Constructing Apuleius: the emergence of a literary artist.

This paper aims to be both a contribution to the history of scholarship and a stimulus to further research. In it I seek to follow some key themes in the scholarly reception of Apuleius' Metamorphoses before and after the work of Ben Edwin Perry, and to show how the critical constructions of literary works are necessarily affected by contemporary ideological prejudices, which change over time as scholarship develops. In particular, I try to trace the emergence after much negative judgement of the modern construction of Apuleius as a careful literary artist, and of the Metamorphoses as a novel worthy of detailed study and a carefully composed, complex and highly allusive literary text, a view which has come into being almost entirely during the twentieth century, and to suggest consequent paths for further scholarly investigation. (1)

1: The Problems of Prejudice--from Antiquity to Perry Anxiety about how ancient prose fiction relates to the conventional canon of literary genres, about its overall quality, and about whether it can be accounted 'proper' literature, goes back to the Roman world. (2) Prose fiction is notably omitted from the most extensive survey of literary genres to be found in ancient literary criticism, the syllabus for the aspiring orator in the tenth book of Quintilian's Institutio, and novels are largely unmentioned in the other literature of the ancient world until the fourth to fifth centuries. Then we find two revealing comments. In his Neoplatonising commentary on the Somnium Scipionis of Cicero, Macrobius condemns prose fiction as merely titillating and more suitable for the nursery than for serious consideration (Somn.1.2.8):

   Auditum mulcent ... argumenta fictis
   casibus amatorum referta, quibus
   vel multum se Arbiter exercuit vel
   Apuleium non nunquam lusisse miramur.
   Hoc totum fabularum genus, quod
   solas aurium delicias profitetur,
   e sacrario suo in nutricum cunas
   sapientiae tractatus eliminat.

   'Our hearing is charmed by plots
   stuffed with the imagined vicissitudes
   of lovers, on which Petronius spent
   much labour, and in which we are
   surprised that even Apuleius often
   sported. This whole kind of story,
   since it aims only at the pleasuring
   of the ears, is expelled by the discussion
   of wisdom from its shrine to take
   refuge in the cradles of nurses'.

Here prose fiction and its trivial concerns are seen as mere empty and false story-telling, insignificant compared to the truth of philosophy which is central for Macrobius himself; Apuleius' combination of Platonic philosophy and novel-writing is seen as surprising and inconsistent. A similar attitude is shown in the supposed attack by Septimius Severus in a letter to the Senate on the character of Clodius Albinus, his rival for the purple in 195--7, in the Historia Augusta (SHA Clod. Alb.12.12):

   Maior fuit dolor, quod illum pro
   litterato laudandum plerique duxistis,
   cum ille neniis quibusdam anilibus
   occupatus inter Milesias Punicas
   Apulei sui et ludicra litteraria consenesceret.

'It is a greater pain to me that many of you have deemed him to deserve praise as a man of literature, when he was busied with some nonsense or other fit for old women and was growing senile amongst the Punic Milesian tales of his friend Apuleius (3) and such literary trivialities'.

This association of novels with old women, suggesting the credulity of their readers and the triviality of their content, is likely to be of earlier origin, since it is seems to be playfully alluded to in Apuleius' Metamorphoses, when the inserted tale of Cupid and Psyche, which is clearly designed to parallel the plot of a Greek romantic novel, is presented as being told by a real old woman, the anonymous housekeeper of a robber-band, who herself introduces her narrative as an old woman's tale in a phrase which seems to pick up the kind of criticisms made by Macrobius and Severus' supposed letter (4.27.8): sed ego te narrationibus lepidis anilibusque fabulis protinus avocabo, 'but I will distract you at once with elegant narratives and with the stories of an old woman'.

Allegory, a mode of analysis which discovers seriousness under apparent frivolity, has naturally been perennially popular in the criticism of Apuleius' Metamorphoses, which unlike Petronius' Satyrica offers clear opportunities for figurative interpretation. The most famous analysis of Apuleius from the ancient world, the reading of the story of Cupid and Psyche by Fulgentius, from the late fifth or early sixth century, (4) offers an elaborate allegory (Myth. 3.6), from which I select a key portion:

   Civitatem posuerunt quasi in modum
   mundi, in qua regem et reginam velut
   deum et materiam posuerunt. Quibus
   tres filias addunt, id est carnem,
   ultronietatem quam libertatem arbitrii
   dicimus et animam; Psice enim
   Graece anima dicitur ... huic invidet
   Venus quasi libido; ad quam perdendam
   cupiditatem mittit; sed quia cupiditas
   est boni, est mali, cupiditas
   animam diligit et ei velut in
   coniunctione miscetur ...

   'They [i.e. Apuleius and his lost
   Greek imitator Aristophontes of Athens]
   have put Psyche's city to represent
   the universe, in which they have put a
   king and a queen, standing for God and
   matter. To these they add three
   daughters, that is the flesh, spontaneity
   (which we call freedom of will)
   and the soul, for the soul is called
   'psyche' in Greek ... She is envied by
   Venus, representing lust, who sends Desire
   to destroy her; but since Desire
   is for both good and bad, Desire loves
   the soul and mixes with it in a
   form of union'

The details of the allegory are not wholly clear, but it is plain that we are dealing with a serious moralising interpretation in a familiar ancient tradition, aimed at uncovering the real truth under the apparent falsity and levity of fiction. (5)

The Renaissance like late antiquity felt that the apparent triviality of Apuleian fiction needed explanation and justification, by allegory if necessary. The elder Filippo Beroaldo, the most distinguished Renaissance editor of Apuleius, clearly addressed this issue in the introduction to his 1500 text of and commentary on the Metamorphoses, (6) where he discusses the writer's intention and plan (scriptoris intentio atque consilium):

   Ego Apuleium quidem nostrum confirmo
   Lucianum Graecum scriptorem
   argumento consimili imitari, verum
   sub hoc transmutationis involucro
   naturam mortalium et mores humanos
   quasi transeunter designare
   voluisse, ut admoneretur ex hominibus
   asinos fieri: quando voluptatibus
   beluinis immersi asinali stoliditate
   brutescimus nec ulla rationis virtutisque
   scintilla in nobis elucescit.... rursus
   ex asino in hominem reformatio
   significat calcatis voluptatibus
   exutisque corporalibus deliciis rationem
   respicere et hominem interiorem qui
   verus est homo ex ergastulo
   illo caenoso ad lucidum habitaculum
   virtute et religione ducibus migrasse
   ... illa vero eruditioribus principalis
   huiusce transformationis
   causa valdeque probabilis videri
   potest, ut videlicet sub hoc mystico
   praetextu Apuleius noster Pythagoricae
   Platonicaeque philosophiae
   consultissimo dogmata utriusque
   doctoris ostenderet et sub hac ludicra
   narratione palingenesiam et
   metempsychosim, id est regenerationem
   transmutationemque, dissimulanter afferret ...

