Emendations on the third book of Lucretius.
quo magis in dubiis hominem spectare periclis 55 conuenit aduersisque in rebus noscere qui sit; nam uerae uoces tum demum pectore ab imo eliciuntur <et> eripitur persona [dagger]manare[dagger].
58 et suppl. LPAFC : om. OQV manare OQV : manet res FC : manere A : mala re Heinze : manu a re Krokiewicz (quae sibi trib. Martin) : homini re Merrill : minaci Morel : mali a re uel iam ab ore uel mala ore Orth : anima re Bury : in amaris Meurig-Davies : manet os uel manetque os MacKay : ibi ab ore Garcia Calvo
Lucretius here observes that one can most easily learn the truth about the character of others when they experience periods of real strife: from the depths of their heart true words are elicited and the mask that they customarily assume is ripped from them. But what of the unmeaning and unmetrical manare that closes 58 in the mss? Almost every editor since Avancius has adopted the Renaissance conjecture manet res (declared by Lachmann (1) to have been made "egregie"), which provides a third and final verb for the clause and which they have typically taken to mean 'the truth remains'. Yet I do not find myself wholly convinced by this emendation (2). Firstly, the expression manet res is odd Latin for the sense intended and this particular usage of res has no exact parallel in Lucretius (3). Secondly, as Kenney rightly observes (4), the emendation truncates the otherwise striking imagery of the mask torn from the face. Finally, it is odd for the third element of these two verses to be asyndetic, after the pair eliciuntur et eripitur, particularly since manet res only qualifies the latter element (5). There is surely scope for further conjecture (6).
It is a lamentable fact, however, that the emendations made since Heinze opened the floodgates over a century ago, have ranged from the poor to the truly incredible. Heinze's own mala re, intended in the sense 'by misfortune' (7), is more Plautine than Lucretian and carries little weight after 55-56. Although Merrill's homini, however unspecific, is not impossible, the closing re, 'in fact', can hardly function in isolation (8). Morel's minaci needlessly qualifies the hypothetical man under discussion and with an adjective that is not obviously suitable. The emendations of Krokiewicz and Bury are two of the more nightmarish results that the palaeographical method can bring: the former introduces not only the rare elision of -u (9) but also that of an iamb, which I do not believe to be Lucretian (10); further, manu is irrelevant (11) and a re without clear sense (12); the latter, which Bury optimistically translated as "from the soul by the actual experience" (13), loses the imagery of a real mask and joins the others in introducing the difficult bare re. Meurig-Davies' conjecture is almost nonsensical and MacKay's manetque os (disregarding his unmetrical alternative) introduces the painfully banal observation that 'the face remains'. Surveying such a field of emendations, M. D. Reeve stated that manet res "satisfied metre, sense and syntax, so well indeed that it puts the modern conjectures to shame." (14) Although I believe that the Quattrocentro emendation does surpass the more recent efforts discussed thus far, I think that better sense and syntax can yet be found by following an alternative path.
Orth seems to have been right with his later conjectures in restoring ore, 'from the face' (an ablative of separation) (15), from the transmitted are. The primary question is how then to complete the fifth foot when the sense is seemingly satisfied by this conjecture. I suggest that Lucretius could have inserted an adjective that denotes the hollow shape of the mask and thereby its vacuity as a screen for one's true nature: eripitur persona caua ore, 'the hollow mask is torn from his face'. Eskuche (16) records 69 Lucretian instances of the elision of a short vowel in this position; for pyrrhic words elided in this sedes one could compare nisi (1,697; 2,17; 5,527), quoque (1,810; 1,912; 3,328; 4,507; 4,554; 6,769; 6,898), sine (1,1054; 6,207; 6,300; 6,310), ubi (2,313; 4,401; 4,575; 5,688; 6,725; 6,1072), ita (3,180), quasi (3,934; 4,815; 6,515) and neque (5,1409). An infinitive is very commonly elided here but adjectival instances can be seen at, e.g., 1,672 (quadrisyllabic) and 5,1291 (trisyllabic). A more prosaic alternative which is metrically more secure is to read ab ore (as in Orth's earlier conjecture), preceding it with ita: 'and the mask is thus ripped from his face'. In either case the corruption arose from the scribe's accidentally taking the verbal unit following persona as an infinitive of the first conjugation (17); once this was done, we may assume that a scribe corrected the nonsense series of letters preceding -are to the closest Latin verb that occurred to him.
