In Silv. 4.4 Statius pays homage to Vitorius Marcellus,Footnote ¹ the young dedicatee of the poem (for this small but meaningful detail, see Stat. Silv. 4.4.45 iuuenes … annos; 4.4.74 iuuenemque … parentem), praising his skills as an orator (4.4.39–45) and foreseeing a brilliant military career for him (4.4.61–4). The last point is highlighted in a brief portrait of Marcellus as a perfect foot soldier and horseman (4.4.64–9):Footnote ²

The whole passage is clear, except for the adverb tarde in line 66. In recent times, apart from Shackleton Bailey (who, however, expresses his doubts in a footnote to his translation), all critical editors and commentators of StatiusFootnote ³ agree that the reading of manuscript M is barely defensible. This suspicion dates long back, and its reason is evident: having been said that Vitorius’ limbs (membra) are suitable for military employment (accommoda bellis), Statius’ statement that Marcellus’ arms are unable to fit his breastplate properly would be self-contradictory.

Emendation therefore seems to be needed. Barth's proposal haud tarde reveals its weakness since in Statius haud never modifies this adverb,Footnote ⁴ while Goodyear's subeant apte Footnote ⁵ and Liberman's subito (subeant) and propere (subeant) can sound better than recte (subeant) or dextre (subeant), which were proposed by Courtney,Footnote ⁶ but they seem to be mere attempts to emend a much-disputed locus criticus rather than decisive conjectures which settle the problem.

However, another approach to emendation has been proposed, and I think that this turns out to be more convincing. It consists in replacing tarde not with an adverb but with an adjective suitable for Marcellus’ arms. In this way the hexameter would perfectly match the structure of a golden line: the verb subeant will be embedded between a pair of adjectives (grauem + the unknown adjective to be referred to lacerti) and a subsequent pair of nouns (thoraca lacerti).Footnote ⁷ Even on the basis of replacing tarde with an adjective, the list of conjectures is quite long:Footnote ⁸ Markland's ualidi (preferred by him to fortes), Otto's faciles, Phillimore's artandi,Footnote ⁹ Coleman's ingentes and Watt's crudi may deserve consideration in terms of sense, but they do not account well for the genesis of the error (LibermanFootnote ¹⁰ points out that, as regards palaeographical evidence, Watt's proposal is the most likely, but I wonder how crudi could have been corrupted into tarde).

Though not totally convincing, Otto's proposal sets up an intriguing and well-balanced contrast (a ‘wohlberechnete[r] Gegensatz’, as he put it) between an adjective that conveys the sense of weight (grauem) and an adjective that points to the agility of Marcellus’ arms (faciles). Ιf this strategy sounds convincing, perhaps a more suitable conjecture can be lenti. This adjective is frequently used with reference to the physique of young people in the sense of ‘supple, but at the same time tough’: for the best examples, see the description of Leander's arms cleaving the water in Ov. Her. 18.58 iactabam liquido bracchia lenta mari Footnote ¹¹ and Her. 19.48 lentaque dimotis bracchia iactat aquis. Note also that lenti … lacerti might be interpreted as the Latin rendering of the Homeric formula describing the suppleness of young limbs: see, for example, Hom. Il. 11.668–9 οὐ γὰρ ἐμὴ ἲς | ἔσθ’ οἵη πάρος ἔσκεν ἐνὶ γναμπτοῖσι μέλεσσιν (‘since my strength is no longer the strength I possessed in my pliant limbs’; here we have old Nestor complaining for his diminished strength); Od. 11.393–4 ἀλλ’ οὐ γάρ οἱ ἔτ’ ἦν ἲς ἔμπεδος οὐδ’ ἔτι κῖκυς, | οἵη περ πάρος ἔσκεν ἐνὶ γναμπτοῖσι μέλεσσι (the description of Agamemnon's ghost); Od. 21.281–3 ὄφρα μεθ’ ὑμῖν | χειρῶν καὶ σθένεος πειρήσομαι, ἤ μοι ἔτ’ ἐστὶν | ἴς, οἵη πάρος ἔσκεν ἐνὶ γναμπτοῖσι μέλεσσιν (Odysseus showing his youthful strength).Footnote ¹²

The proposed correction has also another significant advantage over other attempts, since it reasonably accounts for the genesis of tarde. There are at least two possibilities: the first and most likely is that lenti was glossed with the adjective tardi, further corrupted into tarde. Alternatively, a passage from lenti to lente is not difficult (especially in a late antique manuscript written in majuscule writing, where the two letters are quite similar); after lente substituted the original reading, a scribe may have altered the corrupt adverb into its synonym tarde in order to give it better sense. Whichever explanation we prefer, I argue that the original reading was lenti.