1. AVDAX VITA, SCELERVMQVE PLENA

The reading audax uita(,) scelerum plena appears in MSS D G F E and is almost universally accepted by the editors, with or without the comma, depending on the individual editor. That said, MSS dFootnote ² T offer the slight but important variant audax vita et scelerum plena, which Pintianus supports (‘scribe’) and which Sillig (both 1831–6 and 1851–8), Mayhoff and André all include in their apparatus critici. (Whether intentionally or unintentionally, Nettleship also prints the conjunction.)Footnote ³ Salmasius proposes audax id, at scelerum plenum (taking audax and plenum as adjectives modifying aliquid), and the editio Hackiana of 1668–9 dutifully reports this proposal in the notes. Later in the same edition, however, Gronovius condemns the emendation (‘infelicissime, mea quidem sententia’), and Harduinius later cites Gronovius approvingly in his notes. (Franzius, Valpy and Alexandre et al. all record the controversy in their notes, as well, but without further comment.) While Dalecampius prints the consensus reading (with the comma) in his text, in a marginal note he also reports the variant audax uita scelerumque plena from either an unnamed uetus or, perhaps, unnamed ueteres—whether ‘Vet.’ indicates the singular or plural here is unclear from Dalecampius's own table of abbreviations. The editio Hackiana, Franzius, Valpy and Sillig 1851–8 all include this reading in their respective apparatus critici, but subsequent editors have not followed their lead, whether intentionally or unintentionally. Given the potential resonance of Horace's integer uitae, scelerisque purus (Carm. 1.22.1), I would like to resurrect this long-forgotten reading, in spite of its weak manuscript support: at the same time, I am not foolhardy enough to suggest that the proposed emendation is at all certain, especially since the resonance of Horace could very well have been what inspired the reading with -que.

2. CVM VIX … SVFFICIENT

The reading quamuis appears in all the manuscripts, and it meets with general approval from most editors, commentators and critics.Footnote ⁴ However, other scholars have objected to quamuis as inconsistent with the logical and rhetorical flow of ideas, and they have accordingly sought a suitable emendation which restores their understanding of the sense of the passage. Pintianus suggests cum uix (‘legendum reor’), and several editors have reported this idea in their notes or in their apparatus critici, although none has introduced the emendation into the text itself.Footnote ⁵ Ianus likewise discerns the need for a correction and suggests a similar alteration of the manuscript reading with quamuis uix, which Mayhoff and André record in their apparatus critici and which Rackham goes so far as to print in his text. In the same vein, Detlefsen ventures quom uix, which Mayhoff, Rackham and André all include in their apparatus critici (André reading ‘quum uix’). In his apparatus criticus, Mayhoff offers quamquam uix, but then questions the need for any emendation (‘ego malim quamquam vix. sed vix parum apte collocatum sit; fortasse nihil mutandum’; cf. Semi in his apparatus criticus). I believe that we come closest to understanding (and, therefore, correcting) the problem by observing the addition of non in MSS d T to produce the reading quamuis … non sufficiant (André in his apparatus criticus). Smith long ago identified the underlying issue and suggested the addition of non: had he further recognized the meiosis in cum uix … sufficiant (≈ cum … non sufficiant), I suspect that he would have unhesitatingly supported the reading of Pintianus.Footnote ⁶ Of the various proposed emendations, I support the reading of Pintianus against the consensus of the manuscripts, as well as against the alternate readings quamuis uix, quom uix and quamquam uix. Of these three options, quamuis uix does not appear to have been used in Classical Latin; quom seems like an unnecessary archaism; and quamquam uix, although it is used at least sometimes elsewhere in Classical Latin (Cic. Phil. 13.6, Fam. 7.5.3; Livy 2.49.11, 7.13.6; Ov. Met. 3.578; Quint. Inst. 7.3.8, 12.1.20), seems like an unnecessary alternative merely aimed at elucidating a hypothetical haplography (that is, quamuis uix vs quamquam uix); in contrast, cum uix requires only the slightest of changes from quamuis and reflects the standard usage of the phrase, as well as the regular use of uix with forms of sufficio.Footnote ⁷