359-361: dicere porro oculos nullam rem cernere posse, sed per eos animum ut foribus spectare reclusis, 360 difficilest, contra quom sensus ducat eorum.
Lucretius here asserts that our senses provide good evidence that our eyes can discern objects themselves and are therefore not mere portals through which the mind 'sees'. As the text of 359 stands, it states 'that the eyes can see nothing'. Although this is unobjectionable in sense, could not Lucretius' point be improved by focusing upon the lack of agency within the eyes: 'that the eyes see nothing by their own powers', i.e. the act of seeing rather takes place from inside the body? We can easily supply this sense by reading per se for posse. per se, 'for themselves', therefore provides an elegant verbal contrast with per eos, 'through them' in the following verse (18). It is curious that some translators seem to have imagined that the transmitted text already bore this emphasis (19). With the common confusion between e and o (20) and r and s (21), pesse or porse would have been at once changed to posse, a corruption that so happened to leave passable sense. per se closes the Lucretian hexameter three times elsewhere (1,445; 2,241; 2,1050) (22).
513-516: addere enim partes aut ordine traiecere aequomst aut aliquid prosum de summa detrahere hilum, commutare animum quiquomque adoritur et infit 515 aut aliam quamuis naturam flectere quaerit.
The employment of hilum in 514 has often occasioned comment but has not to date led to a published emendation. Lucretius employed hilum, 'a tiny bit', as a noun in the nominative (3,220), accusative (3,518 (subject); 3,1087) and ablative (5,1409) singular cases and as an adverbial accusative (3,518; 3,783; 3,813; 3,830; 3,867; 4,379; 4,1268; 5,358). In all of these instances he placed the word at the close of the hexameter, preceding it by a negative particle: two exceptions to the latter rule exist, 3,514 above and 4,515. Heinze (23), however, rightly made the observation that in these two prima facie anomalous instances, the semantics of the sentences' verbs are virtually negative: in the present passage and at 4,515 (et libella aliqua si ex parti claudicat hilum) detrahere and claudicat possess an inherently negative force, thereby giving Lucretius the licence to employ hilum in these contexts without an explicitly negative particle (24). In the word's other occurrences in Latin literature, i.e. in Plautus, Ennius, Lucilius and potentially two other scenic poets, such a negative particle always precedes. Furthermore, in no author is the word modified by an adjective. To return to Lucretius' practice, hilum is only employed as an adverbial accusative with intransitive or passive verbs, where there is no potential for its being taken as a verbal object. 3,514 therefore presents a potential anomaly: if hilum is to be taken as an adverb, it stands as an accusative alongside aliquid (and therefore potentially as an object of detrahere) and is no longer parallel with the occurrence of hilum four lines later, where the word serves as an accusative noun (25). If, on the other hand, it is to be taken as an accusative noun, aliquid could not stand in apposition with it and would require emendation (26).
In view of 518, it seems most natural to take hilum as the accusative object of detrahere (modified by prosum, 'a very little amount') and to emend aliquid accordingly. In the margin of his copy of Munro's Lucretius (27), A. E. Housman suggested reading aliquod, that is the adjectival counterpart of aliquid. Such a change, however, introduces the unparalleled anomaly of modifying hilum by an adjective and therefore cannot be commended with confidence. It would perhaps be easier to emend aliquid to an adverb of suitable force. I suggest aliqua, 'in some way', 'at all'. Although the adverb is not otherwise attested in Lucretius, it is frequent enough in poetry of the first century B.C. (28). The force of the verse is therefore 'or in some way detract a very small amount from the total': indefinite aliqua suitably widens the scope of how quiquomque could bring about a change in their nature (29). If a was miscopied as i (30), or open a in early minuscule as ic, the resultant text would have been inevitably 'corrected' to aliquid; alternatively, a scribe unaware of the restrictive use of hilum could have assimilated the adverb to a more familiar neuter form (31).