   'I for my part maintain that our
   Apuleius imitated the Greek writer
   Lucian by means of a similar plot,
   but wished under the cover of this
   story of metamorphosis to mark out as
   if in passing the nature of man
   and human character, so that a warning
   should be given that asses can be
   made of men: since when submerged
   in bestial pleasures we become like
   brutes with the stupidity of the ass,
   and no spark or reason or virtue
   shines out in us ... Again, the
   retransformation from ass to man signifies
   the spurning of pleasures and the
   stripping off of fleshly delights so as to
   take note of reason, and that the
   inner man, which is the true man, has
   passed from that foul prison to a habitation
   of light through virtue and religion....
   to the more learned the following
   can indeed seem the principal
   cause of this transformation and
   an extremely probable one, namely so
   that under this mystic pretext, our
   Apuleius, an expert in Pythagorean
   and Platonic philosophy, might demonstrate
   the teachings of both masters
   and introduce in a concealed way under
   the cover of this frivolous
   narrative palingenesis and metempsychosis,
   that is rebirth and transformation....' (7)

It was in the Renaissance, too, that Apuleius' novel came under attack from a different angle, that of its style. Though there is some ancient evidence for an African provincial accent in the speaking of Latin, (8) and Isidore in the seventh century could claim that Latin had been corrupted by contact with barbarian languages in the expansion of the Roman Empire (Orig. 9.1.7), it is to the Renaissance that we owe the argument that provinces generated their own debased regional dialects of Latin, and the invention of 'African Latin' in particular, a topic that was to be the subject of particularly heated scholarly debate in the late nineteenth and earlier twentieth centuries. (9) The cult of Ciceronian style amongst many humanists (10) led to the denigration of other Latin styles and to the categorisation in particular of post-Ciceronian Latin as decadent, (11) though Beroaldus and several other humanists prized and defended the exuberant and non-classical diction of Apuleius. (12) A letter of Erasmus written some twenty years after Beroaldus' edition (Ep. 1334, from 1523) clearly uses this rhetoric against Apuleius, characterising his style along with that of his fellow-African Tertullian as studied, affected and obscure, showing provincial over-striving for stylistic effect:

   Mihi veterum dictionem variam consideranti
   videtur vix ullos provinciales
   feliciter reddidisse Romani sermonis
   simplicitatem praeter aliquot,
   qui Romae a pueris sunt educati.
   Nam et Tertulliano et Apuleio suus
   quidam est character et in decretis
   Afrorum, quae multa refert Augustinus
   contra Petilianum et Crescentium,
   deprehendas anxiam affectationem
   eloquentiae, sed sic, ut Afros agnoscas.

   On considering the varying styles
   of the ancients, it seems to me that
   hardly any writers of provincial origin
   have successfully rendered the purity
   of the Roman language, except for
   those who were educated at
   Rome from their youth. For Tertullian
   and Apuleius have their own particular
   stylistic stamp; in the decrees written
   by Africans, too, which are
   cited in abundance by Augustine writing
   against Petilianus and Crescentius,
   you will find an anxious affectation
   of style, but such as to enable
   you to recognise them as Africans.

Some of Erasmus' great contemporaries put it more succinctly, claiming that the style of Apuleius' novel was as asinine as his hero's metamorphosis. Thus Melanchthon (Eloquentiae encomium 29): recte Apuleius, qui cum asinum repraesentaret, rudere quam loqui mallet, 'rightly did Apuleius in depicting an ass choose to bray rather than speak'. Vives (De tradendis disciplinis 1.3), picking up Melanchthon's gibe, differentiated between the unacceptable style of the Metamorphoses and the more restrained style of some of the other works, thinking no doubt of the more Ciceronian Apologia: Apuleius in asino plane rudit, in aliis sonat hominem, 'Apuleius in his ass clearly brays, in his other works he sounds a human note'.

The already established twin criticisms of the Metamorphoses, frivolity of subject-matter and barbarism of style, are deployed in the most influential work of literary criticism on the novel before the nineteenth century. (13) The Traite des Romans et leurs origines (1670) of Pierre-Daniel Huet, tutor of the Dauphin (son of Louis XIV) and originator of the Delphine Classics, translator of Daphnis and Chloe and Bishop of Avranches, concludes by arguing that novels can be defended as serious literature if they have a didactic underlying purpose and are excellent in material and style:

   'Now Fictions being nothing but
   narrations, true in appearance and false
   in effect, the minds of the simple,
   who discern only the bark, are pleased
   with this show of truth, and very well
   satisfied. But those who penetrate
   further, and see into the solid, are
   easily disgusted with this falsity, so
   that the first love the falsehood,
   because it is concealed under an appearance
   of truth, and these others are distasted
   with this Image of truth, by
   reason of the real falsehood which is
   couched under it, if this falsehood
   is not otherwise ingenious, mysterious
   and instructive, and buoys itself
   up by the excellence of the invention and art.' (14)

At the beginning of his work, Huet had defined the key purpose of novels as moral didacticism, and it is interesting to see literary excellence achieving almost equal prominence by the end in the passage just cited. The opening is uncompromising:

   'The chief end of a Romance, or (at
   least) that which ought to be, and
   which the Composer ought to propose
   to himself, is the Instruction of the
   Reader, to whom he must always present
   Vertue crowned, and Vice punished.' (15)

Literary entertainment and pleasant style is then justified in a Lucretian-style argument that literary honey is needed to sweeten the moral cup owing to human pride:

   'But as the spirit of man naturally
   hates to be taught, and self-love does
   spurn against Instructions, 'tis to
   be deceived by the blandishments of
   pleasure, and the severity of Precepts
   to be sweetened by the agreement
   of Examples.' (16)

But for Huet neither Petronius nor Apuleius, though not without their merits, provided attractive style or fitting didactic content. Petronius contained 'agreeable and ingenious fictions, but very often too wanton and immodest,' (17) and his style was too affected: 'he is somewhat too much Painted and Studied, and degenerates from that natural and majestic simplicity of the happy age of Augustus.' (18) Apuleius' novel was 'an Italian fiction very divertising and full of Wit,' (19) but 'he has not at all retrencht the smuttiness which was in the Originals which he had followed' and 'his style is that of a Sophist, full of affectation and violent figures, hard, barbarous and befitting an African.' (20)

That this low view of Apuleian Latinity was alive and well a century later is shown by David Ruhnken's preface to Oudendorp's (posthumous) edition of 1786.21 Ruhnken repeats the familiar charge of barbarous style, but grounds his accusation more precisely in the wholesale appropriation of archaic language in the Metamorphoses when Apuleius had the 'better' example of Cicero before him. He argues that the Antonine writers Gellius and Apuleius (the palimpsest of Fronto was not to be discovered for another generation) could indeed write like Cicero and other classical authors but chose to adulterate this style with excessively archaic vocabulary (I--II):

   Non ii quidem optimos illos omnino
   reliquerunt, sed tamen cum eorum
   imitatione scriptores ex ultima
   antiquitate repetitos coniunxerunt, ut
   modo cum Cicerone, Caesare, Livio
   et similibus, modo cum Evandri matre
   loqui viderentur.

   'For their part these writers did
   not leave aside these excellent models,
   but yet combined with imitation of
   these the use of writers sought out
   from the most distant antiquity, so
   that sometimes they seemed to be
   conversing with Cicero, Caesar, Livy
   and the like, sometimes with the
   mother of Evander'.

But in this deliberate lexical obscurantism, Ruhnken continues, Apuleius had done his linguistic research and was largely echoing archaic usage as opposed to rankly inventing terms (II--III):

   Is igitur tantum abfuit, ut sibi in hoc
   genere temperaret, ut potius e casca
   vetustate eam orationem conflaret,
   quam nemo, nisi qui multum temporis
   in ea Latinitate cognoscenda contriverit,
   sese sperat assecuturum. Scio
   viros eruditos esse, qui non omnia huius
   modi verba ab antiquis scriptoribus
   sumpta, sed temere et pro libidine
   conficta putent ... sed ego
   libentius sequar Oudendorpium ...
   bene iudicantem, nihil Appuleium
   sine exemplo scripsisse ...

   'He was so far from restraining himself
   in this sphere that he preferred to
   throw together that style of his from
   primitive antiquity, that no-one can
   hope to follow unless he has spent a
   great deal of time in learning that
   kind of Latin. I know that there are
   scholars who believe that all his
   words of this kind are not taken from
   ancient writers but are wilfully and
   capriciously invented ... but I would
   prefer to follow Oudendorp's sound
   view that Apuleius wrote nothing without a model'.

This partial defence of exquisite lexical archaism interestingly echoes the programmatic statements of Fronto, who was (as already noted above) yet to be rediscovered.22 But Ruhnken's relative moderation here does not detract from his overall negative judgement on Apuleius' archaism, which is followed by an uncompromising attack on his 'swollen' style as a whole (IV):

   Est sane ista antiquitatis affectatio
   molesta Appuleium legentibus. Nescio
   tamen, an molestior sit tumor
   Africanus, quo orationem, in iis
   quidem libris, quos a doctis legi
   voluit, praeter modum et pudorem inflavit.

   'That affectation of archaism is
   irritating enough for Apuleius' readers.
   But even more irritating may be the
   African tumidity, with which he in
   flated his style beyond the limits
   of shame, at least in those books which
   he wished men of learning to read.'

