1he Recollections o Lncolpius

ANCILN1 NARRA1IVL

Supplementum 2
!
!
"#$%&'$()!*&('#!
Maaike Zimmerman, +,$-.'/$%0!&1!2'&,$,3.,!!
Gareth Schmeling, +,$-.'/$%0!&1!4)&'$#(5!2($,./-$)).!
leinz lomann, +,$-.'/$%6%!789$,3.,!!
Stephen larrison, :&';</!:='$/%$!:&)).3.5!>?1&'#
Costas Panayotakis ,reiew editor,, +,$-.'/$%0!&1!2)(/3&@ !

A#-$/&'0!*&('#!
Jean Alares, B&,%C)($'!D%(%.!+,$-.'/$%0!
Alain Billault, +,$-.'/$%E!F.(,!B&<)$,5!G0&,!HHH
Lwen Bowie, :&';</!:='$/%$!:&)).3.5!>?1&'#!
Jan Bremmer, +,$-.'/$%0!&1!2'&,$,3.,!
Ken Dowden, +,$-.'/$%0!&1!*$'I$,3=(I!
Ben lijmans, Lmeritus o Classics, +,$-.'/$%0!&1!2'&,$,3.,!
Ronald lock, +,$-.'/$%0!&1!D&<%=.',!:()$1&',$(5!G&/!A,3.)./!
Niklas lolzberg, +,$-.'/$%6%!B8,C=.,!
Irene de Jong, +,$-.'/$%0!&1!AI/%.'#(I!
Bernhard Kytzler, +,$-.'/$%0!&1!J(%()5!K<'9(,!
John Morgan, +,$-.'/$%0!&1!L()./5!D@(,/.(
Ruurd Nauta, +,$-.'/$%0!&1!2'&,$,3.,!
Rudi an der Paardt, +,$-.'/$%0!&1!G.$#.,!
Costas Panayotakis, +,$-.'/$%0!&1!2)(/3&@!
Stelios Panayotakis, +,$-.'/$%0!&1!2'&,$,3.,!
Judith Perkins, D($,%!F&/.;=!:&)).3.5!L./%!M('%1&'#!
Bryan Reardon, Proessor Lmeritus o Classics, +,$-.'/$%0!&1!:()$1&',$(5!H'-$,.!
James 1atum, K('%I&<%=!:&)).3.5!M(,&-.'5!J.@!M(I;/=$'.!
Alons \outers, +,$-.'/$%0!&1!G.<-.,!

D<9/C'$;%$&,/!
Barkhuis Publishing
Zuurstukken 3 961 KP Lelde the Netherlands
1el. -31 50 3080936 lax -31 50 3080934
inoancientnarratie.com www.ancientnarratie.com
1he Recollections o Lncolpius!

1he D(%0'$C( o Petronius as Milesian liction











Gottskalk Jensson

BARKlUIS PUBLI SlI NG
&
GRONINGLN UNI VLRSI1\ LI BRAR\
GRONINGLN 2004


Bkin er tileinkuJ dtrum mnum
Ragnhildi og Lllen
Book design: Barkhuis Publishing
Dust-jacket design: 1eikna Design Inc.
Printed by: Drukkerij Giethoorn ten Brink

