Bryn Mawr Classical Review 95.05.08

P.G. Walsh, Apuleius, The Golden Ass. Oxford: Clarendon Press, 1994. Pp. lv + 277. $55.00. ISBN 0-19-814932-8.

Reviewed by John L. Penwill, La Trobe University (

The lack of an accessible and accurate English version of Apuleius' Metamorphoses has long been felt by academics who offer courses in Roman literature in translation. Teaching the work via Graves's Penguin (even in Grant's 1990 revision) is a recipe for despair, while Lindsay's 1960 rendition is too distracting in its stylistic eccentricities and less than accurate paraphrases. By far the best recent translation is that of J.A. Hanson (1989), but that is buried in what has become a two-volume Loeb and consequently far too expensive to prescribe for undergraduate students. Thus OUP's decision to commission a new translation of Apuleius' novel by a scholar who has made a significant contribution to Apuleian studies is a welcome move, and one hopes that this version will quickly be issued as a paperback in the World's Classics series.

It is, I think, in its value as a teaching text rather than its appeal to the 'general reader' that this translation will be judged. The latter market will still find Graves's version more 'user friendly' and will not be bothered by the barriers it imposes against treating the text in a scholarly way. Walsh's version is generally more faithful to the Latin (with the qualifications outlined below), preserves Apuleius' division of the work into eleven books and gives the chapter numbers in the margin, thus making it far easier for students to use in conjunction with the critical literature. That W. has eschewed the practice that reached its nadir with Graves of arbitrarily dividing the work into chapters of the translator's invention is highly to be commended; it is as important for Apuleius' mutuus nexus as for Ovid's carmen perpetuum that the reader be allowed to experience the structure imposed on the narrative by the author's book-division. My one quibble in this regard is the heading The Tale of Cupid and Psyche that separates 4.27 from 4.28 and the fact that the old woman's narration from 4.28 to 6.24 lacks the quotation marks that are rightly employed for the tales of e.g. Aristomenes, Thelyphron and Charite's servant.

However, despite this advance in format, one cannot call the translation an unqualified success. It is not particularly easy to read, it tends towards the verbose, it has a number of stylistic infelicities and certain key passages are not rendered as accurately as they should be. The language does not flow, exhibiting an awkwardness that seems to derive from a concentration in the act of translation on individual words and phrases rather than sentences, and the choice of words often leaves something to be desired. Take this example from 10.3:

The young man did not defer his obedience to his sick parent's command. He made for her chamber with a brow puckered with an old man's concern, for this was an obedience which in a sense he owed to his father's wife and his mother's brother.... The young man for his part even at that moment had no suspicion that anything was amiss; with modest demeanour and without prompting he asked her the causes of her present illness.
The first sentence corresponds to Apuleius' nec adulescens aegrae parentis moratus imperium. 'Did not defer his obedience to' is too stilted a rendition of nec ... moratus, and loses in comparison with Hanson's 'responded unhesitatingly to' -- a version both more succinct and more accurate. 'With a brow puckered with an old man's concern' (senili tristitie striatam gerens frontem) in its awkward repetition of 'with' both does violence to Apuleius' taut grammatical structure and again lacks the felicity of Hanson's 'his brow furrowed with an old man's sadness'. And 'The young man for his part even at that moment...', while accurate in the lexical sense, is very laboured for at iuuenis etiam tunc. This tendency towards verbosity is a recurring feature of W.'s translation technique, and shows in the total word-count: at a very rough calculation, W.'s version comprises 110,000 words as opposed to 87,000 for Hanson and 98,000 for Graves.

Inappropriate and anachronistic elements are also an irritant. At the end of Book 5, for example, we read 'This was how the two goddesses sucked up to Cupid, seeking to win his favour, though he was absent, by taking his part, for they feared his arrows.' Typically the main clause comes first and is followed by a set of accretions linked to it by conjunctions and prepositions, making an awkward 28-word sentence out of Apuleius' tightly constructed 10. And to translate the main verb blandiebantur as 'sucked up to' gives the passage an aura of school slang which is hardly in keeping with the overall tone. There is a similar lack of decorum in W.'s rendering of the priest's speech at 11.15, where the impressive rhetoric is suddenly deflated with an almost petulant 'So she [Fortune] can now head off...'; surely the most appropriate rendering for the eat nunc that begins this sentence is the literal one: 'Let her now go...'. In a different vein, Fotis sounds anything but the sultry sex-kitten when she says to Lucius in W.'s rendering of 2.7's quae dulce condiens et ollam et lectulum suaue quatere noui, 'The spices which I incorporate are sweet. I'm an expert in pleasurably shaking a bed as well as a pot.' No hint of Catullus' quassa lecti argutatio here! (The sex scenes are generally coyly done and are not one of W.'s strong points -- cf. esp. 2.16-17.) Of the more glaring anachronisms one can cite the appearance of Peter Pan in Juno's speech at 5.31 (doubly incongruous in a context where we have just encountered the real Pan at 5.25), Bridewell for Tullianum as the place of incarceration for the eunuch priests at 9.10 (since this requires a note to explain the joke anyway, it would surely have been better to leave 'Tullianum' in the text as with meta Murciae at 6.8 rather than introduce this out-of-place English concept) and the hour of Prime at 11.20 (suggesting that we have somehow been transported into a medieval monastery).

