The Bellum ciuile has been the subject of three major editions in the past thirty years, attributable to D.R. Shackleton Bailey, R. Badalì and G. Luck.Footnote 1 The existence of these three works highlights the resurgence of sustained interest surrounding Lucan as of the 1970s, with the publication of two significant works, Lucan: An Introduction by F. AhlFootnote 2 and the collective volume of Entretiens à la fondation Hardt,Footnote 3 yet it also demonstrates the difficulty in establishing the text of the Pharsalia.
From a methodological perspective, the focus of our study is not limited to the contributions of the three most recent editors cited in the previous paragraph: all editions of Lucan's text produced since the editio princeps by G.A. Bussi have been consulted, and the elements that are useful for the discussion have been retained.
The passage studied in this paper is found towards the end of the first book. In this passage, the poet describes a procession being held as part of the procuratio prodigiorum. This ceremony is by no means ordinary: the circumstances are dramatic. The scene occurs after crossing the Rubicon in 49 b.c.; Rome is in a state of panic. Dark omens accumulate, announcing a disaster: comets, lightning, a solar eclipse, the eruption of Etna, altars divided by flames, sinister birds, ghosts, etc. Dark oracles delivered by the Cumaean Sibyl and predictions from the Galli announce the wrath of the gods (1.564–5). Terrified by this accumulation of dark omens, the Romans call upon the Etruscan soothsayer Arruns not only to have a better understanding of the cause of this divine anger but also to assuage it. The procession leads citizens and priests around the town. Line 1.593 states that the procession served as a means of purification, clearly indicated by the phrase festo lustro, which literally means ‘by a solemn lustration’. As such, all the main priestly colleges of Rome were present.
The Latin text reads as follows:
599 tum: tunc MZY || 600 reuocant: referunt SGv || Cybeben: -belen PGUVM2Z2 || 602 festus: festis codd. recentiores || 603 laeto: lento D2
Study of this passage leads us to focus on three aspects where the text is uncertain: the meaning of the verb reuocant at line 600, the variant readings festus and festis at line 602 and the use of the epithet laeto at line 603.
1. LOTAM … REVOCANT … CYBEBEN (600)
At line 600, Van JeverFootnote 5 has noted that the expression lotam … reuocant … Cybeben poses difficulty owing to the redundant character of lotam associated with reuocant or renouant, verbs that evoke the bathing of the goddess. In order to overturn Van Jever's claim that the participle lotam is redundant, lotam could simply be considered as a reinforcer, a structure that is not unique to Latin poetry.Footnote 6 Another approach was put forward by the medieval commentator Arnulf d'Orléans and finalized by Hudson-Williams:Footnote 7 lotam … Cybeben must be understood in terms of lauationem Cybebes, a usage in which the participle is equivalent to a verbal substantive.Footnote 8 The verb reuocare would therefore mean ‘to remember’, ‘to re-enact’. This latest interpretation is the interpretation to which Shackleton Bailey returns in the critical apparatus of his edition, which supports the fact that the quindecimuiri re-enact the bathing of Cybele. In fact, the verb reuocant seems more appropriate than the reading referunt which appears in MSS S and G. referunt seems to be a gloss inserted into the text alluding to the return of Cybele in the city after the lauatio. However, the quindecimuiri are referred to as the keepers of ancient religious traditions in lines 599–600, whether with regard to the Sibylline books or to the lauatio ritual. It is, therefore, more plausible that the verb at line 600 emphasizes the lauatio strictly speaking,Footnote 9 rather than solely the route taken by Cybele's chariot during the ceremony. The same argument can be made against Bentley's reuehunt, adopted by G. Luck in his recent edition,Footnote 10 as well as against interpretations looking to attribute this meaning to the verb reuocare.Footnote 11 Furthermore, it seems unlikely that the role of the quindecimuiri was to return the statue of Cybele to the city of Rome. Since the lauatio is intended to symbolize the arrival of the Magna Mater in Rome, it may well contain clues as to the unfolding of the ceremony within recitations on the initial transportation of the statue in Rome. Livy certainly mentions the decemuiri de sacris faciundis in this regard: they are responsible for considering the way in which the statue can be taken to Rome.Footnote 12 However, the focus here is not on the transportation of the statue from Italy but rather on the journey from Pessinus to Italy. Once the statue reached Italian soil, the decemuiri delegated the responsibility for its transport to the Roman matrons.Footnote 13 All sources mention the role of Claudia Quinta on this occasion, whether to say that she was the person who enabled the statue to be moved,Footnote 14 or to say that she was merely one of the women to whom the statue of Cybele was entrusted.Footnote 15 Further to this, Ovid emphasizes the hopeless efforts of men to move the statue.Footnote 16 As such, it seems unlikely that the quindecimuiri would have been responsible for transporting the statue, since they did not play this role during the arrival of the statue of Cybele re-enacted by the lauatio. Furthermore, diverse descriptions of Cybele's return procession to Rome during the lauatio refer to a festive and popular ceremony, accompanied by pantomimes,Footnote 17 which, again, does not seem to correspond to a role befitting the quindecimuiri. For these reasons, we can conclude that the verbal action at line 600, attributed to the quindecimuiri, does not refer in all probability to the transportation of the statue of Cybele.