603-606: quid dubitas tandem quin extra prodita corpus imbeciHa foras in aperto, tegmine dempto, non modo non omnem possit durare per aeuom 605 sed minimum quoduis nequeat consistere tempus?
The mss provide in 605 evidence for one of the three instances of masculine aeuos in Latin literature (32). The noun occurs 50 times in De rerum natura, never in the nominative and never unambiguously neuter. Nonetheless, neuter aeuom is well attested as the norm in Republican and Augustan literature (33). Of the other two passages where the manuscripts transmit a masculine form, one occurs elsewhere in the poem, at 2,561: aeuom debebunt sparsa per omnem/disiectare aestus diuersi material (561-562). Each Lucretian instance has been taken to support the other and, it must be confessed, it is not impossible that Lucretius did anomalously assign aeuom to the masculine gender (34).
Yet there is room for doubt: the scribal error of adding -m to a final vowel is exceptionally common in Lucretius (35) (and indeed at 1,1024 Q writes omnem for omne at line end); at 2,561, therefore, omne may have originally closed the verse. For there here seems to have been some motivation for Lucretius to have rejected his typical word order of adjective-preposition-noun (when the adjective and noun are metrically equivalent). It seems that metrical necessity lay behind the reversal, as would indeed have been the case if aeuom was employed in its typical gender, thereby demanding that trochaic omne close the line. I would therefore read omnem for omne at 2,561 (36) and omne haec (a welcome pronoun referring to the subject, anima, last mentioned in 600 but kept in focus by prodita and imbecilla in 603-604) for omnem at 3,605 (37). If these emendations are accepted, we are left with a strikingly isolated instance of masculine aeuom in Plautus (Poen. 1187). Although the gender is there not guaranteed either by metre (the phrase closes an acatalectic anapaestic octonarius) or the support of the grammarian tradition, most manuscripts present uitalem aeuom. Since this is the only occurrence of aeuom in the playwright, emendation is inherently risky. Nonetheless, I do not think it would be excessively bold to suggest that the text presents an error of assimilation for uitale aeuom.
674-678: nam si tanto operest animi mutata potestas, omnis ut actarum exciderit retinentia rerum, 675 non, ut opinor, id ab leto iam longiter errat; quapropter fateare necessest quae fuit ante interiisse et quae nunc est nunc esse creatam.
The final line of this passage is another that I find unsatisfactory as transmitted, although it has not to my knowledge yet occasioned comment. Lucretius seeks to dispel the possibility of metempsychosis and concludes thus: 'therefore it must be confessed that the soul which existed before has perished and that which now exists has now been created'. If the two instances of nunc bear the same force and their natural temporal sphere, the resultant sense is untrue: the souls that humans now possess have not 'now' been created. A translation such as P. M. Brown's reads well but does not do justice to the Latin: "the soul which now exists has only now been created" (38), as if a different adverb, such as modo or nuper, were written for the second nunc; I cannot believe that the first nunc means 'now' but that the second has the Plautine sense of 'just now'. Since quae nunc est inevitably refers to the present but one's soul is not, according to Lucretius, created in the present, I see no viable way of taking the two adverbs with the same meaning, as the structure of the Latin demands (39). The sense we would rather expect, in opposition to quae fuit ante in 677, is 'has been created thereafter', that is, after the dispersal of the previously existing soul. I therefore suggest that the latter nunc is an accidental Perseverationsfehler for a word such as inde, post or hinc. Since in minuscule the ductus of ind is similar to that of nunc, inde (commonly used by Lucretius in the temporal sense 'thereafter') strikes me as being the most probable of these suggestions.