3. AMPLITVDINI ANTEMNARVM … ALIA VELA

The manuscripts uniformly agree on the reading amplitudini antemnarum … uelorum alia uela, and scholars before Urlichs leave the passage untouched, except for the alternation in forms between dissimilated antemna and assimilated antenna.Footnote ⁸ Urlichs, however, contends that the manuscript reading is meaningless and that, instead, uelorum ought to be moved up in the sentence into the prior clause: amplitudini uelorum antennarum … alia uela (‘velorum steht in den Hdss. nach addi, wo es keinen Sinn gibt. anten. gehört zu arbores’). Detlefsen adopts this suggestion into his text, as does Mayhoff; Rackham follows suit, but without any indication that the reading is, in fact, an emendation. I agree with Urlichs that uelorum does not make sense in the text as is, but I disagree with the solution which he offers: Semi is right to comment that ‘velorum post amplitudini tantulerunt editores sed frustra’. Rather, I would contend that uelorum ought to be expunged altogether as a marginal gloss which has crept into the text, whether as an explanation for antemnarum (OLD s.v. antemna records the meaning ‘sail’ as ‘poet.’) or as an explanation for the forms of alia which follow—in which case, one could delete either uelorum or uela as the marginal gloss. Simply put, uelorum alia uela strikes me as needlessly redundant, and scholars routinely ignore uelorum in their translations of the passage.Footnote ⁹

4. PRAETERQVE ALIA IN PRORIS ET ALIA IN PVPPIBVS PANDI

As Salmasius and Gronovius attest, all the manuscripts concur in reading praeterque in proris et alia in puppibus pandi, and this is the reading noticed by Sillig 1831–6 in his apparatus criticus and subsequently adopted into most modern texts.Footnote ¹⁰ Sensing that something was still not right, however, Salmasius ventures praeter quae in proris, but Gronovius rebukes him for the suggestion and vouches for the manuscript text (‘Salmasius corrigit: praeter quae in proris, oblitus hoc praeterque Plinio in deliciis esse: …. Praeterque in proris et alia in puppibus, est, et praeterquam in proris etiam alia in puppibus vela pandi’). Despite its apparent lack of manuscript authority, the vulgate reading praeterque alia in proris(,) et alia in puppibus pandi enjoyed a certain popularity over the years, appearing in most of, if not all, the editions before Sillig 1851–8.Footnote ¹¹ Since then, however, only a few scholars have printed this vulgate reading in their text or even included it in their apparatus criticus.Footnote ¹² I would like to express my support for the addition (if such it is) of alia between praeterque and in proris in order to balance the alia which follows between et and in puppibus, and to take this part of the sentence as the climax of the tricolon which begins with super eas tamen addi alia uela.

5. DENIQVE TAM PARVO SEMINE NASCI

The reading denique tam paruo semine nasci appears in all the manuscripts, and the passage is left intact by all scholars before Mayhoff.Footnote ¹³ In his apparatus criticus, however, Mayhoff speculates about the possible loss of the preposition e (‘an e tam paruo?’) and compares HN 17.72 e tam paruo gigni arbores. The proposal has not caught the attention of many subsequent editors: Rackham goes so far as to adopt the emendation into his text (with the rather compressed note ‘e add.? Mayhoff’ in his apparatus criticus), while Semi simply leaves it in his apparatus criticus, but no other editor even bothers to mention it. All the same, despite the apparent lack of interest in the idea, I would like to revisit this suggestion, even if only to bolster the argument against it. The reading e tam paruo gigni arbores in 17.72 certainly does bear a strong resemblance to the reading denique tam paruo semine nasci in 19.5 (e tam paruo ~ tam paruo semine, gigni ~ nasci), but there are two details worth emphasizing when comparing the two passages. First, the reading in 17.72 is itself an emendation, albeit a relatively secure one, proposed by Gelenius: in his apparatus criticus, Mayhoff reports that etiam is the manuscript reading, while ex tam is the vulgate, as well as the origin of e tam.Footnote ¹⁴ Second, the explanation for the difference between e tam paruo and tam paruo semine appears to lie in the difference between the substantival use of paruo in 17.72 and its adjectival use in 19.5: HN 27.60 paruo semine milii nascente (the reading paruo semine is secure) suggests that Pliny may have distinguished between substantival e paruo (perhaps in order to avoid any confusion with substantival paruo ‘at a low price’: OLD s.v. paruus 4d) and adjectival paruo. Ultimately, regardless of the reading adopted (with or without e), the sense of the text is clear.Footnote ¹⁵

