Hostname: page-component-74656f9884-j7cmm Total loading time: 0 Render date: 2026-11-07T12:09:08.215Z Has data issue: false hasContentIssue false
  • English
  • Français

A NOTE ON LUCRETIUS, DE RERUM NATURA 3.361*

Published online by Cambridge University Press:  16 April 2014

Abigail Buglass
Abigail Buglass*
Affiliation:
Trinity College, Oxford
*
abigail.buglass@trinity.ox.ac.uk
Rights & Permissions [Opens in a new window]

Extract

A conjecture made by Lambinus in 1565 proposes replacement of difficilest (present in the three authoritative manuscripts O, Q and V) at 3.361 with desiperest. Although commonly printed until the end of the nineteenth century (including in Lachmann's famous edition and Heinze's edition of Book 3, but excluding Wakefield's and Munro's), in the twentieth century the conjecture fell out of favour, both in editions of the entire De rerum natura (DRN) and of Book 3; the manuscript reading of difficilest has been kept by editors since 1900: Duff, Bailey, Martin, Kenney, Müller, Smith and Brown. But with this consensus there are problems, and there are several reasons for preferring Lambinus' conjecture.

Information

Type
Shorter Notes
Information
The Classical Quarterly , Volume 64 , Issue 1 , May 2014 , pp. 413 - 417
Copyright
Copyright © The Classical Association 2014 
Dicere porro oculos nullam rem cernere posse,
sed per eos animum ut foribus spectare reclusis,   360
difficilest, contra cum sensus ducat eorum;
sensus enim trahit atque acies detrudit ad ipsas,
fulgida praesertim cum cernere saepe nequimus,
lumina luminibus quia nobis praepediuntur.   (Lucr. 3.359–64)

Moreover, to say that the eyes can discern nothing, but that the mind looks out through them as through open portals, is difficult, when their own feeling leads us to the opposite conclusion; for it is their feeling that draws us and pushes us on to the very eyeballs; especially since we are often unable to perceive glaring objects because our bright eyes are hindered by the brightness …Footnote 1

A conjecture made by Lambinus in 1565 proposes replacement of difficilest (present in the three authoritative manuscripts O, Q and V) at 3.361 with desiperest.Footnote 2 Although commonly printed until the end of the nineteenth century (including in Lachmann's famous edition and Heinze's edition of Book 3, but excluding Wakefield's and Munro's), in the twentieth century the conjecture fell out of favour, both in editions of the entire De rerum natura (DRN) and of Book 3; the manuscript reading of difficilest has been kept by editors since 1900: Duff, Bailey, Martin, Kenney, Müller, Smith and Brown.Footnote 3 But with this consensus there are problems, and there are several reasons for preferring Lambinus' conjecture.Footnote 4

The normal usage of dicere and Lucretius' own practice lead me to conclude that the expression as transmitted is incorrect. The sentence begins with dicere and is completed in sense by difficilest.Footnote 5 In order for it to make sense, the sentence construction relies upon an understanding of dicere as ‘argue’, ‘declare’, ‘make definite’, or similar. But this use of dicere is seen rarely, and not at all with an accusative and infinitive construction.Footnote 6 Here, it is most likely to convey simply, ‘say’. dicere … difficilest is awkward here unless the author is referring to a specific instance of a physical difficulty with speech – which he is not.Footnote 7 Bailey's translation of dicere as ‘maintain’ might improve the expression, but it is therefore unlikely.Footnote 8 The point of these lines is not that it is difficult to express a particular theory, or even that it is difficult to reason or to assert (were we to accept this unusual interpretation of dicere). Rather, Lucretius rejects the very notion that the eyes are mere windows through which the soul sees, stating instead that it is they which allow sight. The assertion here is that the reasoning does not make sense at all.