Nevertheless, understanding the possible meaning of reuocant remains complex: the interpretation proposed by Hudson-Williams seems ingenious, but, in our opinion, encounters several difficulties. First, parallels conveyed by the critic in order to underpin ‘resurrect’ as the meaning of reuocant are not fully relevant: it concerns the restoration of ancient laws, both in LivyFootnote 18 and in Juvenal.Footnote 19 As such, the verb reuocare concerns general rules we should like to see brought back as opposed to a specific action as implied by the regime lotam … Cybeben in our text. The parallel drawn with Seneca the YoungerFootnote 20 appears more relevant, since it concerns religious celebrations (sacrum). However, once again, the object of reuocare is a general term rather than a specific ritual.Footnote 21 Furthermore, Hudson-Williams's interpretation presupposes that the object of the verb is not the action represented by the lauatio but rather the noun Cybeben, completed by the participle lotam, which would represent a substantive.Footnote 22 The structure poses difficulty owing to the absence of corresponding structures with the verb reuocare in Latin literature, which suggests that the proposition should not be adopted and that a different meaning should be found for line 600.
An alternative interpretation proposed for this passage by a number of editors and commentators consists of understanding the verbal action of line 600 as a reference to a purification ritual.Footnote 23 It is for this reason that Burman and Withof propose the conjectures renouant and curant for reuocant Footnote 24 respectively, whilst Getty, following Heitland, suggests interpreting reuocant to mean ‘restor[e] to her original purity’.Footnote 25 However, parallels suggested in a bid to confirm this meaning for the verb reuocare are not wholly convincing. If we consider examples taken from Virgil (G. 4.282; Aen. 1.214, 1.235), the verb reuocare is not used in any form of sacred context to invoke purification. Conversely, it appears to us, whether via conjecture or via interpretation, that these scholars have correctly understood the intended meaning of line 600. Lucan must be alluding here to purification using water. In fact, this is the actual purpose of the lauatio ceremony,Footnote 26 and the reminder of this concept of purification is particularly relevant within the context of the festo lustro (593), included in our passage. The equivalent by Silius Italicus, quique ostia Tusci | amnis amant tepidoque fouent Almone Cybelen,Footnote 27 where the poet uses the verb fouere to mean ‘reinvigorate, purify’, appears to coincide with the meaning proposed by Burman, renouant. The verb renouare with its derivatives is the appropriate Latin word to invoke purification, restoring to an original state:Footnote 28 an example of this can be found in Livy's works on the subject of ritual ceremoniesFootnote 29 and in Cicero's works in relation to a temple,Footnote 30 the world,Footnote 31 or even a painting (in the literal sense) and the Republic (in the figurative sense).Footnote 32 Whilst Withof's conjecture, curant, is close in meaning, it must be noted that it is not credible in terms of palaeography: on the other hand, variation between renouare and reuocare is not uncommon in Lucan's manuscripts. It appears again at 4.204 (where MSS F and L contain the readings reuocata and renouata remaining from the tradition) as well as at 10.311 (where MS G is the only manuscript to contain the reading renouat).