ISSN 1568 3540
ISBN 90 80390 8 1








Table oI contents
PreIace IX


PART 1 NARRATIVE 1

1.1 Text, Context and Identity 3

1.2 The Desultory Voice oI Encolpius 29


PART 2 STORY 85

2.1 Sorting the Fragments 87

2.2 Retrospective Soliloquies and Dialogues 136

2.3 Rewriting the Satvrica (My Turn) 174


PART 3 GENRE 189

3.1 Ancient Narrative in Personis 191

3.2 The Hidden Genre 245


Abbreviations 303

Bibliography 305

Index 319

Abstract 329



PreIace
Like probably every modern student oI the Satvrica, I began working with a
text that did not make a whole lot oI sense to me. While reading the schol-
arly literature, Iurthermore, I was struck by the ubiquitous exceptionalism oI
twentieth-century Petronian studies. Numerous articles and books written
within the last hundred years contain emphatic statements to the eIIect that
Petronius is unlike any other author and the Satvrica a unique work to which
no ordinary rules apply. It soon became clear to me that the very paradox oI
the Satvrica and its authora paradox which had, as I Iound out, been cre-
ated by scholars themselves not much more than a century agohad come to
Iunction as a hermeneutic barrier in reading the Satvrica. Because Petronius
was thought to be so exceptional, his text became virtually inexplicable and
readers gave up trying to interpret the work as a coherent whole. Instead,
most scholarly work concentrated on bits and pieces oI the preserved text
which could be useIully studied without having to deal with the problems oI
the genre or the narrator, beyond reaIIirming the negative modern thesis that
the one was synthetic and that behind the other hid the author.
Unsurprisingly, then, the conservative wish to respect the premises oI the
discipline and the institutional pressures to come up with new things to say
about this ancient text have lately generated what are, in my view, some
rather bizarre readings. One may be told variously that Encolpius` Iictional
autobiography is 'the narrative equivalent oI a play, or that as a text it 're-
sists its own interpretation, or even that it is 'anti-narrative, communicat-
ing only through the Iigures oI language. Meanwhile, there has been no ex-
amination oI the modern conception or reception oI the Satvricaclearly by
now bankrupt as such, but all the same providing a basis Ior downbeat con-
servative scholarship and avant-garde theorizing alikeviz. that it was writ-
ten to give a novelistic, even realistic description oI the author`s times, or the
image oI Petronius as an original Italian genius, 'perhaps the only Roman
who created his art independent oI the Greeks. In the last sections oI this
study I attempt such an examination, but I have by no means exhausted the
subject and much more could be said about the prejudices motivating the
invention oI the modern Petronius.
II we can relieve it oI the baggage oI its nineteenth-century reception, the
Satvrica will turn out to be both less than it has recently been thought to be
PREFACE X
and more than we had previously hoped. It can be thought oI as a compli-
cated literary game, inIormed by a sophistic reading oI the Homeric Odvs-
sev, but its rules are at least consistent and can be studied. Its humor and
message are scholastic in a positive sense, learned and playIul. The Lucianic
author oI the Erotesa text which, as I show in my study, has much in
common with the SatvricajustiIies such story telling in the prologue as
relaxation Ior the educated scholar who is weary oI unceasing attention to
serious topics. But scholars are serious beings and their Iun is not without a
darker side. Preoccupation with shady topics is indeed a characteristic oI the
Satvrica, its tone is oIten sarcastic and the story hopelessly obscene. One
aspect oI this kind oI literature is its examination oI the ancient belieI system
oI scholars, the scholarly view oI the world. Reading it with attention today
could provoke an examination oI the modern reader`s attitudes. Gian Biagio
Conte argues quite correctly that the Satvrica is not mainly trying to advance
a pseudo-aristocratic grudge against uneducated upstarts like Trimalchio but
is, equally, an analysis oI the rich but certainly conIused and sometimes sor-
did mentality oI the learned. The genre involves a playIul dismantling oI
scholarly preconceptions, a kind oI Saturnalia Ior the literati.
Petronius` Satvrica is a derivative text in two senses: Iirstly, it plays the
genre-derivative game oI satire and parody, and secondly, our Latin text by
Petronius Arbiter looks and Ieels like a Roman palimpsest, a reworking oI a
preexisting Greek Satvrica, most likely called just that, !"#$%&'(. The pro-
posal that Petronius` text is a palimpsest has not been made beIore, and it
was not an easy one to make. Such a hypothesis is, oI course, the polar op-
posite oI the belieI in Petronian originality which has been unshaken since
Mommsen`s days and held by German, Italian, French, British and American
scholars alike. Great scholars have been ridiculed Ior suggesting that
Petronius had imitated a preexisting Greek genre or even borrowed a motiI
Irom Greek Iolktales. In Iact, the Iew scholars who, like the German philolo-
gist Karl Brger, dared to suggest that Petronius was writing a traditional
work never argued Ior the possibility oI a Roman palimpsest. What is meant
by a 'Greek model in Petronian scholarship is never a single Greek text
adapted by Petronius but either a 'serious (tragic) type oI Greek novel to be
parodied a la Heinze, or some hypothetical Greek genre which is designated
by some such label as 'realistic, 'comic, 'criminal or the anachronistic
'picaresque (Irom the Spanish word picaro), with its German translation
Schelmenroman. Although a rather obvious one, were it not Ior a scholarly
blind spot, the possibility oI a straightIorward adaptation Irom an otherwise
lost Greek text has not been entertained beIore, not even when scholars have
attempted to list all the hypothetical possibilities (Jensson 2002, 88).
PREFACE XI
The present book is a substantially revised version oI my dissertation,
written in Rome and Toronto Irom 1994 to 1996, and deIended at the Uni-
versity oI Toronto in November 1996. At the time the text was accepted
without changes, and thereIore I should perhaps employ the Horatian topos
and pride myselI on having waited until the ninth year beIore publishing. But
I cannot claim to have done so out oI modesty or a desire to create a perIect
work. In Iact, the dissertation has been a copyrighted text in the public do-
main Ior most oI this time. Having completed it, moreover, I Iurther devel-
oped my redeIinition oI the generic term 'Milesian in a paper I read at a
graduate seminar in Toronto in March 1997, and again in a reworked Iorm
under the title 'Milesian Tales: Short Stories or Novels? at the CAC Meet-
ing in June 1997 in NewIoundland. The arguments advanced in these talks,
which were not published as such but are now integrated into this study, are
in many ways similar to those presented by Stephen Harrison at the Gronin-
gen Colloquia on the Novel (May 1997) in a paper published a year later in
the homonymous series under the title 'The Milesian Tales and the Roman
Novel. At the time neither oI us knew oI the other`s work. I should also
mention that an article I published in Ancient Narrative 2, 'The Satvrica oI
Petronius as a Roman palimpsest, is a byproduct oI the present study. Un-
Iortunately, it is not impossible that this book, because its completion has
taken so long, does not adequately reIlect relevant literature published since
1996. I have, however, tried to take recent work into consideration in my
rewriting, and in the meantime I have undertaken to review signiIicant new
books (see bibliography); Iormulations developed in those reviews have
admittedly contributed in places to the present text.
At various stages I have beneIited Irom the advice oI helpIul readers and
reIerees, and many colleagues and Iriends have provided generous help to
me while writing and rewriting this book. I owe them all an immense debt.
Originally my readership was composed oI a select Iew, the members oI my
committee, all oI whom were encouraging and ready with advice. I am cer-
tainly most grateIul to my supervisor Roger Beck, who never showed signs
oI losing Iaith when my initial attempts were unsuccesIul, and later read
draIts oI individual chapters thoroughly and wrote useIul comments in the
margin, several oI which have Iound their way into the present text. I would
also like to extend my gratitude to Brad Inwood, who in the oIIice oI gradu-
ate coordinator was more actively involved than was required oI him. It has
been my good Iortune to work with Hugh Mason, whose knowledge oI an-
cient prose narratives proved invaluable, and Christopher Jones whose
graduate seminar on the ancient novel constituted the beginning oI the work
that has led to the present book. Alison Keith, Eric Csapo and Catherine
PREFACE XII
Connors deserve special thanks Ior their helpIul suggestions and Iriendly
advice. Neither should I Iorget to mention the valuable insights oI Gerald
Sandy whose report on my thesis I have made use oI in my rewriting. Arthur
James, Patricia Fagan and Robert Nickle, Iellow graduate students at To-
ronto, Irequently lent patient ears to my discursive experimentation relating
to the vast problems oI the Satvrica. I also wish to remember the staII oI the
Robarts Library and the adjacent Thomas Fisher Rare Book Library which
houses many Petronian treasures bequeathed by the late Gilbert Bagnani.
Similarly, my warm thanks go to Ann-Marie Matti oI the Department oI
Classics at the University oI Toronto.
I owe to Maaike Zimmerman the Iact that my book has now Iinally been
printed. AIter reading a copy oI the dissertation I had sent to a colleague oI
hers in Groningen, she wrote me to suggest the possibility oI publishing it as
an Ancient Narrative Supplement. I am also grateIul to Minna SkaIte Jensen
Ior reading my work and encouraging me to publish, and to Tarrin Wills,
Michael Chesnutt, Matthew Driscoll and Christopher Sanders, at The Arna-
magnan Institute in Copenhagen, where most oI the rewriting took place,
who have provided much good advice on the English language. My Iriend
Claudia Neri deserves warm thanks both Ior generously oIIering to design
the dust jacket and Ior all the help she provided while I was writing the dis-
sertation. Last but not least, I wish to acknowledge the vital support oI my
wiIe Annette Lassen who has read the entire text in typescript and suggested
many improvements. Despite all the help I have received the Iollowing text
will surely still contain imperIections. It goes without saying that I alone am
responsible Ior the remaining errors, misunderstandings and inIelicities.