In addition to reasons of sense and normal Latin usage, the use of desiperest is far more in accordance with Lucretius' argumentative strategy. Lucretius finishes his refutation of the ‘doorway’ theory of ocular vision at 3.359–66 with a reductio ad absurdum at 3.367–9: if the eyes are mere doors, why not remove them altogether, by which people might see even more clearly? The escalating verbs ducat … trahit … detrudit (3.361–2) are combined with the polyptoton of lumina luminibus (3.364) and the use of acies (3.362), whose primary sense is ‘blade’ (when oculos would have scanned), to draw the reader towards a vivid realization of the pain felt in the eyes at light which is too bright.Footnote 9 Here, Lucretius builds up to his reductio ad absurdum, and it can be no coincidence that the solution suggested for adherents to the theory is a painful one, namely, removal of the eyes (exemptis oculis, 3.368). On argumentative and rhetorical grounds, desiperest is much more suited to the argument at this point in the text, for it acts as a signpost to the idiocy of those who would propose, and adhere to, such a theory.

On examination of Lucretius' use of Latin more generally, one finds that although dicere is never seen with difficile(st), facile(st) or non/haud facile(st) in DRN,Footnote 10 it is seen with desiperest. Each coincidence of dicere and desiperest occurs in an infinitive construction and in a similar context to the argument found at 3.359–64, where Lucretius disparages or derides a philosophical view which is at variance with Epicurean teachings. At 3.359–69, the refutation of the ‘doorway’ theory is an illustration of the argument at 3.350–8 that the body has sensus; itself part of the broader point that the soul and body are united (3.323–69), and which, in turn, leads to the crux of the matter: the soul will die along with the body (3.434–44).Footnote 11 Later, the same construction of dicere … desiperest is used three times: first where Lucretius argues against the immortality of the soul by claiming that the immortal soul cannot be part of something mortal (3.800–2); secondly where he dismisses a fear of blasphemy, and even uses the construction dicere porro … desiperest (5.156–65); and thirdly where he disputes the notion of a single being who invented the nomina rerum (5.1041–3). Bailey uses these examples to strengthen a case against Lambinus' conjecture, where in fact they defend it: the sentence structure and vocabulary of the conjecture is more in keeping with Lucretius' style, and the similar contexts of every other occurrence (especially the example at 5.156–65) increase this possibility.Footnote 12 The correspondences in the text are striking, even down to the enjambment occurring alongside desiperest, which is seen also at 3.361. The appearance of dicere at two other points in DRN together with vocabulary pertaining to madness or foolishness strengthens this argument further: dicere porro … perdelirum esse videtur (1.690–2), and aequa videtur enim dementia dicere utrumque (1.704).Footnote 13 These examples form part of a refutation of the belief in fire as the fundamental element (1.690–704), as well as making a broader criticism of those who ignore the evidence provided by their senses. Here, as at 3.359–69, one finds Lucretius building up to an accusation of idiocy at the end of his rational argument. His fondness for using the language of madness where people neglect the evidence of their own senses is clear:

perdelirum esse videtur.
nam contra sensus ab sensibus ipse repugnat
et labefactat eos, unde omnia credita pendent …   (1.692–4)

The language here, especially of 1.693, is very close to the language of 3.361–2:

contra cum sensus ducat eorum
sensus enim trahit atque acies detrudit ad ipsas …   (3.361–2)

Knowledge of Lucretius' rhetorical strategy and of the normal usage of the Latin words in question render Lambinus' desiperest more suited to the argument than the unsatisfactory and rather bland difficilest of the manuscript tradition. Lambinus' desiperest entirely befits the indignant and ironic tone of Lucretius' arguments at this point.

Footnotes

*

I should like to thank Gavin Kelly sincerely for his suggestions and for his time when first writing up this note. David Butterfield was kind enough to discuss several points with me, and I am fortunate to have benefited from his expertise. For his interest and his advice in the earlier stages of thinking about this textual issue, I owe thanks to Dominic Berry; for reading a later version of the note, to Stephen Harrison. Not least, I gratefully acknowledge the comments of the editor, and of the anonymous reader, whose learned suggestions (particularly on the more ‘poetic’ and polemical Lucretius) helped me to make a stronger case for resurrecting Lambinus' conjecture.

References

1 Unless stated otherwise, all translations are W.H.D. Rouse's, from Smith, M.F., Lucretius. De Rerum Natura. With an English Translation by W.H.D. Rouse (Cambridge, MA and London, 1992 2)Google Scholar.