Here Ruhnken seems to be referring to issues of syntax rather than vocabulary, to the deliberate and exuberant accumulations of effects of sound, balance and sentence-structure which were later to be so notably attacked by Norden (see below).

Half a century later in the preface to his important edition of 1842, (23) Hildebrand could only assent to Ruhnken's judgement on Apuleius' style in the Metamorphoses, pointing again to its 'African' exuberance and the contrast with the more Ciceronian works (xxiv):

   Mihi vero accuratius libros perlegenti
   haec statim quaestio sese obtulit,
   cur in Metamorphosi auctor turgidum
   plane et conquisitum et ad taedium
   usque luxurians genus dicendi
   affectasset: qua quidem verborum
   copia nimiisque eloquendi ac fucatis
   illis pigmentis tanto lectorem fastidio
   affecit, ut nisi animus vivida imaginandi
   et speciosa describendi alliceretur
   ratione, librum statim e manibus
   deponeremus. Quodsi Florida
   praecipue tamen Apologiam ceterosque
   Apuleii libros accuratissime perlectos
   cum Metamorphosi contuleris, unius
   eiusdemque esse scriptoris
   omnia vix tibi persuaseris.

   'As I read carefully through the
   works this question presented itself to
   me immediately--why the author in
   the Metamorphoses had affected a
   style which was plainly swollen
   and recherche, and exuberant to the
   point of over-satiety. He affects
   the reader with such disgust with his
   abundance of words and his excessive
   and artificial colours of speech,
   that were it not that our mind is
   attracted by his vivid powers of imagination
   and his decorative manner of
   description, we would put down the
   book at once. But if you compare the
   Florida and especially the Apologia
   and the other works of Apuleius with
   the Metamorphoses after the
   closest of readings, you will have
   difficulty in convincing yourself that
   all these works are by one and the same author.'

Hildebrand attributed this exuberance again to Apuleius' African origin and suggests that the Metamorphoses were composed early in his career at Rome, a view which has recently been notably revived (xxv): (24)

   Omnes enim Afros constat antiqua quaeque
   affectasse.... Si vero Apuleium
   saepius quam ceteros Afros hanc
   antiquitatem plane obsoletam
   sectatum fuisse apparet, minus dedita
   opera quaesitae affectationi hoc
   tribuo, qua sui temporis perversitatem
   corruptumque iudicium perstringere
   voluerit, quam iuvenili eius ingenio
   et quae in adulescentulis saepius
   observari potest, verborum inusitatorum
   captationi, qua orationem
   vividiorem reddi et exornari putant.

   'For it is generally agreed that all
   the African writers affected whatever
   was ancient ... but if it clearly
   emerges that Apuleius pursued this
   wholly outdated archaism more often
   than other African writers, on reflection
   I attribute this less to a search
   for affectation and more to his
   youthful spirit and to something which
   is often observable in young men,
   the taking up of unusual words, with
   which they think that their style is
   made more vivid and ornate.'

But it was in the matter of allegory, untouched by Oudendorp, that Hildebrand mounted a defence of Apuleius. In a discussion of some ten pages he generates a Platonising allegory which is recognisably based on Fulgentius, of which I cite his outline summary (p. xxxviii):

   Psychem castam intelligo atque
   pudicam animam, qualem a summo numine
   profectam Plato iam disseruit. Cupido
   sive Eros coelestis ille est et
   sanctus amor qui pudicae et purae
   insitus animae a natura est, unde eius
   nominatur maritus, quia artissimo
   naturae vinculo cum Psyche coniunctus
   est.

   'I understand Psyche to be the chaste
   and innocent soul, in the form that
   set out from the supreme deity according
   to Plato. Cupid or Eros is that
   heavenly and holy love which is
   naturally inherent in the chaste and pure
   soul, and is hence called her husband,
   since he is joined with Psyche in
   the closest of nature's bonds.'

In the second half of the nineteenth century, with the rise of the concept of 'African' Latin, anti-Apuleian scholars tend to focus on the issue of outlandish language rather than frivolous content. The most notable treatment is that of Norden in Die Antike Kunstprosa (1898), who like Huet before him emphasised the sophistic element in Apuleius. Norden, writing with youthful polemic in his first major book, opposed the then fashionable idea of African Latin, but delivered an attack on Apuleian Latinity which held no punches:

   Bei ihm feiert der in bacchantischen
   Taumel dahinrasende, wie einer
   wilder Strom sich selbst ubersturzende,
   in ein wogendes Nebelmeer wuster
   Phantastik zergehende Stil seine Orgien;
   hier paart sich mit dem ungeheuerlichsten
   Schwulst der affektierteste
   Zierlichkeit; alle die Matzchen,
   die dem weichlichsten Wohlklang dienen,
   werden in der verschwenderischsten
   Weise angebracht, als da sind Alliterationen, Ohren
   und Augen verwirrende Wortspiele,
   abzirkelte Satzteilchen mit genauester
   Korresponsion bis auf den Silbenzahl
   und mit klingelndem Gleichklang
   am Ende. Die romische Sprache, die
   ernste wurdige Matrone, ist
   zum prostibulum geworden, die
   Sprache des lupanar hat ihre castitas
   ausgezogen.

   'In him the style which rages away
   in Bacchic ecstasy, which rushes
   along like a wild torrent, which vanishes
   into a misty surging sea of depraved
   imagination, celebrates its
   debaucheries; here the most affected
   delicacy is coupled with the most
   monstrous bombast; all the tricks
   which serve to produce the most feeble
   euphony are applied in the most
   extravagant manner, such as alliterations,
   word-plays which bewilder eye
   and ear, precisely-shaped clauses
   with the most exact correspondences,
   even as far as the number of syllables,
   and with a jingling harmony at the
   end. The Roman language, a serious
   and worthy matron, has become a
   prostitute; the language of the brothel
   has stripped away her chastity.' (25)

In the nineteenth century, where the modern novel was so central to general literary output, appreciation of this ancient novelist was inevitably linked with critical battles over the novel more generally, battles not entirely vanished by the day of Ben Edwin Perry. (26) In the essay 'The Art of Fiction' of 1884, the earliest of his contributions to the criticism of the novel, Henry James suggests that it is only recently that the novel had become a respectable subject of literary criticism, (27) and indeed the novel had become increasingly socially, morally and intellectually acceptable since Jane Austen's determined defence of it under fire in the fifth chapter of Northanger Abbey (1818, written 1798), a development which is well traced by Trollope in his Autobiography of 1883. (28) For his critics in the early twentieth century, Apuleius could thus be received as a late and non-canonical writer of barbarous style, best known for his work in a genre which was barely intellectually respectable.

This was the context in which Rudolf Helm entered on his project of editing all the (non-philosophical) works of Apuleius, which occupied him for the first decade of the twentieth century. (29) Helm was a junior colleague of Eduard Norden in Berlin for the whole of that time, but clearly did not share Norden's disdain for Apuleian style. In the 1910 Latin preface to the Florida which gathers Helm's views on all the Apuleian works, he argues (rightly I think) against the view that the statement in the prologue of the Metamorphoses that the narrator learnt Latin late in life applies to Apuleius himself:

   Sed hoc quidem vix credet qui eum
   scriptores antiquos haud sine fructu
   legisse animadvertet, qui optime eum
   verba elegisse sentiet ut rebus et
   personis accommodaret, qui denique
   intelleget summam iam artem inesse
   in hoc libro, quem primum scriptum esse putant.