Part 1
Narrative

1.1 Text, Context and Identity
1.1.1 Rewriting the Satvrica

This study is an attempt to interpret the Satvrica in accordance with its origi-
nal design as an extended Iictional narrative, in deIiance oI the severe limi-
tations imposed by the Iragmented state oI the extant text. Despite the copi-
ous measures oI material still extant Irom the original Satvrica (175 pages in
the standard edition),
1
anyone wishing to advance a literary interpretation oI
the work Iaces the daunting task oI working with an extraordinarily Irag-
mented text. As the result oI obscure events in the textual history oI the Sa-
tvrica, the modern text must be reassembled Irom Iour diIIerent and overlap-
ping traditions, the most extensive oI which owes its preservation largely to
late sixteenth-century printed editions.
2
Beyond these Iour textual traditions
there are several important allusions to the work in late ancient sources (e.g.,
Servius, Boethius, Sidonius Apollinaris and Fulgentius), and some loose
poems Irom separate traditions attributed to Petronius.
When this modern patchwork oI a text is read through, it becomes appar-
ent Irom the contents that what we have Iirst opens late in the storyproba-
bly no earlier than in the latter halI oI the originaland breaks oII beIore
there is an end in sight. That this is the case is also indicated by manuscript
evidence which assigns the extant text to books 14, 15 and 16.
3
The extant
material could well derive Irom more than three books and thus take us be-
yond book 16, but we cannot say how many books there were in all. II the
book numbers are to be trusted, the scale oI the original appears to have been
at least Iour times the extant text.


1
Mller, K. 1995. Petronius Satvricon reliquiae, Stuttgart and Leipzig.

2
The Long Fragments (L), see Richardson 1993. The other textual traditions are the Short
Excerpts (O), the Florilegia ()), which cover much less, and provide practically no other
material than L; and Iinally there is the Cena Trimalchionis (H), which represents the only
continuous and whole text, as Iar as it goes, and is preserved in a single manuscript, the
Traguriensis or codex Parisiensis latinus 7989. For attempts to explain the obscure history
oI the text, and especially the unclear relationship between L, O and ), see Mller 1983,
381II.; Reardon 1983, 295II.; Richardson 1975, 290II.; van Thiel 1971, 2II.