Lastly, in order to defend our proposed correction, it seems necessary to address the argument we had made against the reading reuocant: is it possible to attribute such an action to the quindecimuiri within the context of the lauatio? Whilst the verb renouare refers to the purification of the statue of Cybele by bathing it, this does not necessarily imply that the quindecimuiri were the ones to bathe the statue in the water. The identity of the officiants of the lauatio is uncertain: Ovid's textFootnote 33 appears to indicate that it is the archigallus, but we cannot eliminate participation of the quindecimuiri in the bathing process. However, such a hypothesis is not required in order to adopt the conjecture renouant. In fact, the role of the quindecimuiri is to monitor good practice of various public rituals, notably in connection with the Magna Mater.Footnote 34 As such, since the quindecimuiri advocate and monitor, at the very least, smooth running of the lauatio, this suggests that they were responsible for restoring the Goddess to her original purity. The traditional reading reuocant, whose meaning is not satisfactory, should therefore be replaced by Burman's conjecture, renouant.
2. FESTVS OR FESTIS (602)?
At line 602, manuscripts and editors swing between the readings festus and festis. festus, which features in all the primary manuscripts as well as in the quotation by the Regulae Aurelii Augustini,Footnote 35 is adopted by Housman, Bourgery, Shackleton Bailey and Luck. festis, a reading which appears in later manuscripts referred to by Grotius,Footnote 36 is preserved in the text by Hosius, Lejay and Badalì. The phrase septemuirque epulis festus or festis clearly refers to the priestly college of the septemuiri epulones, responsible for announcing and organizing sacred banquets, notably the epulum Iouis. The singular-number noun which was often used to refer to a member of the college is septemuir epulonum.Footnote 37 As a dative describing epulis, festis would mean ‘the septemuir in charge of ritual banquets’. HousmanFootnote 38 considers that the construction is abrupt and that a verbal adjective such as curandis would be expected to complete the phrase, modelled on other titles such as quindecimuir sacris faciundis or duumuir iure dicendo.Footnote 39 Furthermore, whilst the juxtaposition of epulis festis that occurs is not unique,Footnote 40 most Latin poets in dactylic poetry avoid the homeoteleuton of placing a noun and an adjective agreeing with it next to each other.Footnote 41
None the less, the uniqueness of this homeoteleuton in Shackleton Bailey's study primarily originates from the number of categories of homeoteleuton he distinguishes. In fact, epulis festis is the only example of a homeoteleuton consisting of a noun followed by an adjective agreeing with it, but comparable examples can be found in Lucan, such as suos cunctos,Footnote 42 uindictam quantam,Footnote 43 or animos quantos.Footnote 44 The argument concerning scarce use of homeoteleuton is, therefore, not an incontestable one. As for the construction of epulis festis understood as a dative, it seems wholly excessive to criticize it: usage of the adnominal dative is very much substantiated.Footnote 45 In addition, traces that could mark the use of the adnominal dative in order to invoke the responsibilities of certain magistrates can be found in epigraphic sources. As such, the duumuiri uiis aedibus sacris publicis procurandis Footnote 46 are occasionally referred to simply as duumuiri uiis aedibus sacris publicis.Footnote 47 The reading festis, therefore, seems perfectly admissible. On the other hand, festus is trickier: HousmanFootnote 48 suggests interpreting epulis as an ablative, since the phrase epulis festus is the equivalent of epulo. Cortius suggests interpreting the phrase septemuirque epulis festus as an hypallage, which he describes as being a frequently used poetic technique.Footnote 49 This reading remains rare: the adjective festus is most often used to describe a moment,Footnote 50 or inanimate realities linked to rituals.Footnote 51 This adjective most frequently refers to participation in a ritualistic event and to the joy that can result from it when it is used to describe peopleFootnote 52 and never to the organizers or officiants of a ritual, whether in literary or in epigraphical sources. Lastly, looking beyond the difficulties in construction concerning the reading festus, it is important to highlight that the variant festis seems more appropriate as it attributes a qualifier to epulis intended to refer to banquet rituals.Footnote 53 Therefore, contrary to choices made by Housman, Bourgery, Shackleton Bailey and Luck, the reading festis should be accepted.