6. INIVRIA AD SVMMAM AVDACIAM PERVENIRE

The text of the culmination of the rhetorical period in 19.4–5 remains especially unclear, and something may very well have dropped out in transmission. The manuscripts generally offer the defective reading iniuria ac summa audacia et peruenire: a later hand in one manuscript (D²) deletes the conjunction et altogether, and this deletion, as we will see, paves the way for the most likely fix. Scholars have suggested a variety of changes for solving the textual conundrum, from such relatively minor adjustments as the insertion of a quo before peruenire (which links the clause with what follows) or the alteration of et to eo (which completes the clause on its own terms) to a number of more radical emendations which attempt to fill in a purported lacuna after et. The vulgate reading iniuria naturae(,) ac summa audacia(,) et quo peruenire appears in de Valle, Beraldus and Chalderius, Caesarius, Erasmus, Dalecampius and the editio Hackiana; thereafter, Franzius, Valpy, Sillig 1851–8, Ianus and Mayhoff all dutifully record the reading in their respective apparatus critici (with Mayhoff inadvertently printing ‘-cia | quo’ across the line-break for ‘-cia | et quo’). Salmasius, noting that neither naturae nor quo has any manuscript authority (‘omnes [manuscripti] carent illa voce naturae. nec illud quo agnoscunt’), recommends the slight change of et to eo in order to produce the reading iniuria ac summa audacia eo peruenire. Elegant in its simplicity, this emendation finds its way first into a note in the editio Hackiana and then into the text of Harduinus, who adds a comma after audacia, as well as his own explanatory note: ‘Ut nempe orbem terrarum vel sic ultro citroque portet.’ Nearly every subsequent editor until Sillig 1851–8 adopts this emendation (with the additional comma after audacia),Footnote ¹⁶ and Sillig 1851–8, Detlefsen, Mayhoff and André all include the reading in their respective apparatus critici. Despite the strong support for eo, however, Sillig 1851–8 rejects the change, posits a lacuna after et and offers his own suppletion in his apparatus criticus: ‘At spreta hac interpolatione post copulam et signa lacunae posui, ut fere talia suppleri possint: et pari damno in mare pervenire.’ (Ianus and Mayhoff both report Sillig's claim of a lacuna but not his proposed suppletion in their respective apparatus critici.) In the text itself, Sillig 1851–8 prints iniuria naturae ac summa audacia et < … > peruenire, a conflation of the vulgate reading with the manuscript text. A year later, Strack suggests an even simpler emendation: ‘Auch dürfte am Schlusse des Satzes die Annahme einer Lucke vermieden werden können, wenn man statt ac summa audacia et pervenire lesen wollte: ad summam audaciam est pervenire.’Footnote ¹⁷ Ianus cites this proposal in his apparatus criticus and offers his own suppletion for Sillig's lacuna (per uehiculum tale ad longinquas terras peruenire), but, in the text itself, he prints iniuria ac summa audacia est < … > peruenire (adopting est from Strack). A few years after that, Urlichs—without any mention of Sillig, Strack or Ianus—offers iniuria ad summa audaciae peruenire, and claims in a note that ‘die Hdss. haben ac summa audacia ei’ (where ei must simply be an error for et). Subsequent scholars—likewise without any mention of Sillig, Strack or Ianus—generally adopt Urlich's reading into their text,Footnote ¹⁸ but Mayhoff relegates it to his apparatus criticus, choosing to print another emendation of his own instead: iniuria, ac summa audacia peruehi mare. In a departure from previous texts, Mayhoff takes iniuria with what precedes and not with what follows: in the absence of any note in the apparatus criticus, it is simply impossible to determine whether the comma is intentional or unintentional (that is, whether Mayhoff wanted it put before or after audacia). Rackham and André record this emendation in their respective apparatus critici; Semi sums up the state of the question succinctly in a note: ‘ac … mare Mayhoff, alii ad summa audaciae pervenire vel aliter.’ I believe that a combination of the readings offered by Urlichs and Strack may come closest to what Pliny may have originally written, and so I would like to propose the emendation iniuria ad summam audaciam peruenire. On the one hand, summa audacia is a well-established collocation (Cic. Font. 44; Caes. BGall. 1.18.3, 7.5.1; Sall. Cat. 18.4; cf. Val. Max. 9.15.2 ad summum audaciae gradum); on the other hand, so is ad summ- with a form of (per)uenio: instances abound of ad summ- where the form of summus serves as an adjective in agreement with a noun, whether concrete or abstract;Footnote ¹⁹ in addition, there are examples of the substantives ad summos (Cic. De or. 2.85 ad summos peruenire), ad summum (all without a complement in the genitive, except Hor. Ep. 2.1.32 uenimus ad summum fortunae)Footnote ²⁰ and, most importantly for the text under discussion, ad summa (all without a complement).Footnote ²¹ When we turn to Pliny himself, we find that he uses ad summa on its own in 9.50 and 17.152,Footnote ²² but that he also uses ad summa with a complement in 9.152 ad summa aquae uentum est and 32.1 uentum est ad summa naturae exemplorumque per rerum ordinem.Footnote ²³ Elsewhere, we find that Pliny twice uses ad summ- with a form of peruenio and that, in both instances, summus modifies a noun: 5.50 usque ad summum Delta perueniunt and 9.22 ad summam magnitudinem peruenire. Taking all of this evidence into consideration, I understand why Urlichs proposes ad summa audaciae, but I believe that Strack's ad summam audaciam rests on better authority and deserves to be the accepted text.