   'But this will hardly be credible to
   anyone who notes that he has read the
   older writers not without profit, who
   feels that he has made an excellent
   selection of words to suit his material
   and characters, who indeed under
   stands that this book (which they think
   was written first) contains the
   highest art.' (30)

The issue of the high quality of Apuleian style in the Metamorphoses is here connected with another argument, that the novel could not have been the earliest work of Apuleius (with which I also agree). (31) But Helm principally pursued another angle of negative criticism, one reflected in other scholarly work of the early twentieth century--the inappropriateness of Apuleius' inserted tales and the general confusion of his narrative line. Both these emerged from increasing scholarly interest in the relationship between Apuleius' Metamorphoses and its Greek model, rightly identified by Helm himself with the lost Greek Metamorphoses of Lucius of Patras, which suggested that Apuleius should be viewed as an inferior Latin imitator and even mangler of a Greek work. This is clearly the mindset behind Helm's view on the inserted tales, quae quamquam lector ut laetetur efficiunt, tamen totum narrationis corpus, ut ita dicam, totum deformant, 'which though they effect pleasure for the reader, nevertheless (so to speak) deform the whole body of the narrative.' (32)

Modern scholars have a very different view, and the consensus, following German work of the generation following Helm and especially the work in English of James Tatum in his important article of 1969, is (to cite Tatum himself) that 'by anticipation of later events; by sensitivity to the narrative 'environment'; by extremely subtle interrelationships between characters in a tale, and the people hearing it; and by thematic relationship to the final "Isisbook", the tales are not simply relevant to the main story, they are in fact essential to its conclusion and its philosophy of human life'. (33) We will return shortly to the key question of the coherence of Book 11 with the rest of the novel, but on the issue of the coherence of the tales with the main narrative Helmian scepticism is outmoded given modern research, which argues rightly for high Apuleian narrative and narratological skill.

The supposed incoherence of the tales was for Helm symptomatic of a larger incompetence, which he attacks in a section entitled De rebus ne glegenter compositis, 'on material carelessly composed.' (34) Though it is undeniable that a long work such as the Metamorphoses, like the epics of Homer and Vergil, has its small inconsistencies, some of Helm's examples are indicative of an unwillingness to accept that Apuleius was capable of subtle intertextuality. For instance, Helm discusses the scene from the narrative of Aristomenes to Lucius in the first book, where Aristomenes relates his own despairing attempt at suicide in a shabby inn, including his farewell address to his bed (Met. 1.16): 'grabbatule, animo meo carissime, qui mecum tot aerumnas exanclasti'. Helm comments: ita ... lectum appellare videtur, non qui per unam noctem mutilo et putri et alieno, sed qui suo et semper eodem usus est. sed non satis perspicio, qua ex conditione rerum haec possint sumpta esse, quae certe ad amantis tristem multo magis quadrarent, 'this kind of address to a bed seems not to belong to someone who has used another's mangled and crumbling bed for one night, but to someone who has used his own and always the same one. But I do not understand from what circumstances these words are derived, which would certainly be much more fitting for a lamenting lover'. Helm, though he is conscious to some degree elsewhere of both Apuleius' use of previous authors and his sense of humour, does not see (as Silvia Mattiacci has most fully explored) (35) that this scene brilliantly parodies several literary sources, and especially the tragic scenes in Euripides' Alcestis where the queen Alcestis, about to die for her husband Admetus, bids a touching farewell to her marital bed (177--80), or in Sophocles' Trachiniae, where the suicidal Deianira does the same (920--22), as well as the scene in Vergil's Aeneid where Dido, bent on suicide, bids farewell to the bed which she had shared with Aeneas (4.648--52). This (again) is one of many cases in which detailed modern research can make a concrete contribution, and where an undemanding view of Apuleian literary texture is less effective for interpretation.

Another key question confronted by Helm is that of the relevance of Book 11 to the rest of the novel: nam ut nos magnopere iuvat de diebus festis piaculisque et pompis Isidis edoceri, ita tota haec descriptio aliena est a hilari lubricaque fabula quae antecedit, 'for though it gives us great pleasure to be given such full information about the festivals, offerings of processions of Isis, this whole description is alien to the amusing and lubricious tale which precedes it.' (36) This fundamental issue was also famously faced by Ben Edwin Perry in his The Ancient Romances of 1967. By that date Perry had spent some five decades working on Apuleius and connected texts, and his fundamental view was that of Helm, and was well paraphrased as follows: 'his method was to look for evidence of res neglegenter compositae and to argue from them that Apuleius was either unwilling or unable to develop and sustain a single idea.' (37) Naturally with this general view, Perry followed Helm in viewing Book 11 as basically inconsistent with the rest of the Metamorphoses. His judgement, though lengthy, is worth citing for its emphasis and for its importance in the scholarly debate, though by 1967 the mood was turning the other way: (38)

   '... the last book of the Metamorphoses
   was added for a very special
   purpose ... to redeem his book from
   the appearance of complete frivolity.
   To publish for sheer entertainment a
   lengthy work of fiction in the form
   of dramatically sput-out witch stories,
   fairy tales and tales of sensational
   or scandalous adventure, all of which
   types of prose narrative were
   looked upon with disdain by his
   contemporaries as trivial old wives'
   tales (aniles fabulae), or tales fit
   only to be told on the street-corner (aureae
   fabulae), was something that
   Apuleius really wanted to do, but did
   not dare to do, without qualifying
   his work in such a way as would leave
   the impression that he had, after all,
   something of serious importance to
   convey by it, which was instructive,
   and high-minded, and thereby worthy
   of an educated writer. Book XI served
   that necessary purpose, but
   only in a very perfunctory and
   superficial fashion. Instead of building
   into the framework of his story-book
   as a whole an ostensible meaning in
   terms of satire, philosophical critique,
   or allegory which would be evident
   from start to finish, as is the case
   in Lucian's novels, Apuleius is
   content merely to tack on at the end
   a piece of solemn pageantry as ballast
   to offset the prevailing levity of
   the preceding ten books. With his
   showman's instinct for the value of
   immediate dramatic effects (which
   often leads him into self-contradiction
   elsewhere), he feels that all he
   needs to do in order to prevent the
   publication of his old wives' tales
   from becoming a scandal in the
   literary world, comparable to that of
   Aristides' Milesiaca, is to make
   a personal appearance on the stage in the
   last act, bow deeply and reverently
   before his audience, and overwhelm
   them with the magic of his eloquence
   on a subject of grave and universal
   import, a subject about which he speaks
   with earnest conviction and sincerity,
   but which does not belong with the
   story of Lucius.' (39)

At the same time, Perry, though critical of Apuleius'intellectual content and level of seriousness, was (like Helm) more positive about his style:

   'Indeed, it is in the realm of
   style that Apuleius has made his most original
   contribution to literature; for that
   style--so highly colored, fanciful
   and rococo, so studiously piquant and
   recherche, and so picturesque, varied
   and opulent--is shaped in large measure
   by his own romantic outlook
   on the world.' (40)

A romantic Apuleius is an interesting concept for modern Apuleian scholars; but it is plainly with this idea that Perry justified his interest in the writer, since he argued that the Metamorphoses was 'deeply permeated with a spirit of belief in the hidden and marvellous potentialities of nature and human life.' (41) Perry certainly did not hold an allegorical view of the Metamorphoses: in a terse footnote he dismisses the work of religious allegorisers such as Merkelbach as 'all nonsense to me,' (42) and he plainly holds that Apuleius' Platonism is too superficial to provide an allegorical interpretation of his novel.

2: Beyond Perry: Present and Future Directions in Apuleian Studies

Thus in Apuleian criticism up to Perry we have seen three main strands in a predominantly negative reception of Apuleius' Metamorphoses: criticism on the grounds of extravagant and decadent style, of frivolous and low-brow content, and of poor literary and narrative technique. Perry himself would agree to the last two of these charges while tolerating Apuleian style as interesting for its exoticism. In this final section I will look at these three areas in turn, trace the development of scholarship in each since Perry and open up future prospects, and conclude by suggesting some further directions which Apuleian scholarship might usefully explore.