3. LAETO … COLLO (603)
At line 603, the adjective laeto, which describes the neck of a member of the Salii, has met with varied responses from editors. The majority of editors kept laeto, with the notableFootnote 54 exceptions of Bentley, who was unsure whether to use the variant lento,Footnote 55 and Luck, who, in the most recent edition of Bellum ciuile, used Jahnke's conjecture, laeuo.Footnote 56 First, the variant lento is not an obvious choice: Bentley questions whether the adjective would be a better fit, since it refers to a lustratio ceremony to attempt to repel civil war. However, descriptions of the priests have not been adapted to this particular event in the passage as a whole. On the contrary, they are all described in several words to identify them through their traditional roles or attributes (palladium for the Vestal Virgin, Sibylline books and lauatio for the quindecimuiri, banquets for the septemuiri epulones, etc.). In addition, the very term lentus would not provide a satisfactory meaning: lentus meaning ‘flexible, supple’ could refer to youth, this being the Salii. However, since the sacred shields they were required to carry were particularly heavy,Footnote 57 it would be surprising for such a quality to be attributed to the necks of the Salii in action. Lastly, lento most likely originates from an alteration to the reading laeto: in MS P (Paris, Bibliothèque nationale de France, lat. 7502), laeto is noted as leto with the normal diphthong marked as a cedilla under the e omitted, as can sometimes occur in manuscripts. The following sequence can be adopted: laeto > leto > lento. The variant lento must therefore be dismissed.
Following Burman's theory, HousmanFootnote 58 defended the reading laeto and noted that the adjective could very well be applied to parts of the body. He also compared Lucan's phrase to gaudentibus umeris in Valerius FlaccusFootnote 59 and suggested that this argument should have drawn a line under other interpretations and conjectural propositions.Footnote 60 We nevertheless feel it necessary to reopen the discussion concerning the reading laeto. Neither Burman nor Housman nor any of the editors to come after them discussed the relevance of using an adjective in this context, and all of them were satisfied with the conclusion that laetus alluded to moves performed by the Salii as part of their tripudium dance.Footnote 61
The adjective laetus and its synonyms can undoubtedly be linked to danceFootnote 62 and more specifically to the tripudium: AcciusFootnote 63 and Quintus CurtiusFootnote 64 connect joy to the tripudium but never in relation to the Salii. Conversely, the dance of the sacerdotes appears to be characterized by movements which emphasize the warlike strength and power of Rome, characteristics conveyed by the tripudium.Footnote 65 In fact, the description of dances linked to weapons in Rome does not relate to the laetitia of the dancers. When Seneca the Younger referred to the tripudium as a dance also intended for Rome's enemies, he expressly emphasized that it was not composed of spineless movements but rather of movements considered uiriles.Footnote 66 It is therefore hardly plausible that Lucan wanted to highlight a certain laetitia regarding the Salii, who were supposed to embody the power of Rome.Footnote 67 Furthermore, the adjective laetus is the only one here to allude to dancing: the line by Lucan is focussed on a different activity typical of this sodality, the activity of transporting sacred shields, ancilia, during festivities to honour Mars.Footnote 68 The shields were attached to bars and were carried over the shoulders of the Salii or of their servants in order to transport them.Footnote 69 In Lucan's text, the Salii carried the ancilia themselves.Footnote 70 As such, it seems impossible that the poet would describe the Salii as dancing in line 603, since they were carrying bars on their shoulders to transport the shields. In addition, since this epithet is in no way characteristic of the Salii, nor of any other sacerdos in Rome,Footnote 71 it should be assumed that the text is corrupt.