769-775: quoue modo potent pariter cum corpore quoque confirmata cupitum aetatis tangere florem 770 uis animi, nisi erit consors in origine prima? quidue foras sibi uolt membris exire senectis? an metuit conclusa manere in corpore putri et domus aetatis spatium ne fessa uetusto obruat? at non sunt immortali ulla pericla. 775
Lucretius here asks why the soul, if it is immortal, desires to leave the body that contains it. The expression quid... sibi uolt in 772 has occasioned some discussion. Bailey translates "why does it wish to?" (40), rightly stating that sibi uelle is not here employed "in the frequent idiomatic sense of 'what does it mean?'" (41). The personification of the animus evident in 773-775 makes it as good as certain that uolt in 772 is to be taken in its root sense of 'wishes'. As a result, the force of reflexive sibi in the same line becomes uncertain: it cannot be possessive and would be irrelevant as an ethic dative; the attempt of Smith to take it as a dative of advantage ("i.e. for its own good") seems special pleading (42). Could Lucretius not have written ibi, which dittography of the preceding s corrupted to sibi? (43) ibi in its temporal sense (as at 3,28; 4,666-667; 4,1183; 5,346; 6,17) would here have a focalising force: "or why does the soul wish to leave its aged limbs at that time (i.e. of death)?": membris senectis provides the temporal framing for this and the three subsequent verses.
Christ's College, Cambridge
(1) K. Lachmann, In T. Lucreti Cari De rerum natura libros sex commentarius, Berlin, 1850, ad loc.
(2) It should be made clear at the outset that I am of the firm conviction that the Italic mss do not represent an authority independent from OQGVU and are therefore only of use as a repertory for Renaissance conjectures.
(3) The least dissimilar employment of res is its occurrence in the formula manifesta docet res (1,893; 2,565; 3,690; 6,139; 6,249) or the collocation res ipsa (2,289; 2,1050; 3,47; 4,396; 5,104; 5,108; 6,469; 6,542); however, I can find no Lucretian parallel for unqualified res meaning 'reality'. S. B. Smith, T. Lucreti Cari De rerum natura libri sex, Madison, 1942, ad loc., wishes to impart far more force to the word than is possible: "res, i.e., the simple fact that the man does fear death"; J. D. DUFF, T. Lucreti Cari De rerum natura liber tertius, Cambridge, 1903, had already glossed the word ad loc. as "'the reality', i.e. a cowardly dread of death". The first modern commentator, J. B. Pius, In Titum Lucretium poetam commentarii, Paris, (2)1514, ad loc., was all the more generous: "Remanet vera non ficta sermocinatio tota dolore vacillans detegente fucos et inania verba".
(4) E. J. KENNEY, Lucretius: De rerum natura Book III, Cambridge, 1971, who states ad loc. that the conjecture "does not complete the striking image of the torn-off mask as might have been expected".
(5) W.A. MERRILL, T. Lucreti Cari De rerum natura libri sex, New York, 1907, ad loc., rightly observes that "a connective before manet is missed". Brieger instead sought to avoid inserting et by reading deripitur for eripitur, but, as commentators have justly objected, the resultant rhythm is unlucretian.
(6) It is no defence of the conjecture to make observations like P. M. Brown, Lucretius: De rerum natura III, Warminster, 1997, ad loc.: "the final monosyllable of manet res... adds to the stark, dramatic effect."
(7) The actual translation of R. HEINZE, T. Lucretius Carus: De rerum natura Buch III, Leipzig, 1897, is "das Ungluck reisst ihnen die Maske ab".
(8) W. A. MERRILL, "Criticism on the text of Lucretius with suggestions for its improvement. Part I, Books I-III", UCPCPh, 3, 1916, 1-46, observes at p. 30 that "Lucretius usually has vera" in agreement to provide this sense. For "usually" read "always".
(9) Otherwise attested only at 1,677; 3,49; 4,1188.
