
Leo's discussion of this poem, remarks "nicht in zwei Gedichte zu tren-
nen"; I do not know whether this is intended as an expression of personal
opinion by the compiler of the index, but it is certainly the direct opposite
of what Leo says. Professor Williams also maintains the unity of 3.20,
but my disagreement with him is more apparent than real, since he agrees
that there must be a pause in the action after line 103 (but be it noted
that in the poem itself there is nothing to motivate this abrupt transition).
What seems to me decisive is the analogy of 1.8, which the symmetrical
scheme of the book proves to be two poems; Williams rejects both scheme
and separation. His reference of 3.20 to Cynthia (491; on 535 he refers
lines 1-6 to her husband on the supposition, discussed below, that she
was a Roman matron) and the dramatic dating of the poem to the begin-
ning of the affair depend on his hypothesis of unitary publication of
Books 1 to 3, which is surely to be rejected.
The opening of poem 20b at last reveals all; he is starting a new affair
with a new girl (amore novo, 16, though as Shackleton Bailey [204] points
out this is not conclusive). It is totally impossible to interpret this em-
phatically repeated primus of a resumption of his affair with Cynthia.
With hindsight we can now see that poems 19 and 20a must be inter-
preted in the light of 20b; just as in 17 Propertius tried to escape Cynthia
with wine, now he tries to escape her with another girl.
He proceeds to explain that they must draft out and seal a proper
marriage settlement (sacra marita, 26; this is of course euphemistic, like
the common elegiac use of maritus; cf. dominae, 29) before the nox prima;
otherwise, if the "marriage" should break up, the wronged party has no
redress from heaven. Thisfoedus with itsfides (24)4 is like that between
Catullus and Lesbia (Cat. 87.3-4, 109.6). Without such a contract there

'Likewise White in Laudatores Temporis Acti, Studies in Memory of W. E. Caldwell

(Chapel Hill 1964) 70.
4Cf.fidus 9; we thus have the complete Ablaut series of this root. This aspect of the
poems is interestingly discussed by R. Reitzenstein, Zur Sprache der lat. Erotik (Sitzb.
Heidelberger Akad. 3 [1912] Abh. 12) 12.
 PROPERTIUS BOOK 3 51

is only libido, not amor, and libido will presently break free. This libido
refers us back to 19.1-2, where Propertius, though asserting that it is
stronger in women, could not deny its presence in men. Here he is
thinking of the lover who has abandoned this girl in consequence of her
failure to secure such a contract. That is the point of 20a.5, at tu, stulta,
deos, tufingis inania verba. Dr Shackleton Bailey (204) interprets this to
mean "you cheat yourself with talk of gods and such-like empty words";
for fingere deos, "to call on the gods deceptively," he compares Lucan
5.159, superis quosfingis (where his misunderstanding can be corrected
from 148) and Stat. Theb. 7.497, quid molles lacrimas venerandaquenomina
fingis (is Polynices supposed to be calling on his tears?). Propertius'
meaning, I think, is that the girl pretends that the man swore by the
gods and used words ("as it might be fides, foedus etc.," Shackleton
Bailey) which she now alleges were empty; but he never used them at all,
and therefore non habet ultores nox vigilanda deos (20b.22). Poem 20b then
ends by cursing men who do make and break such contracts; Propertius
thus recurs to the idea of fidelity imposed on the man and thereby rein-
forces the protestations of 20a.10.
Thus throughout 20a and 20b there is no hint at all of female infidelity.
This would be incredible if the poems were addressed to Cynthia; we
should have to suppose a tactfulness and chivalry which are the last
qualities displayed by Propertius in such situations, qualities indeed
which the conventions of his literary genre would hardly allow him to
display.
In poem 21 Propertius tries yet a third, the last available (9), remedy
for his love, since others have failed (5). He will now run away from it and
go on his travels, a remedy envisaged in 1.1.29, tried in 1.17, and recom-
mended by Ovid in his Remedia Amoris 213-248 (Ovid also deals with
wine 803-810 and suggests a second mistress 441-488). Poem 22, exhort-
ing his friend Tullus to return to Italy, at first sight seems to be irrelevant
to Propertius and Cynthia; but in fact there is an elaborately contrived
contrast with 21. The return of Tullus from Asia directs the reader's
mind back to 1.6, which referred to his departure for Asia; and that poem,
with its contrast between Tullus and Propertius, is crucial for the poet's
attitude to his life. Propertius there praises Tullus for his devotion to
his state duties and indicates that his own surrender to love is by com-
parison contemptible: nequitia (26), cessatio (21). Now, years later,
turpis amor (3.21.33) is driving Propertius from Italy to find consolation
in the artistic riches of Athens (25-30; compare doctas Athenas with
an mihi sit tanti doctas cognoscereAthenas? [1.6.13] and note how the situa-
tion has changed); he will either find relief in spatia annorum et (Scaliger;
aut codd.) longa intervalla profundi (31) or he will meet a mortis honesta
dies (34). Tullus on the other hand actually has been separated from
52 PHOENIX