3
Chapters 1.126.6 are likely to represent Iragments Irom book 14. Book 15 most likely
began at 26.7, with the arrival oI a new day. How many books are represented by the rest
(likely more than one), or where divisions are to be placed, is problematic. For a Iurther dis-
cussion oI the problem, see, e.g., Mller 1983, 410II.
4 1 NARRATI VE
These statistics, however, do not tell the whole story, since the inIorma-
tion that we have is by no means limited to what is Iully preserved oI the
text. As in other extended Iictional narratives, internal allusions abound. In
theory at least, the entire context oI the Satvrica is implicit in every pre-
served sentence. A Iew words can sometimes supply enough inIormation to
outline the contents oI entire episodes. As it happens, the inIormation avail-
able to us has not been recognized Ior what it is worth. Because oI the nine-
teenth-century misunderstanding oI certain crucial external Iragments, and
the general lack oI interest in the larger Iorm oI the work in the twentieth
century, there is a strong tendency in the scholarship to trivialize the signiIi-
cance oI internal allusions, despite the Iact that the extant text is conditioned
by earlier episodes. My Iindings in this study indicate that we can improve
considerably our knowledge oI the plot in the missing early parts oI the
story, leading up to the point where the extant text begins.
There is no denying that the study oI the Satvrica has been greatly ham-
pered by the incompleteness oI the text, but another Iactor oI major conse-
quence is also a general, but not necessarily justiIied perplexity regarding the
authenticity oI what is extant. Even the current approach to the text, which
was initiated and deIined by Franz Bcheler`s critical edition oI 1862, Iails
to show adequate respect Ior the tradition. As I show in section 1.2.4 oI my
study, the narrative oI the Satvrica properly represents 'spoken discourse to
be listened to, as opposed to a 'written text to be read, which explains why
the style is so colloquial and improvised in Ilavor. This allows Ior unusual
Iorms, colloquial syntax, brevity in descriptions, loose ends, and quirky tran-
sitions. Recently published Greek papyri oI prosimetric Iiction indicate that
it may be unreasonable to expect a Iully polished literary text in this genre.
4

Such a text does not invite rationalizing emendation. In Iact, there is no rea-
son to be overly skeptical about the tradition. As Iar as it goes, it seems reli-
able in preserving an unusual and very diIIicult text. In any case, nothing is
known about the circumstances surrounding the loss oI most oI the text, and
very little is known about the origin and relationships oI the Iour overlapping
traditions. There exists thereIore no reliable historical basis on which to
ground a systematic emendation oI the text. Under these circumstances edi-
torial conservatism seems obligatory.
The history oI editorial responses to our text, however, reveals an aston-
ishing unwillingness to accept the received tradition, as is demonstrated by
the unusually great number oI misguided attempts at repair, ranging Irom
creative rewriting, to arbitrary reshuIIling oI Iragments, to wholesale pruning

4
See Stephens and Winkler 1995, 365, n. 19.
1. 1 TEXT, CONTEXT AND I DENTITY 5
oI supposedly alien elements. The record shows that the text oI the Satvrica
has throughout history been threatened by a strange license taken with it by
scribes and scholars alike. This attitude is discernible already in the methods
oI the earliest 'editor oI the work, the collector oI the Short Excerpts (O).
This unknown scribe seems to have been chieIly interested in sampling po-
etic and literary passages, and censoring pornographic material oI the 'ho-
mosexual type. He appears to have attempted to stitch together the gaps
where he had leIt out material, implying new continuity by means oI clever
juxtaposition.
5
From the same manner oI license stemmed Nodot`s publica-
tion (Paris 1694) oI a supposedly complete textpassed oII as the transcript
oI a newly Iound manuscriptthat proved to be a Iorgery. This literary hoax
is an entertaining story on its own, but the promised Petronius restitutus is
little more than the present Iragments with the gaps imaginatively supple-
mented.
6
Again the creative approach was attempted by Marchena (1800),
who Ileshed out the Iragmentary Quartilla episode with sexually explicit
material that owed little to the obscenity oI the original.
7

The erotic nature oI the Satvrica is certainly a Iactor contributing to the
licentious approach which is traditionally taken to this text. Many oI the
diIIiculties that have beset the text throughout the ages derive Irom the
irreconcilability oI the work`s pornographic material with the moral values
oI scribes and scholars.
8
Whether in the libidinous additions oI Marchena or
in the censorious logic that guided the excerptor oI the Short Fragments (O),
the sexually explicit material in the Satvrica has been crucial Ior its textual
preservation.
Today, this clash oI values can be seen in the diIIerent reception aIIorded
the least obscene part oI the work, the Cena Trimalchionis, and the consid-
erably more indecent main body oI the Satvrica`s text. The Cena, which is
regularly published separately and sets the boundaries Ior most learned
commentaries (e.g., Friedlnder, Marmorale, Perrochat and Smith),
9
is also
traditionally the Iocus oI halI oI all the scholarship on the Satvrica.
10
This
exaggerated concentration on one third oI the whole continues to produce
readings oI the Cena out oI context, and is ultimately responsible Ior the