(10) For my emendation removing the one instance of an elided iamb typically retained by Lucretian editors (4,741), see my "Lucretiana quaedam", Phil., 152, 2008, 111-127, at 117-118.
(11) Equally irrelevant is Martin's comparing 4,843 when offering the conjecture as his own.
(12) A. KROKIEWICZ, T. Lucreti Cari De rerum natura liber tertius, Lublin, 1921, says of his emendation ad loc. "cum litteris proximum est, tum sensui satis videtur convenire", which claim few will believe. His subsequent comparison of manu at 2,869 is nil ad rem.
(13) R. G. BURY, "Lucretiana", PhW, 58, 1938, 701-704, at 702.
(14) M. D. REEVE, "The Middle Ages and early Renaissance" in S. GILLESPIE & P. HARDIE (edd.), The Cambridge Companion to Lucretius, Cambridge, 2007, pp. 205-213, at p. 210.
(15) For a similar employment of the word in the poem, cf. 2,705 and 3,497.
(16) G. Eskuche, "Die Elisionen in den 2 letzten Fussen des lateinischen Hexameters, von Ennius bis Walahfridus Strabo", RhM, 45, 1890, 236-264 and 385-418, at 409.
(17) Both (c)auaore and (it)abore differ little in minuscule from -anare.
(18) For similar use of the phrase cf. 2,133; 2,241; 2,910; 2,1092; 3,145; 3,554; 3,561; 3,565; 3,633.
(19) "To deem the eyes, then, of themselves survey nought in existence" (J. S. WATSON), "pretendre, d'autre part, que les yeux rien voir par eux-memes" (A. ERNOUT), "to say that the eyes themselves can see nothing" (W.E. LEONARD), "die Augen (selber) konnen keinen Gegenstand sehen" (E. ORTH); cf. P. J. SCHRIJVERS, "La regard sur l'invisible" in O. GIGON (ed.), Lucrece (Entretiens sur l'antiquite classique 24), Geneva, 1978, pp. 77-121, at pp. 109-110: "La these y [= 3,359-369] est exposee selon laquelle les yeux ne peuvent rien voir par eux-memes: ils sont l'instrument dont l'esprit se sert pour regarder, comme a traverser une pourte ouverte."
(20) o for e: e.g., 1,932; 2,301; 2,928; 4,577; 4,696; 4,879; 5,654; 5,675; 5,1068; 5,1150; 6,13; 6,51; 6,589; e for o: e.g., 1,777; 4,581; 5,675; 5,906; 5,997; 6,237.
(21) s for r: e.g., 2,736; 5,1147; 6,1122; r for s: e.g., 1,668; 4,270; 6,48.
(22) As I argue in a forthcoming article, I believe that per se should also be restored for the very improbable sese closing 2,250.
(23) R. HEINZE, op. cit., ad loc.
(24) WALDE-HOFMANN s.v. rightly term its use "kunstlich ohne Negation". A similar licence can be seen in Lucretius' apparent creation of the adverb perhilum, which closes 6,576 (although still modifying a potentially negative verb, uacillant) and seems to be a metrically convenient equivalent of Cicero's perpaulum (Orat. 2,150; 2,234; Fin. 1,19).
(25) at neque transferri sibi partes nec tribui uolt/immortale quod est quicquam neque defluere hilum (517-518).
(26) Writing illum for hilum, a change that can be traced back as far as the Giuntine edition (Florence, 1512) and presumably lurks in some Italic mss, is of course nonsensical; similarly unprofitable is Havercamp's also printing aliud for aliquid: "Vidistine umquam fedius aliquid, aut vulgarius? Imae cathedrae puero vix dignum esse hoc tentamen confiteberis." are the damning words of G. WAKEFIELD, T. Lucretii Cari De rerum natura libri sex, Glasgow, (2)1813, II ad loc.
(27) Surviving at St John's College, Oxford, in Housnan Cabinet 1.
(28) Verg. Ecl. 3,15; Ov. Met. 15,300; Tr. 2,213; 3,4b,75; Pont. 3,4,28 (also attested in Cicero and Livy).