Rome by time and distance; but unlike Propertius he can now return to
an honourable life in Italy (22.39-42) with the prospect, first, of continu-
ing his civic activity (40-41), second, of a loving wife and legitimate
offspring (41-42). The Greece which appeared so attractive to Propertius
in his flight from Italy is now depicted as the home of monsters and hor-
rors (27-38), whereas Italy is shown in the most attractive colours
(17-26). The contrast between poems 21 and 22 thus accentuates the
humiliation which Cynthia's love has brought to him, and secures the
reader's understanding for his attempt to break free.
Poem 23, on the loss of the writing-tablets which once carried love-
messages, if considered on its own appears, like 18, to have nothing to do
with the theme of Propertius' increasingly desperate struggles to escape
from his love. But when we have read poems 24 and 25, hindsight will
again make us see that the poet has once more been deliberately keeping
us in suspense, so that the climax of the break in 24 and 25 will come with
all the more effect for its surprise. It is a generally valid principle that
poems should be read as self-contained wholes, including within them-
selves everything needed for their understanding; but we have seen
several cases in which the full meaning depends on context and setting.
I believe the same to be true here; it is hard to avoid understanding the
loss of the tablets which once carried love-messages as in some degree
symbolic of the rupture with Cynthia,5 though Propertius has avoided
any phraseology to suggest this. Similarly one cannot but think that
24 and 25 partly symbolize the poet's farewell to erotic elegy before he
turns to aetiological, just as Ovid symbolized the same transition by
writing a poem on the cure of love.
It can accordingly, I hope, be seen that towards the end of the book
at any rate (for I do not deny that poems 1-5 form a group) Propertius
needed a method of arrangement allowing linear progression; he could
not have achieved his striking, though carefully prepared, climax by
arranging the poems in groups with interlacing correspondences to each
other.6

APPENDIX: THE STATUS OF CYNTHIA

PROFESSOR GORDON WILLIAMS in his book Tradition and Originality in
Roman Poetry argues that Roman Elegy envisages love-affairs with
married Roman women; but his discussion is, I think, over-influenced by
a desire for originality and too low a regard for traditional beliefs (542,
5And not just with Cynthia, but with love in general; note the plural puellas in line 5.
"This article was with the editor before the appearance of that by R. J. Baker in AJP
90 (1969) 333, which coincides with mine in some points and does not persuade me to
change my view on others.
 PROPERTIUS BOOK 3 53

"contrary to received opinion"). In the first place, the case of Lycoris,

which does not suit his thesis, is played down (541): secondly, Tibullus'
explicit statement about Delia (1.6.67-68) has to be twisted out of its
plain sense (536). The status of Cynthia is discussed on 529 sqq. The first
piece of evidence produced is 2.23, but Williams himself admits that this
poem does not necessarily offer precise evidence; Propertius (cf. 3.14.23-
24) is elaborating on a contrast suggested to him by Philodemus in a
lost epigram quoted by Horace Sat. 1.2.120-122 (note si exierit vir), and
Horace himself ibid. 127 sqq. (compare vir rure recurrat with Propertius
2.23.20, hodie vir mihi rure venit). Again, 2.6.41-42 is agreed to provide
no firm basis. In discussing 2.7 Williams begins by expressing, quite
reasonably, disapproval of enquiring why Propertius could not have
married Cynthia, but then proceeds to answer this question. There are
two possible answers; either Cynthia was married already, or her legal
status made marriage impossible. Williams chooses the first answer,
which suits his thesis; but he produces no solid argument against the
second, since his whole discussion is based on the unwarranted assumption
that the projected law was similar in its provisions to those which were
actually enacted at a later date. In fact this poem too is neutral. In the
fourth place he argues that 2.32.43-46 necessarily imply that Cynthia
and Lesbia are in all relevant respects comparable; this would be a valid
inference if we were reading Aristotle. And finally 2.13.9-12, a general
statement not to be tied down to Cynthia.
On the other side Cynthia-Hostia is associated with Lycoris-Volumnia
by 1.8 and 2.16; as the mistress of Gallus followed a soldier to the Rhine,
so Cynthia is willing to accompany a praetor to Illyria. 1.8a in fact is
evidently much indebted to Gallus.
KING'S COLLEGE,LONDON