But these are irritants, not serious errors, and it is by its accurate rather than its mellifluous rendition of the Latin that a translation of this kind should ultimately be judged. The translator of the Metamorphoses does not have long to wait before his or her ability is tested to the limit; a competent negotiation of the syntactical and narratological minefield of 1.1 is essential in order to allow the reader to interact appropriately with this enigmatic text, and thus requires great care. It is unfortunate, therefore, that W. does not do all that good a job with it. The first words are at ego tibi. In his note on 1.1 (p. 241), W. quotes with approval Hanson's own note on this opening phrase: 'The work opens as if in the middle of a literary discussion.' But there is no hint of this aura of debate in W.'s translation of it. Instead of the emphatic at ego, 'But I', we find a diffident 'What I should like to do is...', a circumlocution that is both inaccurate and symptomatic (cf. 'what she said was' for ait at 1.13 and 'this is how she began' for incipit at 4.27). There is nothing in the Latin that corresponds to these words, any more than to the 'I want you to' at the beginning of W.'s second sentence. Further problems arise as we proceed. Exordior does not mean 'So let me begin', quis ille? does not mean 'Who is the narrator?' -- that is interpretation, not translation -- and paucis accipe does not mean 'Let me briefly explain'. The bald juxtaposition of first, third and second person vanishes in this expansive and obfuscating rendition, just like the opening ego tibi. In this post-Calvino era any serious discussion of 1.1 and of the work as a whole must include the question of the relation between narrator and reader, and to be genuinely useful a translation must give a text on which this can be based. We thus require a more exact and exacting rendition of en ecce praefamur ueniam siquid exotici ac forensis sermonis rudis locutor offendero than 'So at the outset I beg your indulgence for any mistakes which I make as a novice in the foreign language in use at the Roman bar' (for an alternative to W. cf. Winkler [1985], 181; Penwill [1990], 225); and in the next sentence the concluding clause ('for the romance on which I am embarking is adapted from the Greek') imports concepts and logical connections that simply aren't there: the Latin fabulam Graecanicam incipimus has no conjunction linking it to the previous clause, and no words corresponding to either 'romance' or 'adapted'. And finally 'Give it your attention, dear reader, and it will delight you' is far too verbose for the pithy lector intende: laetaberis and displays that disregard for subject and object that we have already encountered in paucis accipe and ut mireris.

Some of these criticisms may appear to be carping and hair-splitting, but it is important that translations of texts aiming at the academic market should render them accurately. One does not want to have to spend time pointing out that there is an ambiguity here or a significant form of expression there that the translator has seen fit to recast; the text must as far as possible be able to speak for itself. There is no 'dear' reader in 1.1; that is a 19th century import. There is a lector optime at 10.2, rendered (again in 19th century fashion) as 'gentle reader'. And there was a studiose lector at 11.23, but he has been edited out: 'Perhaps the reader's interest is roused, and you are keen to inquire' is not a legitimate rendering of quaeras forsitan satis anxie, studiose lector; in order to include this as one of the narrator's addresses to the reader, one will be forced to explain that this is an erroneous translation.

Finally a few words about the introduction and the explanatory notes. The introduction canvasses what is known about the life and other writings of Apuleius, goes on to offer some interpretive comment on the Metamorphoses, and concludes with short sections on the manuscript tradition, Nachleben and other translations followed by a reasonably full bibliography. That I disagree with much of the interpretive comment (particularly on 'Cupid and Psyche') should surprise no-one who has read my own writings on this novel, but that is a matter for academic debate outside the province of this review. Suffice it to say that it is disappointing to find the 'Augustine fallacy' (the idea that the work is -- must be -- in some sense autobiographical, with Apuleius lurking behind the mask of Lucius) and the 'Platonist fallacy' (Apuleius was a Platonist, Apuleius wrote the Metamorphoses, therefore the Metamorphoses must be in some sense a Platonist tract) still very much in vogue, and that while the seminal work of Winkler receives honourable mention, it appears to have had very little influence in W.'s thinking. It accords with this conservative critical stance that the Nachleben section ends in the 19th century, ignoring the novel's 20th century avatars in the work of authors such as Kafka, Bulgakov and Nabokov. The notes are largely explanatory rather than interpretive (as they should be) and are useful sources of information about the meaning of names and mythological allusions; the only error I have spotted is 'Hadrian' instead of 'Trajan' in the note on 7.6. There are likewise useful remarks in both introduction and notes on intertextual relationships with earlier Roman authors, though that with Ovid's Metamorphoses is underplayed. Giving Apuleius' work what W. acknowledges to be its proper title (Introduction p. xix) would help to highlight the fact that this is a (and in my view the) significant relationship.

The translation, though, is the thing; and in spite of the fact that my comments have been largely critical, I will conclude by saying that provided it is published in an affordable format this is without doubt the translation I would prescribe for students studying the work in English. 1.1 will always present massive problems for the Latinless reader (and not just for the Latinless reader!), and one can do far more with W.'s version than with Graves. It would provide a welcome incentive to reinstate an important European author in the courses that I teach.


  • Graves, R., tr. (1990). The Golden Ass by Lucius Apuleius, revised with an introduction by Michael Grant. Harmondsworth.
  • Hanson, J.A., tr. (1989). Apuleius: Metamorphoses, 2 vols. Cambridge Mass. and London.
  • Penwill, J.L. (1990). 'Ambages Reciprocae: Reviewing Apuleius' Metamorphoses' in The Imperial Muse: Flavian Epicist to Claudian, ed. A.J. Boyle, 211-35. Bendigo.
  • Winkler, J.J. (1985). Auctor & Actor: A Narratological Reading of Apuleius's The Golden Ass. Berkeley and Los Angeles.