2 The conjecture is found listed in the emendations at the end of D. Lambinus' 1565 pocket edition of Lucretius: Titi Lucretii De Rerum Natura Libri Sex (Paris, 15652). Where desiperest is printed, it is invariably accompanied by Lambinus' dicat rather than ducat. This is attractive, but it is a separate question. The choice is between seeing dicat balancing dicere at 3.359 or ducat anticipating the physical language of trahit at 3.362.

3 Lachmann, K., Caroli Lachmanni in T. Lucretii Cari De rerum natura libros commentarius (Berlin, 1850)Google Scholar, on 3.361, supports the 1563 conjecture: ‘Lambinus optime desiperest contra cum dicat; cui Wakefieldus et Forbiger cum contra dicunt, desipiunt’; as do Creech, T., Titi Lucretii Cari de rerum natura libri sex, quibus interpretationem et notas addidit Thomas Creech (Oxford, 1695)Google Scholar, and Heinze, R., T. Lucretius Carus de rerum natura. Buch III. Erklärt von Richard Heinze (Leipzig, 1897)Google Scholar. For those who do not print Lambinus' conjecture, see: Wakefield, G., T. Lucreti Cari De rerum natura libri sex, comm. illustr., et cum animadversionibus R. Bentleii aliorum subinde miscuit G. Wakefield (London, 1796–74)Google Scholar; Forbiger, A., T. Lucretii Cari de rerum natura libri sex, ed., annotationem adiecit A. Forbiger (Leipzig, 1828)Google Scholar; Munro, H.A.J., T. Lucreti Cari, De Rerum Natura, Libri Sex (London, 1886 4)Google Scholar, who, with Duff and Müller, prints difficilest with dicat; Duff, J.D., T. Lucreti Cari, De rerum natura. Liber tertius (Cambridge, 1903)Google Scholar; Bailey, C., Titi Lucreti Cari. De Rerum Natura. Libri Sex (Oxford, 1947)Google Scholar; Martin, J., T. Lucreti Cari, De Rerum Natura, Libri Sex (Leipzig, 1969)Google Scholar; Kenney, E.J., Lucretius De Rerum Natura Book III (Cambridge, 1971)Google Scholar; Müller, C., T. Lucreti Cari De Rerum Natura Libri Sex (Zürich, 1975)Google Scholar; Smith (n. 1); and Brown, P.M., Lucretius De Rerum Natura Book III (Warminster, 1997).Google Scholar

4 David Butterfield suggests to me that one could also read credere for dicere. This is a valid alternative solution to the problems of expression in the transmitted text. It should be noted, however, that Lambinus' solution is easier palaeographically. The visible and sonic similarities of difficilest and desiperest make it perfectly plausible that an unusual form was replaced by a more common one in copying the text. Another explanation for the route of corruption is that in Lucretius both difficilest and desiperest occur naturally in line-initial position (except for difficilest at 5.1168). For difficilest see 3.361, 5.527, 735, 1168 (one might note too Catullus 76.13, 14); for desiperest see 3.802, 5.165, 1043.

5 See Bailey (n. 3), on 3.359–62.

6 See TLL 5.1.984.12–985.61 s.v. dicere, iii.c.i. Although dicere can mean ‘declare’ or similar when with a direct object, no example in the accusative and infinitive construction relies solely on this meaning for sense. Cf. OLD 2, s.v. dicere, 2a.