In broader academic terms, the prejudice which underlies the view of Apuleius as late, decadent and second-rate is one which is under general challenge in modern classical studies. Driven partly by the vast body of existing research on traditionally canonical periods and genres, recent scholarship has focussed on writers and writings beyond conventional bounds and has re-evaluated texts previously viewed as 'marginal'; the Greek and Roman novels have recently been identified as a genre which has raised its standing as a result of this process, (43) and though there is naturally a danger of making exaggerated claims for the interest and quality of previously neglected texts, the avidity with which scholars have taken to the novel and the enthusiasm of their students who have seen the ancient novels become increasingly central to classics courses suggest that these texts provide substantial subjects of study. It is not perhaps coincidental that this movement within the classics has coincided with the more general challenge to the literary canon in the 'culture wars' of the 1980's and early 1990's. (44) As we shall see below, this ideological metamorphosis has affected scholarship on Apuleian language, literary texture and narrative technique, changing the common perception of the Metamorphoses: no longer an incoherent and marginal work in barbarous Latin, it is now viewed as an elegantly expressed, intertextually complex and narratologically intriguing central work of Roman literature.

Apuleian style and language

Between the time of Norden and Perry most analysis of the style of the Metamorphoses was devoted to listing its syntactical features (45) or to working out the balance in its diction between everyday and elevated Latin. (46) Callebat's later work (47) has used his analysis of the mixed register of the style in the Met. to argue that the richness and exuberance of Apuleian style represents a lively and vivid view of the world and a genuine awareness and seeking after religious experience and philosophical truth. Rather different and perhaps less speculative is the work by scholars such as Tatum, who like Norden ascribes many of the features of Apuleius' style to his sophistic intellectual context and his continuation of the Gorgianic tradition of rhetorical Asianism. (48) This is also true of perhaps the most useful recent brief analysis of Apuleian style, that of Kenney, (49) who also views Apuleius as an Asianist but as belonging to the central tradition of Latin literary style. Importantly, he stresses the poetic elements in Apuleius' language in the Metamorphoses, and suggests that Apuleius' exuberance derives partly from this feature rather than from mere eccentricity.

More thought is needed here on the way in which artistic prose replaced poetry as a vehicle for complex and ambitious literature under the high Roman empire, the very period where Apuleius is writing; the scattered evidence needs to be brought together and analysed in a detailed and literaryhistorical manner. We need in effect a general history of Latin prose in the second century from (say) Pliny the younger to Tertullian, in which Apuleius can be placed in his proper context. This should cover not only the development of the language of prose through the use of archaic and poetic vocabulary, but also issues of genre and literary history, especially the way in which more flexible prose genres such as the novel and the miscellany tend to replace conventional poetic genres such as the epic, something which has been only briefly touched on in the work of by scholars such as Marache and Steinmetz. (50)

One interesting feature of recent work is a revival of discussion of African Latin. Lancel has distinguished between the supposed lexical Africitas identified by nineteenth-century scholars and the tendency of writers from Roman North Africa to a high rhetorical style with exuberance and euphony as its main features, (51) which fits well Juvenal's quip that Africa is the nutricula causidicorum (7.14). There is no real evidence for lexical Africitas, he argues, but the shared general stylistic features demand further research. Kenney went beyond this in speaking of Apuleius' 'creative and poetic attitude

to the language, suggesting the experimental exploitation of an adopted tongue, as by an Ennius or a Nabokov'. (52) However, it seems likely that from his early education in Madauros Apuleius learnt Latin before Greek, though he may have known the vernacular Punic before either, and Latin is therefore for him the natural tongue of literary composition rather than an artificial form of discourse; and if learning Latin in an African context engendered such style, we would expect more of it in writers of similar background such as Fronto, who shares Apuleian archaism but not his exuberance. Most recently, Petersmann has shown that there is clear evidence from inscriptions and grammarians for some elements of pronunciation, orthography and even morphology as especially prevalent in Roman North Africa, but he denies that this extended to the written language and argues that such elements may be survivals from the archaic vulgar Latin of earlier Roman settlers. (53)

On this front Bakhtinian formalism might be a useful theoretical model. Though Bakhtin himself wrote about Apuleius in his work on the chronotope, novelistic space and time, (54) he did not discuss his language in detail apart from pointing to the potential openness created by Apuleian bilingualism. (55) Bakhtin's concept of heteroglossia can be profitably pursued further in an Apuleian context. Heteroglossia is roughly speaking the range of informal, varied, dialectal or colloquial languages which differ from and oppose official or literary discourse. (56) In Bakhtinian terms, a parodic novel such as the Metamorphoses is a site of continuous dialogue between this subversive linguistic tendency and official or elevated literary language. This would seem to fit well with the combination between colloquial, archaic and literary elements consistently detected by detailed research into the style of Apuleius' novel, and a study which could draw together the detailed linguistic evidence while applying a Bakhtinian model would be an exciting prospect; this key idea has been partly applied at a more thematic level in Finkelpearl 1998, who shows that the Met. seems to fit well the Bakhtinian notion that the novel naturally subverts and ironises more elevated literary genres, given its largely ironic relationship to epic, but could be pursued in more detail.

Frivolity of content

Many modern readers still feel the urge to interpret Apuleius' Metamorphoses as an allegorical text, feeling that the surface story is too insubstantial and frivolous for so apparently ambitious a literary work, and thinking (unlike Perry) that the novel can be unified through an allegorical interpretation. The most extreme example of this tendency is Merkelbach: his view that the whole novel, and especially the story of Cupid and Psyche, represents an allegorical version of the conversion of Lucius to Isiac religion in the last book, though much attacked, has been sustained by the same writer a generation later. (57) Other accounts have been advanced claiming that the novel is a Platonising allegory, a view as we have seen which goes back at least as far as Beroaldus in the Renaissance. (58) In my view these attempts to make Apuleius intellectually respectable in terms of uplifting content are misconceived, for reasons which say more about intuitive modern views on the supposed didactic and moralising function of high literature than about the Metamorphoses. (59) The programmatic prologue of the Met. itself claims that the reader will receive pleasure from reading this text (1.1 lector intende, laetaberis): there is no requirement to make further demands of ideological or improving content, and such content is in any case not the sole criterion for intellectual interest. What we have in the Metamorphoses, as (I would argue) in Ovid's homonymous epic poem, is a type of literature where the prime intellectual element derives not from a didactic message but from a complex and allusive literary texture.

One aspect helpful here is the connection of Apuleius with the Second Sophistic, which has been recently strongly emphasised. (60) Here is a literary environment where epideictic-type performance, whether in person or on the page, is a prime criterion of literary standing, and where literary texture and stylistic features such as archaism (or pure 'Attic' Greek) are often more important than the didactic content of a piece of writing. The Metamorphoses constantly demonstrates the cultural capital (to use Bourdieu's term) (61) of its author and his elite literary education, and this sociological aspect of literary culture in the second century AD, already raised for the Greek sophists, (62) could be given more emphasis in the study of Apuleius. (63) The Met. is clearly an act of self-definition and self-advertisement by an elite intellectual, and deserves further study as such.

Literary and narrative technique

The reassessment of the literary and narrative texture of the Metamorphoses has been perhaps the most important achievement of the post-Perry generation of Apuleian scholars. In terms of literary texture, the careful analyses of Apuleian style made by Callebat and others have laid the ground work for the type of thoroughgoing literary and linguistic commentary to be found in the latest volumes of the Groningen Commentaries on Apuleius, (64) where ten or more pages of commentary are dedicated to each page of Latin text, recognising the depth of allusion and literary ornament underlying Apuleian language, and showing that apparent peculiarities are due to conscious archaism, subtle intertextual echoes, and recognisable planning. These commentaries have been supplemented by an increasing number of studies. (65) These works and others of similar content make the kind of connections long familiar to readers of the scholarship on Latin poetry, showing a wide range of literary allusion and a subtle adaptation of diverse material to a particular novelistic context. This has allowed research on the literary texture of Apuleius to connect with what one might call the 'intertextual revolution', the way in which literary allusion and its various forms have come to dominate the most interesting recent work on Latin poetry, for example in the work of Conte, Thomas, Hinds, Barchiesi and Hardie. (66) The fruits of applying intertextual literary analysis to the Metamorphoses are still to be wholly harvested.

The recognition of complex literary texture has also led to a revaluation of the issue of the novel's readership. Earlier views, famously including that of Perry himself, that the implied readership of ancient novels was at a cultural level below that of the readers of elevated ancient texts such as epic and tragedy have now been refuted, not only through our increased knowledge of the circulation and production of literary texts, but also through the aforementioned scholarly analysis of novelistic texts which shows their highly literary and allusive nature. Novels like the Metamorphoses can thus be expected to engage in literary allusion at the same level as (say) the Aeneid or Ovid's Metamorphoses and achieve readerly recognition.