5
See Mller 1983, 420II.
6
See Stolz 1987.
7
Rose 1966, 268II., prints the supplements.
8
Slater`s 1990, 40, denial oI the pornographic nature oI the work overlooks much material,
and relies on a too narrow deIinition oI pornography.
9
It is, however, not correct to say that there exist no commentaries on the whole oI the work;
see, e.g., Paratore 1933, vol. 2, and Pellegrino 1975, 205443. Many editions and transla-
tions also provide important commentaries in the Iorm oI notes.
10
See the bibliography oI Schmeling and Stuckey 1977.
6 1 NARRATI VE
rampant generalizations about the whole work based on that part alone. It
seems no coincidence that this section oI the text which most easily meets
with moral approval should also be the best preserved, and vice versa. Most
likely, the great loss oI text was caused, not by accident, but by biased ne-
glect or deliberate destruction. We should thereIore be wary oI the circularity
oI the argument that is oIten used to account Ior the modern neglect oI the L-
tradition in comparison with the much studied Cena Trimalchionis. Its inIe-
rior quality as text (real and hypothesizeddiscussed Iurther below) in com-
parison to the Cena is hardly a legitimate rationale Ior continued disregard,
since it was negligent attitudes in the Iirst place that led to such poor preser-
vation.
It goes without saying that the wildly creative restorations oI Nodot and
Marchena were rejected by modern editors, who have nevertheless Iailed to
see the analogy with their own sometimes no less radical manipulation oI the
text. Related to the modern neglect oI the Long Fragments is their indirect
subversion by an over-conIident indication oI 'new lacunae. A deIense Ior
this editorial practice could point out that the Long Fragments (L), where
they overlap with the early sections oI the Cena, 27.1 37.6, are mislead-
ingly presented as an unbroken text and Iail to indicate about eight short
passages Iound in the Cena. This is certainly a cogent sign oI poor quality in
the L-tradition, although not a Iormal prooI, since the argument involves
generalizing about the whole text Irom a Iew passages that could, theoreti-
cally, be oI inIerior quality. It would, however, be wrong-headed to deny the
importance oI this evidence and it can surely be accepted. At this point,
however, caution is needed, Ior although we may assume that there must be
lacunae, we have no method oI determining where they are and what is miss-
ing. To clariIy this point, a brieI examination is needed oI the maniIest but
unmarked gaps in the L-tradition, where it overlaps with the Cena. This is,
aIter all, the only place in the whole text where we can check the nature and
location oI the lacunae against some evidence.
What strikes one Iirst is that all eight gaps in the overlapping area are
small, one to Iour lines each, and thereIore no crucial material Ior under-
standing the narrative has been lost. At least two seem the result oI careless-
ness in copying. The eye oI the copier has missed a line because the same
word occurred in two lines in a row (34.7 vinum and 35.4 super).
11
More to
the point, halI oI the lacunae would have gone completely unnoticed (27.4I.,
28.3, 28.6I., 29.9I.) without the singular advantage oIIered by the complete
text in the Cena manuscript, since in those places where the gaps occur there

11
van Thiel 1971, 66.
1. 1 TEXT, CONTEXT AND I DENTITY 7
is no discernible break in continuity.
12
As Ior the other Iour (30.5I., 34.4,
34.7, 35.4), three oI them could neither have been identiIied with certainty,
nor supplemented without the Cena. This leaves us with only one lacuna
(35.4), which would sooner or later have been noticed and could partly have
been supplemented because it occurs in the middle oI a Iixed zodiacal cata-
logue, a context which determines that scorpio must be the missing sign,
although the item placed on top oI it would have remained a mystery (editors
have not even accepted the evidence oI the Cena manuscript on this ac-
count).
The editio Tornaesiana (Lyons 1575), a handy little book in octavo and a
major witness to the L-tradition, provides us with an excellent prooI that our
estimation is more than mere pessimism. Uncontaminated with the Cena (the
Traguriensis manuscript had not yet been discovered) and clearly attempting
to be critical in its presentation oI the text, the Tornaesiana prints the section
in question without noticing the Iirst seven lacunae at all. However, the edi-
tor, Tornaesius, suspects the eighth one (35.4), suggests a possible supple-
ment in case oI a lacuna, and gets the missing words super scorpionem cor-
rectly, although not the item on top.
13
SigniIicantly, he also hypothesizes
three lacunae which are not in H (28.1, 31.2I., 37.1), two oI which are con-
jectures, as is careIully noted on the margin,
14
and all oI which were rejected
by Mller.
15


