Zimmerman's edition of this extraordinary novel (Oxford 2012) offers a well-constituted text based on consistently reliable editorial judgement, together with a substantial and informative preface and a full bibliography. But the last critical word has never been said, and a study of Book I has thrown up the following selection of passages which it is suggested may deserve further editorial attention.
1.1.4 linguam Atthidem:
an early correction of linguam atidem, the reading of Laurentianus 68.2, the sole authority for the text of Met. The hybrid form of the adjective is not in itself grounds for suspicion: witness Atthidum … choris at Sen. Phaedra 106; but –em as the termination of a Greek feminine adjective is surely odd, the more so as supposedly penned by a native Greek speaker, who received part of his education at Hymettos Attica. It seems surprising that no one has, it appears, suggested reading Atthida.
1.7.3 illuuiem.
This necessary correction of F's eluuiem is ascribed to Oudendorp. In fact he reported it from Price's note, where it is printed without comment; his text reads eluuiem. This is one of a number of places where corrections that he has merely reported from earlier editions are credited to Oudendorp.
1.9.3 officiosis roncis raucus appellat.
Apuleius is likely to have written rhonchis, as in Greek ῥέγκος. No Latin writer was more attentive to how his words would sound when read aloud. (For the word-play with raucus cf. Lara Nicolini, Ad (l)usum lectoris: etimologia e giochi di parole in Apuleio, Bologna 2011, p. 54 n.139.)
1.10.1 quae cum subinde ac multi nocerentur.
Various supplements have been proposed to cure the broken-backed syntax, but my concern is with multi nocerentur, which has also troubled critics. It is true that noceor passive as if from a transitive verb is attested as early as Vitruvius 2.9.14 larix … non solum ab suci uehementi amaritate ab carie aut tinea non nocetur, suggesting that perhaps by Apuleius' time it was current colloquial usage. Be that as it may there is an alternative reading that is nearer to the ductus than any of the emendations reported by Zimmerman and has impeccable classical credentials: multis noceretur, ‘harm was being done to many’. This impersonal use of noceor is twice attested in Livy, at. 3.55.8 hac lege iuris interpretes negant quemquam sacrosanctum esse, sed eum qui eorum cui nocuerit sacrum sanciri; itaque aedilem prendi ducique a maioribus magistratibus, quod, etsi non iure fiat – noceri enim ei cui hac lege non liceat – tamen argumentum esse non haberi pro sacrosancto aedilem; and 5.14.7 nulla incolumi re cui ferro aut igni noceri posset, (this latter misleadingly classified by OLD s.v. 1b).
1.14.2 destinatae iam cruci candidatus.
The phrase has been suffered to pass unchallenged, but why cruci and not crucis, the usual case for what is (here ironically) ‘sought’. Did Apuleius write crucis, ‘a candidate for the waiting cross’?
1.18.1 manticam meam umero exuo:
an early correction of F's mantica mea; Helm compared Ov. Met. 2.419 exuit hic umero pharetram. A plausible alternative is to keep F's text and read mantica mea umero exuta, the change of exuta to exuo being due to umero preceding. This seems to me at least as plausible as Hildebrand's mantica mea umeros exuo, and it has the additional advantage that it eliminates the asyndeton, so picking up cum dicto and bringing out the simultaneity of words and action.
1.19.7 haud ita longe radices platani lenis fluuius … ignauus ibat.
Prepositional longe takes a bit of swallowing; the obvious sense is ‘not very far away’, as at 1.15.3, 6.18.1, 8.21.1 non longe, Caes. BAfr., 4.8.2 haud ita longe. That being so, the easiest way out is to supply a transitive verb to govern radices. Better, I suggest, than Helm's tentative praeteribat and more plausibly palaeographically is subibat, sub having been omitted after –us preceding. Though we now know that it is not what Ovid wrote, it is more likely than not that Apuleius' text of the Ars Amatoria read subitur at 1.620.
1.26.6 ac nequiquam, defectum iam, incerta uerborum salebra balbutire.
The editorial commas (Martos an exception) merely serve to bring out the awkwardness of the phrasing, so unlike the usual easy flow of Apuleius' narrative. Did he write nequiquam iam defectum, the iam omitted by haplography and then reinserted in the wrong place? That sequence is a well-attested cause of textual error.