A comparable reversal has taken place in the analysis and valuation of Apuleian narrative technique. In terms of macro-narrative, most modern scholars are agreed against Perry that the Metamorphoses shows a conscious and consistent thematic unity and that Book 11 is organically linked to Books 1--10, even if they disagree about the nature of the unifying thread; (67) the specific connections between the conversion-narrative of Book 11 and the adventures of Lucius in Books 1--10 uncovered by close scholarly analysis are too many to be ignored. (68) In terms of micro-narrative, I alluded above to Helm's heading 'de rebus neglegenter compositis', and there is a similar section in Perry's chapter on Apuleius which is a natural extension of his belief in the work's overall lack of unity and the 'bolting-on' of Book 11. These concerns with illogicalities, especially in the embedded narrative stories of Aristomenes and Thelyphron, have some marginal validity, but too often impose twentieth-century standards of realism and consistency which both Homer and Vergil would fail; they also exclude deliberate narrative confusion and misleading which has since been seen as a characteristic Apuleian trait, and effects such as dramatic irony or the second-time reader. Thus Byrrhena's words at 2.31.2, hunc tua praesentia nobis efficies gratiorem, 'you will make this day more pleasant for us by your presence', plainly look forward to the forthcoming appearance of Lucius in the mock-trial at the festival of laughter. Perry regards this as impossible, since neither Byrrhena nor Lucius know that the trial will take place, as it is the result of unexpected events yet to happen, (69) but this view ignores the perspective of the omniscient narrator and the second-time reader, for whom this statement is proleptically amusing and dramatically ironic. (70) This passage has a close parallel in another utterance of Byrrhena, where, as Lucius looks at the statue of Actaeon, she says (2.5.1) tua sunt cuncta quae vides, 'everything you see belongs to you', both offering him the run of her house and suggesting without knowing it that Lucius too will come to disaster through voyeurism and bestial metamorphosis, though here there are also cleverly misleading elements (Lucius will not die, and will be saved rather than destroyed by the intervention of a goddess). (71)

Some countervailing voices were available even in Perry's own time, (72) but one of the most important developments in post-Perry Apuleian scholarship has been the general emergence of the view that the Metamorphoses is not only a work of careful thematic unity (whether or not through Isiac and Platonic colour) but also one in which all the resources of narrative technique are consistently exploited to subtle effect. Two important articles published shortly after Perry's book deserve mention here as symptomatic of this upwards revaluation of Apuleian narrative technique: that of Tatum, (73) which first adequately demonstrated the way in which the inset tales reflected the main plot of the novel, and that of Smith, (74) which pointed to the careful and subtle effects achieved by the separation of narrator and author. The most significant piece of work, however, is that of Winkler's Actor and Auctor, (75) a book which has been rightly characterised as epoch-making in Apuleian studies, (76) and on which I will not linger here. Its suggestion that the Metamorphoses is like a detective novel in which the ending leads to reassessment of the previous plot has been a vital contribution to arguments for overall unity and coherent structure, while its constant stress on Apuleian narratological tricksiness, hermeneutic playfulness and deconstructive tendencies has provided many creative answers to apparent difficulties and inconsistencies as well as demonstrating the text's narrative competence and indeed virtuosity.

Some future directions for research (77)

The work of Winkler shows that the application of literary-theoretical considerations can make decisive contributions to the interpretation of Apuleius. His use of narratology, along with its use by the Groningen group in the later volumes of their commentary, provides a springboard for much further analysis, whether or not specifically narratological in form. (78) Other areas of theory await fruitful employment. Relatively little feminist analysis has been applied to the Metamorphoses; though some work has been done on the representation of Psyche as a feminine type (79) and on the possible colouring of the narrative of Cupid and Psyche by its aged female narrator, (80) a feminist analysis of the function and characters of (e.g.) Pamphile, Photis and Byrrhena, or of the poor view taken of female sexuality in the 'adultery-tales' of Book 9 would be a very interesting development. The Metamorphoses might well respond well to postmodernist analysis, at least in its post-Winklerian form: the subversion by gaps and slippages of the master narrative of error, punishment, conversion and rescue is a key element in Winkler's deconstructive analysis, but has a recognisably postmodern tendency, as does the stylistic combination of traditional literary language, recherche archaism and exuberant neologism, and the overall aspect of hermeneutic playfulness and parody. A current project to study Bakhtinian formalism in the context of the ancient novel will surely have much to say about Apuleius from a Bakhtinian perspective, whether in connection with his notion of heteroglossia, discussed above, or through the use of his concepts of the carnivalesque (the Isis festival of Book 11?) or the chronotope (how 'everyday' is the space and time of the Met, and does it vary at significant points?).

A post-colonialist approach to the Metamorphoses fits (but would be complicated by) its status as an appropriation of a Greek plot (along with that plot's implicit opposition to Roman authority--cf. e.g the account of the soldier in books 9.39--10.1, cf. Onos 44--5). Here Apuleius' origins and location for most of his career in Roman North Africa might allow his analysis as a peripheral figure vis-a-vis the cultural core of Rome; this would not be in terms of the empire writing back (there are few signs of subaltern resentment in the Met.), but in terms of the evident anxiety of the Met. to belong to metropolitan Roman culture. (81)

Another area where there has been much work but where much is still needed is that of reception. The relative obscurity of Apuleius' Metamorphoses in late antiquity and the middle ages (82) is more than outweighed by its significant influence on the emerging novelistic literature of the Renaissance and early modern period, (83) and there is much significant nineteenth century material. (84) Looking the other way chronologically, intertextual analysis should continue to look at further earlier genres which have a particular influence on the literary texture of the Met.; here important work on Roman declamation and on New Comedy is in progress. (85)

Overall, there seems little doubt that future research will continue to stress the complexity, learning and subtlety of Apuleius' Metamorphoses, and its importance as a major and influential work of classical literature. Both these ideas would have been somewhat problematic for Ben Edwin Perry's view of this work, but we should not forget that but for his foregrounding of a neglected genre throughout his academic career, and particularly at its end in his Sather Lectures, the study of the ancient novel might well not be where it is today.

Bibliography

Adlington, W., 1566. The Eleven Bookes of the Golden Asse, London: Henry Wykes.

Alpers, K., 1980. 'Innere Beziehungen und Kontraste als "hermeneutische Zeichen" in den Metamorphosen des Apuleius von Madauros', WJA 6, 197--207.

Altick, R.D., 1957. The English Common Reader, Chicago: Chicago UP.

Ax, W., 1996. 'Quattuor linguae latinae aetates. Neue Forschungen zur Geschichte der Begriffe "Goldene" und "Silberne Latinitat" ', Hermes 124 : 220--240.

Bakhtin, M., 1981. The Dialogic Imagination [ed. M. Holquist], Austin: Texas UP.

Barchiesi, A., 2001. Speaking Volumes: Narrative and intertext in Ovid and other Latin poets, London: Duckworth.

Bernhard, M., 1927. Der Stil des Apuleius von Madaura, Stuttgart: Kohlhammer

Beroaldo, P, 1500. Commentarii a Philippo Beroaldo conditi in Asinum Aureum L. Apuleii, Bologna: Faelli.

Bitel, A., 2000. Quis Ille? Alter Egos in Apuleius' Golden Ass, Oxford D.Phil.thesis.

Bolgar, R.R., 1954. The Classical Heritage, Cambridge: CUP.

Bourdieu, P., 1984. Distinction: a social critique of the theory of taste, London: Routledge.

Bowie, E.L., and Harrison, S.J., 1993. 'The Romance of the Novel', JRS 83, 159--178.

Brzenk, E.J., 1978. 'Apuleius, Pater and the Bildungsroman' in Hijmans and van der Paardt, 231--238.

Brock, M.D., 1911. Studies in Fronto and his Age, Cambridge: CUP.