7 This is not how Lucretius conveys difficulty of expression or argument in any case. Where difficile occurs, the infinitive which it modifies is never dicere, but the following: inlustrare (1.137), credere (1.487), vitare (4.1146–7), pervincere (5.99), ponere (certum) (5.526–7), docere et vincere (verbis) (5.735), and (rationem) reddere (verbis) (5.1168). Of these seven examples, five refer explicitly to difficulty in the expression of ideas (while 1.487–8 communicates the idea of problems in understanding them), yet not one example uses dicere to convey this: when discussing the communication of philosophy, Lucretius prefers periphrasis. This is also seen where difficile is avoided with dicere, at 3.730: neque … dicere suppeditat; and 4.174–5: nemost | qui possit neque eam rationem reddere dictis. Compare too 1.833, 2.381, 763, 5.1053 for the use of (per)facile and an infinitive (never dicere) to form a phrase generally referring to ease of expression. For comprehension of these ideas and facile, see: 2.462, 3.158, 856, 4.633, 749, 5.1282, 6.423, 532. nil est quod possis dicere at 1.430 could be problematic, since it seems to imply difficulty with expression similarly to dicere … difficilest. But the phrase ‘there is nothing which you can call [wholly distinct from body and separate from void]’ puts an emphasis on the nil est. Lucretius' point is that there is nothing of a third nature, and that there is nothing which can be called separate from body and void: in this case, it is not something which is debatable. Cf. 3.1024 for an example of posse with dicere. Similarly, dicere contra at 4.484 might be considered to mean something like ‘argue against’, but it equally makes sense as simply ‘contradict’. Far from being problematic, these examples strengthen the argument that Lucretius illustrates difficulty of expression without using dicere … difficilest.

8 Bailey (n. 3), on 3.361. Here, Bailey cites three passages in defence of difficilest: 5.526, 735 and 1168. He concedes that ‘ponere certum, docere, and rationem reddere go more easily with difficilest than dicere here’, but insists that ‘there is no need to accept desiperest from Lambinus, comparing iii.802, v.165, 1043’. Cf. Munro (n. 3), on 3.361, who admits that when the meaning of dicere is ‘pressed’ the construction with difficilest is problematic, but keeps difficilest. Conversely, see Heinze (n. 3), on 3.361, who argues against Munro and Bailey that difficilest cannot make sense with dicere, and also cites Lucretius' habit of using desiperest.

9 See TLL 1.400.8–45 s.v. acies, i.a and i.b. Cf. OLD 2, s.v. acies for the primary meaning. The observations of Kenney (n. 3), esp. on 3.362, are valuable here. My thanks to the anonymous reader for their instructive suggestions at this point, in particular.

10 See n. 7 for the uses of difficile in DRN; see n. 4 for instances of difficilest. Looking at equivalences, haud facile (est) occurs just twice in DRN (in the same sentence), with evellere (3.327–8) and extrahere (3.330), while non facile does not occur at all. Likewise, difficile's direct opposite, (per)facile, is never used with dicere. Where it is seen as an adjective (as opposed to an adverb), it is with the following: exponere (verbis) (1.833), exsolvere (2.381), cognoscere (4.663, 5.1282, 6.423), investigare (4.701–2), factu (4.889), docere (5.1052–3), (placidam ac pacatam) degere (vitam) (5.1154), obsistere (6.331) and reperire (6.532). Moreover, when used as an adverb, facile is seen with agnoscas (2.402), cognoscere (2.462, with possis), (rationem) reddere (2.763, with possis), inditur (2.1125, 1136), tolerarit (aevom) (2.1171), noscere (3.158, with possit), effervescit (in ira) (3.295), insequitur (3.400), (hoc) adcredere (3.856, with possis), concrescere (4.133), penetrare (4.197, with queant), spoliatur … et repletur (4.377–8), (inter se) iunguntur (4.726), haerescit (4.742), commovet (4.746), cognoscere (4.749, with possis), expletur (4.1093), insuescat (4.1282), penetrare potest (5.698), caperetur (5.929), fregere (5.1018), docere … suadereque (5.1052–3), dempsit (honorem) (5.1114), cedebant (5.1291–2), conlaxat (6.232–3), dissoluit (6.352), fit uti des (6.801), insinuatur (6.802–3) and ardescunt (6.898).

11 See also e.g. 3.455–8, 3.526–47 and 3.819–29.

12 Compare Bailey (n. 3) on 5.321, where a repetition at 5.238 ‘is decisive’ in accepting a conjecture made by Bernays, J., T. Lucreti Cari De rerum natura libri sex (Leipzig, 1862).Google Scholar

13 And at 2.978, dicere callent can illustrate wisdom in speaking as opposed to folly.