12
27.45, cum has ergo miraremus lautitias, ,, Trimalchio digitos concrepuit . ('While we
were admiring these reIinements, ,, Trimalchio snapped his Iingers .)
28.3, iam Trimalchio unguento perfusus tergebatur, non linteis, sed palliis ex lana mollis-
sima factis. tres interim iatraliptae in conspectu eius Falernum potabant. ,, hinc involutus
coccina gausapa lecticae impositus est . ('Awash in Iragrant oil Trimalchio was being
rubbed down, not with linen cloths, but with towels made Irom the soItest wool. Meanwhile
beIore his eyes his three masseurs were drinking Falernian wine. ,, Then wrapped in a bright
red dressing gown he was placed in a litter).
28.68, sequimur nos admiratione iam saturi et cum Agamemnone ad ianuam pervenimus.
,, in aditu autem ipso stabat ostiarius prasinatus . ('We Iollowed having by now satisIied
our appetite Ior wonder and arrived in the company oI Agamemnon to the door ,, At the
very entrance stood a porter wearing green .)
29.930.1, interrogare ergo atriensem coepi, quas in medio picturas haberent. 'Iliada et
Odvssian` inquit. ,, iam ad triclinium perveneramus . ('I began to ask the janitor about
the images they had in the middle. The Iliad and the Odvssev`, he replied. ,, We had now
reached the triclinium .)
13
An asterisk in the text, page 34I., reIers to the Iollowing marginal note: deest fortassis super
scorpionem eiusdem nominis piscem, i.e., the phrase possibly missing is 'super . piscem.
14
Where it says, 'non est nota in v[eteribus]c[odicibus]`, and 'deest aster in v.c.`
15
Surprisingly so, perhaps, considering that editor`s penchant Ior accepting such suggestions
and incorporating them into the text.
8 1 NARRATI VE
Since we generalized about the quality oI the whole L-tradition on the
basis oI its overlapping with the Cena, we must also recognize the implica-
tions oI this low success-rate, only a part oI one corruption out oI a total oI
eight, Ior the likelihood oI locating and correctly Iilling lacunae by guess-
work. Moreover, since the exceptional case oI a missing item in a Iixed cata-
logue is not likely to present itselI oIten, the little success that was shown
can in practice be reduced to none. It is salutary to remember that modern
editors have Iew tools in their hands which were not available also to the
editor oI the Tornaesiana, since this sort oI emendation relies entirely on the
editor`s 'Ieeling Ior the Latin language and logic oI the story.
The unlikelihood oI improving the text, however, has by no means weak-
ened the conIidence oI modern editors in hypothesizing lacunae and lacuna-
related corruptions in the text oI the Satvrica. In the present standard edition
(Mller 1995), no less than six new editorial lacunae have cropped up in the
very same overlapping area that we have been discussing (27.1 37.6). On
the whole, no editor was more productive in this Iield than Bcheler (Berlin
1862), over Iorty oI whose new editorial lacunae Mller has incorporated
into the modern text. I count no Iewer than seventy-Iive dotted (.) new
lacunae in the 1995 edition, which must be added to the one hundred and
Iourteen asterisked (*) old ones, i.e., lacunae librorum auctoritate testatae.
16

These asterisks derive Irom the L-tradition, which uses them to mark
lacunae, although they can signiIy various other things as well.
17
Sometimes
lacunae seem to have entered the text out oI ambiguity about the meaning oI
the asterisks.
18
Given the obscure origins oI the L-tradition, whose main
witnesses are printed editions, some oI the 'authoritative lacunae are likely
to be conjectures in the Iirst place. In recent editions, however, not even the
Cena itselI has been spared new lacunae, whose single witness, the manu-
script Traguriensis, nevertheless presents an unbroken text.
19
In the 1995
edition, there are over twenty speculative lacunae in the Cena. By 'specula-

16
For comparison Ernout`s more conservative edition (cinquieme tirage, 1962) has 36 dotted
lacunae and 121 asterisked ones.
17
The O-class manuscripts lack indications oI lacunae and present a continuous text. The
Tornaesiana, an L-class witness, inserts one or more asterisks into the text, and uses two
kinds oI asterisks (six points lacunae; Iive points marginal gloss). Other L-class wit-
nesses use multiple asterisks to indicate corruption in the text. They are variously placed
within running lines, at the end or the beginning, or in otherwise empty lines. They can also
indicate generic breaks in the discourse, especially beIore and aIter speeches and poems.
18
See Di Simone 1993, 88 n.5, Ior a case oI such misunderstanding involving both Mller
1983 and van Thiel 1971.
19
'|T|he lacunae, iI they exist, being Iew and unimportant, notes Gaselee 1915, in the
introduction to his Iacsimile edition and transcript oI the manuscript.
1. 1 TEXT, CONTEXT AND I DENTITY 9
tive I mean that the introduction oI these gaps into the text oI the Satvrica is
based on subfektive Deutungen, i.e., pure guesswork. Moreover, the majority
oI the new editorial lacunae have been inserted without arguments Ior their
support. By comparison, Helm`s edition oI the Metamorphoses oI Apuleius
has Iour lacunae in a text that is oIten diIIicult to make sense oI and consid-
erably longer than the Satvrica. Such steady accretion oI textual ruptures has
made the reading oI the Satvricanever an easy aIIairincreasingly diIIi-
cult.
Other such licenses with the text involve the hypothesized dislocation oI
passages, this too on the ground oI subfektive Deutungen. A case in question
is three elegiac distichs (14.1), which apparently come out oI the blue in the
middle oI Encolpius` narrative. Bcheler, partly on the authority oI Anton`s
1781 edition, postponed these verses by about Iive lines and appended them
to words uttered by the character Ascyltos in that passage. There, Ior the
most part, they have sat ever since.
20
This amounts to preIerring the opinion
oI Anton and Bcheler to the authority oI the best witness available oI an L-
class manuscript, a sixteenth-century transcription in the hand oI Pierre
Daniel.
21
Nor is it necessary to Iind an explicit speaker Ior the verse. It is in
the nature oI prosimetry that the narrator can dispense with declarative state-
ments when switching Irom one discourse type to another. Besides, the mov-
ing oI these distichs does not improve the text, since in their new place they
disrupt the continuity oI the passage that was there beIore.
22
Less serious,
perhaps, but belonging to the same type oI arbitrary editorial practices, is the
isolation Irom their proper context oI two other distichs (80.9 vv. 58),
through Bcheler`s introduction oI lines to mark them oII in the printed text.
These unprecedented marks proved to be an inIluential Iactor in one recent
critic`s reading oI the Satvrica, which presupposes a contextual autonomy