Callebat, L., 1968. Sermo Cotidianus dans les Metamorphoses d'Apulee, Caen: Univ. de Caen.

--1998. Langages du roman latin [Spudasmata 71], Hildesheim: Olms.

Carver, R.H.F., 1991. The Protean Ass: The Metamorphoses of Apuleius from Antiquity, Oxford D.Phil.thesis.

Coleman, K.M., 1988. Statius Silvae IV, Oxford: OUP.

Conte, G.B., 1986. The Rhetoric of Imitation, Ithaca: Cornell UP.

D'Amico, J.F., 1984. 'The Progress of Renaissance Latin Prose : The Case of Apuleianism', RenQ 37 : 351--392.

Doody, M.A., 1997. The True Story of the Novel, London: HarperCollins.

Dowden, K., 1994. 'The Roman Audience of the Golden Ass', in J. Tatum (ed.), The Search for the Ancient Novel, Baltimore: Johns Hopkins UP, 419--434.

Farrell, J., 2001. Latin Language and Latin Culture, Cambridge: CUP.

Fick--Michel, N., 1991. Art et Mystique dans les Metamorphoses d'Apulee, Paris: Univ.de Franche-Comte.

Finkelpearl, E.D., 1998. Metamorphosis of Language in Apuleius. A Study of Allusion in the Novel, Ann Arbor: Michigan U.P

Gianotti, G.F., 1986. 'Romanzo' e ideologia: studi sulle Metamorfosi di Apuleio, Naples: Liguori.

Gorak, J (ed.), 2001. Canon vs. Culture: Reflections on the Current Debate, New York--London: Garland.

Hardie, P.R., 2002. Ovid and the Poetics of Allusion, Cambridge: CUP.

Harrison, S.J., 1999 (ed.). Oxford Readings in the Roman Novel, Oxford: OUP.

--2000. Apuleius: A Latin Sophist, Oxford: OUP.

Hays, B.G., 1996. Fulgentius the Mythographer, Ph.D. Diss. Cornell.

Heine, R, 1978. 'Picaresque novel versus allegory' in Hijmans and van der Paardt, 25--42.

Helm, R., 1905. Apulei Opera Quae Supersunt II.1: Apologia, Leipzig: Teubner.

--1907. Apulei Opera Quae Supersunt I: Metamorphoseon Libri XI, Leipzig: Teubner.

--1910. Apulei Opera Quae Supersunt II.2: Florida. Leipzig: Teubner.

Hijmans, B.L., Jr. and van der Paardt, R.Th. (eds.), 1978. Aspects of Apuleius' Golden Ass, Groningen: Bouma's Boekhuis.

Hijmans Jr. B.L et. al., 1995. Apuleius Madaurensis Metamorphoses Book IX, Groningen: Forsten.

Hildebrand, G.F., 1842. L. Apuleii Opera Omnia I, Leipzig: Cnobloch.

Hinds, S.E., 1998. Allusion and Intertext, Cambridge, CUP.

Horsfall Scotti, M.T., 1991. 'Apuleio nel mondo tardo-antico', AAPel 66 [1990] : 75--88.

Huet, P-D., 1715 [tr. S.Lewis.]. A Treatise of Romances and their Origins, London: Hooke and Caldecott.

James, H., 1957 (originally 1884). 'The Art of Fiction' in id. [ed. L. Edel], The House of Fiction, London: Macmillan, 23--45.

Katz, P.B., 1976, 'The myth of Psyche: a definition of the nature of the feminine', Arethusa 9, 111--118.

Kennedy, G.A., 2001. 'The Origins of the Concept of a Canon and its Application to the Greek and Roman Classics' in Gorak (ed.), 105--117.

Kenney, E.J., 1990a. Apuleius: Cupid &Psyche, Cambridge, CUP.

--1990b. 'Psyche and her mysterious husband', in Russell, (ed.), 175--198.

Klein, U, 1967, ' "Gold-" und "Silber"-Latein', Arcadia 2 : 248--267.

Krautter, K., 1971. Philologische Methode und humanistische Existenz. Filippo Beroaldo und sein Kommentar zum Goldenen Esel des Apuleius, Munich: Fink.

Lancel, S., 1987, 'Y-a-t'il une Africitas?', REL 63, 161--182.

Marache, R., 1952. La critique litteraire de langue latine et le developpement du gout archaisant au IIe siecle de notre ere, Rennes: Plihon.

--1957. Mots nouveaux et mots archaiques chez Fronton et Aulu-Gelle, Paris: Presses Univ. de France.

Mattiacci, S., 1993. 'La lecti invocatio di Aristomene: pluralita di modelli e parodia in Apul. Met.1.16', Maia 45, 257--267.

May, R., 2002. A Comic Novel? Roman and New Comedy in Apuleius' Metamorphoses, Oxford D.Phil Thesis.

Medan, P., 1925. La latinite d'Apulee dans les Metamorphoses, Paris: Hachette.

Merkelbach, R., 1962. Roman und Mysterium in der Antike, Berlin--Munich: Beck.

--1995. Isis Regina--Zeus Sarapis, Stuttgart: Teubner.

Norden, E., 1898, Die Antike Kunstprosa vom VI. Jahrhundert v.Chr. bis in die Zeit der Renaissance II, Leipzig: Teubner.

Oudendorp, F., 1786. Appulei Opera Omnia I [ed. Ruhnken, D.], Leiden: van der Eyk en Vygh.

Penwill, J.L., 1990. 'Ambages Reciprocae: Reviewing Apuleius' Metamorphoses', in Boyle, A.J., (ed.).The Imperial Muse, II: Flavian Epicists to Claudian, Bendigo, Vic.: Aureal, 211--235.

Petersmann, H., 1998. 'Gab es ein Afrikanisches Latein? Neue Sichten eines alten Problems der lateinischen Sprachwissenschaft', in: B. Garcia Hernandez (ed.), Estudios de linguistica latina, Madrid: Ediciones Clasicas, I. 25--36.

Robbins, D., 2000. Bourdieu and Culture, London: Sage.

Russell, D.A. (ed.), 1990. Antonine Literature, Oxford: OUP.

Sandy, G.N, 1978. 'Book 11: Ballast or Anchor?', in Hijmans and van der Paardt, 123--140.

--1997. The Greek World of Apuleius [Mnemosyne Suppl.174], Leiden: Brill.

Schlam, C.C., 1990. 'Apuleius and the Middle Ages' in Bernardo, A. and Levin, S. (eds.), The Classics in the Middle Ages, Binghamton, NY: CMERS, 363--369.

Schmitz, T., 1997. Bildung und Macht: zur sozialen und politischen Funktion der zweiten

Sophistik in der griechischen Welt der Kaiserzeit [Zetemata; Heft 97], Munich: Beck.

Smith, W.S., 1972. 'The Narrative Voice in Apuleius' Metamorphoses', TAPA 103, 513--534. [reprinted in Harrison (ed.), 1999, 195--216]

Steinmetz, P., 1982. Untersuchungen zur romischen Literatur des zweiten Jahrhunderts nach Christi Geburt [Palingenesia 16], Wiesbaden: Steiner.

Stramaglia, A., 1996. 'Apuleio come auctor: premesse tardoantiche di un uso umanistico', Studi Umanistici Piceni 16 : 137--161.

Tatum, J., 1969. 'The tales in Apuleius' Metamorphoses', TAPA 100, 487--527 [reprinted in Harrison (ed.) 1999, 157--194]

Thibau, R., 1965. 'Les Metamorphoses d'Apulee et la theorie platonicienne de l'eros', Studia Philosophica Gandensia 3, 89--144.

Thomas, R.F., 1999. Reading Virgil and his Texts, Ann Arbor: Michigan UP.

Trollope, A., 1946 (originally 1883). An Autobiography, London: Norgate and Williams.

van Mal--Maeder, D., 2001. Apuleius Madaurensis Metamorphoses Livre II, Groningen: Forsten.

--2003, 'La mise en scene declamatoire chez les romanciers latins' in Panayotakis, S., Zimmerman, M., Keulen, W.H. (eds.), The Ancient Novel and Beyond, Leiden--Boston: Brill, 345--355.