Many, if not most, of these readings are likely to remain contested; that said, I do hope that I have succeeded in some measure in advancing the discussion about an important passage. With the text of 19.4–5 on a somewhat more secure footing, I would like to tackle a translation of it by way of a closer examination of the clause structure of the period and its rhetorical effect.Footnote ²⁴

The clause structure outlined above divides the passage into two roughly equal halves following the opening maxim audax uita, scelerumque plena: each half, in turn, consists of five sections which nicely balance and, in some cases, directly parallel one another. To aid comprehension of the overall syntax, I have italicized the infinitives and the subjunctives, as well as the participles which complement them, and put in bold the various conjunctions, as well as the relative pronoun, which carry the flow of ideas from one stage to the next. Above all else, denique clearly marks the division of the period into its two halves. Elements A₁ and A₂ describe both the plant's growth (aliquid seri ~ denique tam paruo semine nasci) and the purpose for which it is grown (ut uentos procellasque capiat ~ quod orbem terrarum ultro citro portet). On a larger scale, both elements B–E₁ and B–E₂ then include a negative (nec ~ neque), which prepares the way for a sed clause. Within these two complementary parts of the period, elements B–C₁ and B–C₂ develop the theme that the cultivation of linum for the manufacture of uela represents an impious act of arrogant excess (et parum esse fluctibus solis uehi, iam uero nec uela satis esse maiora nauigiis ~ tam gracili auena, tam non alte a tellure tolli, neque id uiribus suis necti).Footnote ²⁵ Thereafter, elements D–E₁ and D–E₂ bring their respective halves of the period to a thunderous finale with an outburst of righteous indignation, as Pliny rails against the provocation of death and then against the audacity of man's attempt to conquer the seas. The sharp contrast between the unnatural size of the yards (cum uix amplitudini antemnarum singulae arbores sufficiant) and the accumulation of still more sails (sed, …, super eas tamen addi alia uela, praeterque alia in proris et alia in puppibus pandi, ac tot modis prouocari mortem) subtly anticipates the similarly sharp contrast between the unnatural origins of linen (fractum tunsumque et in mollitiem lanae coactum) and the impiety of taking to the seas (sed … iniuria ad summam audaciam peruenire). Through this examination of the clause structure, we can discern how Pliny uses repetition and variation in order to develop his train of thought and, at the same time, how he stays focussed on the essential theme of audacia, especially in the elegant balance between the tricolon crescens in elements E_1A–C and the powerful rhetorical flourish in element E₂. In this period, as in so many others throughout the Naturalis historia, Pliny proves himself to be quite the skilled stylist.

Any attempt at a translation of 19.4–5 must begin, of course, with due consideration of the existing versions.Footnote ²⁶ Pliny was first translated into Italian (by Landino, published in 1476) and subsequently into French (first by du Pinet, 1562), Spanish (first by Hernández, completed in the 1560s but published only in 1966–76), English (first by Holland, 1601) and German (first by Denso, 1764–5):Footnote ²⁷ modern multivolume translations of the encyclopedia into English, French, German and Italian have likewise all been undertaken in the past several decades.Footnote ²⁸ I have relied on these versions (and others) for my own translation of the passage, laid out according to the clause structure analyzed above. Truth be told, no translation can do justice to the elegance, sophistication and force of the original, but, all the same, I do hope that this English rendering at least clarifies my understanding of the Latin in light of the various proposed emendations.