20
Pellegrino 1975, however, has put them back in their original place, although, in the Iirst
volume oI his multivolume edition (1986), they have unnecessarily been marked oII by
asterisks to indicate lacunae.
21
See Richardson 1993, 8198.
22
Ascyltos says: 'mihi plane placet emere, quamvis nostrum sit, quod agnoscimus, et parvo
aere recuperare potius thesaurum quam in ambiguam litem descendere` ('OI course I
want to buy it, though it`s ours, as we know, and to recover the treasure Ior a small Iee
rather than to get into an uncertain lawsuit`). |here the verses have been inserted| Sed
praeter unum dipondium sicel lupinosque quibus destinaveramus mercari, nihil ad manu
erat ('But apart Irom one sekel worth two pennies and lupine seeds we intended to use as
money, we had nothing at hand). Fraenkel`s emendation [sicel] quo~ lupinos[que
quibus], accepted by Mller, is too cumbersome. In support oI the received text, see
Schmeling 1992b, 53136. The *+',-. or *+/,-. (Greek Iorms oI the Semitic sekel or
shekel) is a coin. For lupine seeds as Iake money, cI. Pl. Poen. 597 II.; Hor. Ep. 1.7.22II.
10 1 NARRATI VE
Ior the verses.
23
Read through, however, this latter halI oI an eight line ele-
giac poem appears completely sound and Iollows Ireely Irom the context oI
the Iirst part.
In another and perhaps more serious case (55.3), a humorous comment
has been struck out oI the text, and a 'poem has been corrected on metrical
grounds, despite its being presented in demonstration oI a character`s igno-
rance and lack oI versiIying skills. In the Traguriensis manuscript, the sole
witness to the humorous comment, the text runs like this:

. statimque codicillos poposcit et non
diu cogitatione distorta haec recitavit:
quod non expectes, ex transverso Iit
et super nos Fortuna negotia curat.`
Distichon Trimalchionis est cum elego suo:
quare da nobis vina Falerna, puer.`
Ab hoc epigrammate coepit poetarum esse mentio .

At once he demanded his notebook, and aIter a meditation
that was not long he read aloud these distorted verses:
What you don`t expect happens transversely,
and over our heads Fortune takes care oI business.`
This is Trimalchio`s distich, with his elegiac line:
So give us the Falernian wine, boy!`
From this epigram the chat moved on to poets .

Bcheler declared as interpolation the phrase, distichon Trimalchionis est
cum elego suo, but the mocking irony which deIines the Iirst two quite un-
metrical lines oI Trimalchio`s 'distorted (distorta) composition as 'the
distichon oI Trimalchio and the third as 'his elegiac line, and ends by call-
ing the entire composition 'this epigram, makes the case Ior a scribal gloss
seem unlikely.
24
In Iact, the term 'distichon is not being used correctly (as
one would expect in a grammarian) but rather to make it all the more obvi-
ous that Trimalchio`s 'poetics do not conIorm to the rules oI versiIication.
Only the narrator, with his ironic and critical stance towards Trimalchio, is
likely to be responsible Ior this catachrestic use oI a technical term. The im-
plication is that, here as elsewhere, Trimalchio is displaying his tendency to

23
Slater 1990, 13.
24
I take elego (sc. verso) to be the adjective elegus, -a, -um. The singular, although it rarely
occurs Ior obvious reasons, is nevertheless logical enough, and is Iound in the grammarian
Diomedes (502 P, elegum metrum).
1. 1 TEXT, CONTEXT AND I DENTITY 11
do things novo more, just as he does in his mad revisions oI classical my-
thology. The narrator`s interruption oI the delivery oI Trimalchio`s poetry to
make the comment in the present oI the narration, and then to resume the
delivery oI the last line to complete the piece, is the kind oI playIulness
which is in keeping with the tradition oI prosimetry.
25
Although the inter-
jected phrase is in plain language, a lively perIormance (with appropriate
switches in tone oI voice) could make it very amusing. The editors oI the
Satvrica have shown insensitivity to the possibilities oI this Iorm by 'cor-
recting the joke. The text oI the Traguriensis is, in my view, not in need oI
emendation here.
Further changes have been introduced into this short passage. On the
analogy oI Trimalchio`s Iirst poem (34.10), which is composed oI two hex-
ameters and one pentameter (an otherwise unheard-oI arrangement in good
literature),
26
Mller concluded (Iollowing Bcheler, Iollowing Heinsius) that
the end oI the Iirst line and the beginning oI the second had to be missing
(because neither line scans), and so he added words to the text (ubique in the
Iirst line and nostra in the second) in an attempt to make the 'poetry oI
Trimalchio conIorm to the rules oI meter.
27
But even aIter these learned im-
provements, the two lines are not a 'distichon, nor is the composition as a
whole an 'epigram. The reason why only Trimalchio`s Iirst composition
scans, but not his second, could be that the Iirst is supposedly recited Irom
memory, while the second is clearly meant Irom the context to be composed
on the spot.
28
We do not, however, need to prove the soundness oI the hu-
morous comment, merely in order to retain it, Ior the burden oI prooI ob-
viously lies with those who wish to assume the license oI rewriting the text.
29