In order to correct this passage, the conjecture laeuo proposed by Jahnke and adopted by Luck is a possibility which needs more elaboration, which neither Jahnke nor Luck has offered. First, the confusion between the adjectives laeuus and laetus is not limited to Lucan's manuscripts: examples of this can be found in Pacuvius’Footnote 72 and Persius’Footnote 73 textual traditions. Furthermore, reference to the sacred shield being positioned on the left could be inspired by Virgil, who depicts Picus carrying the shield in his left hand.Footnote 74 Moreover, both representations of the transport of the sacred shields on ancient stonesFootnote 75 specifically depict characters carrying bars on their left shoulders with shields attached. As such, this suggests that the clarification provided by the adjective laeuus could therefore correspond to the scene described by Lucan at line 603. In addition, the use of laeuus to denote the left side of a body part is not limited to Latin literature.Footnote 76 Nevertheless, the conjecture laeuo should probably not be accepted. Reference made to the left does not seem to be of importance in terms of the transportation of the sacred shields. In fact, the transportation of the sacred shields using a bar placed on either the left or the right shoulder seems completely irrelevant. As proof of this, a third representation exists on a sella curulis Footnote 77 depicting the Salii or their servants carrying the shields on a bar, which appears to show a man resting the bar on his right shoulder.Footnote 78 This suggests that nothing prevented the Salii from carrying the ancilia on their left shoulder. Therefore, it seems unlikely that Lucan would have written laeuo … collo if this element was not ritualistic, which means that Jahnke's conjecture should be dismissed.
What other possible solutions could be envisaged for line 603? We suggest two possible corrections for this line, in order of plausibility. The first conjecture proposed is lecto … collo. Other than the fact that the words laetus and lectus are frequently confused in manuscripts,Footnote 79 use of the participle lectus is suggested to us in Lucan 9.478–9: quae lecta iuuentus | patricia ceruice mouet. As such, the use of hypallage indicates that a Salian must be patrician and is particularly honoured, as a priest.Footnote 80 Furthermore, this proposition would form a favourable parallel with the line that follows, where the head of the flamen is described as generoso uertice.Footnote 81 The poet would therefore be placing emphasis on the nobility of the Salian and the flamen by using an adjective to describe a part of their body containing a distinctive attribute.Footnote 82 Nevertheless, the use of lectus here concerning a part of the body would be unprecedented, whereas the two phrases lecta iuuentus (in Book 9) and generoso uertice (at 1.604) are customary.
The second conjecture proposed is laeso … collo. This correction is inspired by another poetic passage, this time found in Juvenal's text (2.125–6): arcano qui sacra ferens nutantia loro | sudauit clipeis ancilibus, which concerns carrying ancilia by means of bars. In Juvenal's text, carrying the sacred shields is described as a tiring activity, with the Salii praised for their endurance in this regard. Similarly, the epithet laeso could allude to the arduous task of carrying the shields by describing the impact this task has on the necks of the Salii. Confusion between laetus and laesus is not uncommon in manuscripts: an example of this can be found in Valerius Maximus.Footnote 83 Nevertheless, it could be objected that the epithet is not particularly common, whilst the other characters are all given relatively customary descriptions. Since the conjectures fail to garner sufficient support here, we prefer to print laeto with obeli in order to signal difficulty to the reader.
To conclude this paper, we first wish to emphasize that we do not seek to close the debate on these lines in Lucan's text. On the contrary, we hope that this paper, along with consultation of all the proposals concerning Lucan's text from Late Antiquity, will serve to renew consideration of both this passage and more broadly the passages of the Bellum ciuile which have not always been discussed by recent editors. The number of errors presumed to have been found in the six lines discussed here should not be surprising for two reasons: first, a corrupt passage is more likely to contain errors. Second, the realia to which this short extract refers were no doubt very exotic for late antique or even medieval copyists, so much so that standard terms may have replaced words specifically describing religious practices in ancient Rome.
We propose three changes to the text edited by Shackleton Bailey: the first (renouant) is a conjecture which was not adopted by editors in the twentieth century; the second consists of choosing an alternative reading to the traditional reading (festis); and the third consists of using obeli to signify probable corruption of the text concerning the adjective laeto. We also propose two conjectures in the apparatus criticus, which could constitute progress in terms of the transmitted text. In addition to their intrinsic value, we hope that these corrections will be of diagnostic value, attracting the attention of commentators on Lucan's work in terms of the often-neglected difficulties posed by these lines. As such, our proposed text of lines 1.599–604 is as follows:Footnote 84
599 tum: tunc MZY || 600 renouant Burman: reuocant O referunt SGv || Cybeben: -belen PGUVM2Z2 || 602 festis: festus O || 603 laeto O: lento D2 an lecto uel laeso?