(29) Although aliqua is there adjectival, cf. its appearance at 4,515 (as above).
(30) As at 1,814; 2,227 (Q); 4,822; 4,1220; 5,187; 6,221; 6,483 (O); 6,832 (O); 6,1234 (O); for the reverse error, cf. 3,566 (Q); 3,835 (Q); 4,437; 4,1124 (Q); 5,2; 5,1212 (Q); 5,1253 (Q); 6,7; 6,777; 6,913; 6,1079.
(31) An emendation of a similar force has been tentatively suggested to me by Prof. E. J. Kenney, viz aliqui, which, though attested only in Plautus, is of a linguistic register often employed by the poet.
(32) The reading is only preserved exactly in O: omnem O : omnen Q : omer V.
(33) Cat. 1,6; Cic. Hort. fr. 110; Verg. Aen. 9,609; Hor. C. 3,5,16; Ep. 1,2,43 (in omne... aeuom), Man. 1,46 and 2,473 (both omne per aeuum).
(34) In contrast is the almost pervasive rejection by editors of masculine lux at 2,806, reading larga for the transmitted largo, and masculine eum at 4,284 (with speculum as an antecedent), a variously emended passage.
(35) Cf., e.g., 1,367; 1,1017; 1,1024 (Q); 2,209; 2,378 (Q); 3,15; 3,47; 3,422 (Q); 3,555; 3,800; 3,906; 3,957 (Q); 3,1000 (O); 4,210; 4,501; 4,511 (Q); 4,530 (Q); 5,241; 5,580 (Q); 5,904; 5,1152; 5,1215 (Q); 5,1421 (Q); 6,155; 6,344; 6,442 (Q); 6,514; 6,818; 6,1067.
(36) This change can be found in the ed. princeps (Brescia, 1473) and doubtless some Italic mss.
(37) omne is printed in the ed. princeps (as n. 36) and the first Aldine (Venice, 1500) but without the introduction of a following syllable to produce a metrically sound line. D. Lambinus, Titi Lucretii Cari De rerum natura libri VI, Frankfurt, (4)1583, ad loc., records that he "olim [i.e. in his first edition (Paris, 1563)] excudi iussera[t], Non modo non omne incolumis", a suggestion he rejected in favour of retaining the transmitted text. S. Gryphius suggested reading omne ut possit, which I do not understand, whereas P. GASSENDI suggested the plausible transposition aeuom... omne.
(38) P. M. BROWN, op. cit., ad loc.
(39) The note of H. A. J. MUNRO, T Lucreti Cari De rerum natura libri sex, Cambridge, (4)1886, II ad loc., "the soul that was before birth has really perished, and on entering a new body has really become a new and different soul" is indeed accurate but does not explain his translation "the soul which was before has perished and that which now is has now been formed".
(40) C. BAILEY, T Lucreti Cari De rerum natura libri sex, Oxford, 1947, II ad loc.
(41) The gloss of D. Lambinus, op. cit., ad loc., is therefore misleading "id est, quid sibi vult, quod anima exit e corpore senectute confecto".
(42) S. B. SMITH, op. cit., ad loc. C. GIUSSANI, T Lucreti Cari De rerum natura libri sex, Turin, 1896-8, III ad loc., wishes to see a conflation of two separate ideas by taking quid both nominally and adverbially, a suggestion I do not find probable: "quid sibi uult exire. Qui son come conflatae due forme della domanda: quid sibi uult cum exit, e quare uult exire."
(43) ita could also bear appropriate sense but is more removed from the paradosis.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||II STVDIA BREVIORA|
|Publication:||Euphrosyne. Revista de Filologia Classica|
|Date:||Jan 1, 2009|
|Previous Article:||Elogio di Epicuro e trasmissione del sapere: semiologia di Lucr. 3,3, te sequor.|
|Next Article:||La celebracion de Flora en los Fastos de Ovidio (V, 183-378): mito de origen y poetica metamorfica.|