--2003a (forthcoming), La Fiction des Declamations, Leiden--Boston: Brill.

--and Zimmerman, M., 1998. 'The Many Voices in Cupid and Psyche', in Zimmerman et al, (eds.), 83--102.

Walsh, P.G., 1970. The Roman Novel, Cambridge: CUP.

Winkler, J.J., 1985. Auctor & Actor: a narratological reading of Apuleius' Golden Ass, Berkeley--Los Angeles: California UP.

Wlosok, A., 1969. 'Zur Einheit der Metamorphosen des Apuleius', Philologus 113, 68--84.

--1999. 'On the unity of Apuleius' Metamorphoses', in Harrison (ed.), 142--156.

Vice, J., 1997. Introducing Bakhtin, Manchester--New York: Manchester UP.

Zimmerman, M., 2000. Apuleius Madaurensis Metamorphoses Book X, Groningen: Forsten.

--et al. (eds.) 1998. Aspects of Apuleius' Golden Ass II: Cupid and Psyche, Groningen: Forsten.

(1) My thanks go to the audience at the panel in New Orleans for discussion of an earlier version of this paper, and especially to Antonio Stramaglia for useful comment on Renaissance Apuleianism; also to Ruurd Nauta for helpful information and discussion on the issue of 'Silver Latin' (see note 11 below). My attendance at the 2003 APA Annual Meeting at New Orleans would not have been possible without generous grants from the British Academy (Overseas Conference Grant) and from the Faculty of Classics at Oxford, for both of which I am most grateful.

(2) On this issue see Kennedy 2001, 115.

(3) Clodius came from Hadrumetum near Carthage and therefore from Africa Proconsularis, the same province as Apuleius. Since Severus also came from the province, from Lepcis Magna, sui must suggest more than origin in the same province, and could imply that Clodius knew Apuleius, a generation older, but this semi-fictionalised biography is unfortunately not reliable on such details.

(4) On the dating of the Mythologiae see Hays 1996, 24.

(5) Fulgentius makes it clear that in his view Apuleius and Aristophontes write merely falsitas, a word he uses twice of their fictions in his analysis at Myth.3.6. On Fulgentius' methods of allegorising in the Mythologiae see the very helpful material in Hays 1996, 93--132.

(6) Cited from Fol.2v. of Beroaldo 1500, with some normalisation of spelling and punctuation. For an assessment of Beroaldo's commentary see Krautter 1971.

(7) Beroaldus' views here were paraphrased and adopted by Adlington 1566, the first English translator of Apuleius, in his preface 'To the Reader'; for this and other Renaissance adherents of allegorical interpretation cf. Heine 1978, 32--33.

(8) See Petersmann 1998 and n.51 below.

(9) For the historiography of the question see Brock 1911, 161--261, and the discussion below, pp.161--162.

(10) Cf. e.g. Bolgar 1954, 249--275.

(11) This is part of the invention of an inferior 'Silver Latin', which goes back at least as far as Erasmus; see Farrell 2001, 90--92, Klein 1967 and Ax 1996.

(12) See D'Amico 1984.

(13) It was reprinted many times in the seventeenth and eighteenth centuries; I cite from the English translation of Lewis (= Huet 1715).

(14) Huet 1715, 97.

(15) Huet 1715, 3.

(16) Id. ib.

(17) Huet 1715, 69.

(18) Huet 1715, 70.

(19) Huet 1715, 34.

(20) Huet 1715, 72.

(21) Oudendorp 1786.

(22) Cf. Marache 1957.

(23) Hildebrand 1842.

(24) Cf. Dowden 1994.

(25) Norden 1898, II. 600--601.

(26) On the battles over the status of the novel in Britain cf. e.g. Altick 1957 110--116, 123--126, 194--198; for Perry's consciousness of the debate cf. Perry 1967, 330.

(27) James 1957, esp. 23 'Only a short time ago it might have been supposed that the English novel was not what the French call discutable'.

(28) Trollope 1946, 194--198, arguing that the novel became respectable in the period 1826--1876.

(29) His editions (all subsequently revised) first appeared as follows: Apologia (Helm 1905), Metamorphoses (Helm 1907) and Florida (Helm 1910).

(30) Helm 1910, xi--xii.

(31) See Harrison 2000, 9--10.

(32) Helm 1910, v.

(33) Tatum 1969, 525 = Tatum 1999, 192.

(34) Helm 1910, xv--xvii.

(35) Mattiacci 1993.

(36) Helm 1910, vi.

(37) Sandy 1978, 124.

(38) Cf. e.g. Sandy 1978.

(39) Perry 1967, 244--245.

(40) Perry 1967, 239.

(41) Perry 1967, 239.

(42) Perry 1967, 336 n.17.

(43) Kennedy 2001, 115.

(44) On the 'culture wars' see especially the essays in Gorak 2001.

(45) The main contribution of Bernhard 1927.

(46) The route taken by Medan 1925 and especially Callebat 1968, which remains the most detailed and helpful work on the topic.

(47) Now conveniently collected in Callebat 1998.

(48) Tatum 1979, 135--159.

(49) Kenney 1990a, 28--38.

(50) Marache 1952, Steinmetz 1982.

(51) Lancel 1987.

(52) Kenney 1990a, 29.

(53) Petersmann 1998. 'African' pronunciation could be recognised as such: cf. Statius Silv. 4.5.45 sermo Poenus with Coleman 1988, 169.

(54) Bakhtin 1981, 111--123.

(55) Bakhtin 1981, 60, 64.

(56) Cf. Bakhtin 1981, 259--422 with the analysis of Vice 1997, 18--44.

(57) Merkelbach 1962 and 1995.

(58) See pp. 146--147 above and Thibau 1965, Gianotti 1986, Kenney 1990b, Fick-Michel 1991.

(59) See Harrison 2000, 235--259.

(60) Cf. Sandy 1997; Harrison 2000.

(61) Cf. Bourdieu 1984, and the comment of Robbins 2000, 32--36.

(62) Cf. Schmitz 1997.

(63) Harrison 2000, 226 n.88 at least mentions Bourdieu, but more can be done.

(64) Cf. Hijmans et al 1995, Zimmerman 2000, van Mal-Maeder 2001.

(65) Most notable perhaps are Walsh 1970, the first sustained argument for complex and allusive Apuleian literary texture, and Finkelpearl 1998, which is the first book-length study of Apuleius'novel to engage with the theory of intertextuality. For other work (especially by Mattiacci and Frangoulidis) cf. Harrison 1999, xxxiv--xxxv.

(66) Conte 1986, Thomas 1999, Hinds 1998, Barchiesi 2001, Hardie 2002.

(67) Cf. e.g. Sandy 1978, and further material cited at Harrison 1999, xxxii.

(68) Useful here are Wlosok 1969 (translated as Wlosok 1999) and Alpers 1980.

(69) Perry 1967, 279--280.

(70) See van Mal--Maeder 2001, 400.

(71) See van Mal--Maeder 2001, 91--92.

(72) See Harrison 1999, xxxv.

(73) Tatum 1969 (= Tatum 1999).

(74) Smith 1972 (= Smith 1999).

(75) Winkler 1985.

(76) Cf. e.g. Penwill 1990, Harrison 2000, xxxiii.

(77) It may be instructive to compare this final section with my views a decade ago (Bowie and Harrison 1993).

(78) A good example here is Bitel 2000.

(79) Katz 1976.

(80) van Mal--Maeder and Zimmerman 1998.

(81) For the Met. and the culture of Rome cf. Dowden 1994.

(82) Cf. Schlam 1990, Horsfall Scotti 1991, Stramaglia 1996.

(83) There is some stimulating material in Doody 1997, and a monograph version of Carver 1991 (which will provide a more detailed and contextualised account) is well advanced.

(84) E.g. Walter Pater's use of Apuleius in his Marius the Epicurean (1885); cf. Brzenk 1978.

(85) Cf. May 2002; van Mal--Maeder 2003 and 2003a (forthcoming).

STEPHEN HARRISON

Corpus Christi College, Oxford