25
CI. Sat. 79.89, sine causa gratulor mihi ('I congratulate myselI without a reason), the
narrator`s comment in the present tense on the immediately preceding verse. The present
tense comes naturally to the narrator when he reIlects upon the narrative or comments on
the experience oI telling the story (e.g., 70.8, pudet referre quae secuntur |'One is ashamed
to tell what Iollows|). The present oI the narrator reIers to the moment oI telling the story,
while the present oI the characters is, Irom the standpoint oI the narrator, in the past. Sen.
Apoc. 5 oIIers an example oI interruption oI poetry Ior the interjection oI a humorous com-
ment, and then resumption. Sat. 108.14, where a non-poetic verbum dicendi, exclamat, is
inserted into the middle oI a line oI hexameter (discussed below), is another such interrup-
tion in the delivery oI poetry.
26
Examples are Iound in the Anthologia Palatina (13.15), but Barnes 1971, 3034, convin-
cingly argues Ior its subliterary status and unconventionality by noting 'that the metrical ar-
rangement is Iound basically in Iunerary inscriptions on plebeian tombstones.
27
Walsh 1970, 128, rejected the changes and argued Ior the appropriateness oI the verses to
the persona oI 'the poetic tiro Trimalchio.
28
Suggested by Slater 1990, 161 n.11.
29
While editors have Iorced Trimalchio`s non-poem (55.3) to scan, no attempts have been
12 1 NARRATI VE
And there is more oI the same. Several truly adventurous attempts have
been made to reorganize the Iragments oI the Satvrica. Two scholars in par-
ticular, Italo Sgobbo (1930) and Helmut van Thiel (1971), have advanced
inIluential theses proposing to reshuIIle, not just paragraphs, but whole epi-
sodes oI the extant text. The Iormer proposed to move the whole oI the
Quartilla episode on the basis oI his understanding oI the work`s topography,
while the latter suggested the rearranging oI Iragments in accordance with
his hypothesis regarding the synthetic nature oI the Long Fragments (L),
although he admits that his complex hypothesis is ultimately not provable,
and, even iI proven, would not oIIer the necessary guidance Ior rearranging
the Iragments.
30

The last item on our list oI misguided reactions to the problematic state
oI the Satvrica`s text is a sweeping and still very inIluential hypothesis about
supposed 'scribal interpolations in the work. Acting on the assumption that
the L-tradition was put together in Carolingian times Irom badly damaged
sources, Mller in his Iirst edition oI the text (aided and inIluenced by Edu-
ard Fraenkel)
31
hypothesized a learned auctor oI the Long Fragments (L)
who was to have interpolated his own explanations into the text to restore
it.
32
As a criterion Ior spotting these 'Ioreign bodies (Fremdkrper), Fraen-
kel, according to Mller himselI, Iormed 'a very personal conception oI the
elegant brevity and precision oI Petronian prose (eine sehr persnliche Jor-
stellung von der eleganten Knappheit und Pr:ision der petronischen
Prosa), which was arbitrarily made to serve as the touchstone oI textual au-
thenticity.
33
The ominous Tacitean term 'elegant (eleganten) demonstrates
clearly the Iallacy oI authorial intention behind this Iurther attempt to med-
dle with the text. The cult oI the author (a product oI the grammar school)

made to correct or represent typographically as verses the poetic lines oI 99.3, although the
imagery ('incultis asperisque regionibus diutius nives haerent, ast ubi aratro domefacta tel-
lus nitet, dum loqueris levis pruina dilabitur |'The snows cleave longer to rough and un-
cultivated regions, but where the soil made meek with the plough is radiant, the light Ilakes
melt away even as you speak`|) as well as the diction (ast, domefacta, tellus and pruina
meaning snow`) are indeed poetic, while the utterance issues Irom the mouth oI a capable
versiIier, the character Encolpius.
30
van Thiel 1971, 9.
31
Mller 1983, 471I.
32
The assumption that the text has been meddled with has been strangely tempting to scholars
throughout history. The initial response to the rediscovery oI the Traguriensis manuscript
by Marino Statileo in the seventeenth century shows this. Two scholars, Johan Christoph
Wagenseil and Adrien de Valois, argued then, on the basis oI the unclassical Latin spoken
by the Ireedmen, that H was a modern Iorgery. Strangely enough, this was in 1666, beIore
the appearance oI the Iirst actual Iorgery, that oI Nodot (Paris 1694).
33
Mller 1983, 472.
1. 1 TEXT, CONTEXT AND I DENTITY 13
can be counted on to insist that the elegantiae arbiter oI a Roman emperor
must have expressed himselI in 'elegant Latin, and then derive the meaning
oI the term Irom the highly developed rules associated with it by the Italian
humanists rather than its rudimentary deIinition in ancient stylistics.
34
And
this despite the Iact that Tacitus is reIerring to this 'title given the consular
Petronius ironically and never indicates that stylistics have anything to do
with his supposed elegantia, but rather reIinement in perverted pleasures.
35

Incidentally, the term is also used ironically (and with no reIerence to stylis-
tics) in the Satvrica about Trimalchio`s unreIined taste.
36
However, in
Mller`s Iirst edition oI the text (1961), no less than one hundred and IiIty
words and passages (some quite long) were printed in square brackets to
mark them oII as interpolations.
The result was not only a scholarly controversy that severely undermined
the edition (and led many scholars to continue using the older edition oI
Ernout |1923|),
37
but also a text which was very hard to read because oI the
ambiguous status oI the bracketed passages. In several subsequent revisions
oI his edition Mller has drastically reduced the number oI suspected loci,
however, without abandoning the idea entirely. Although in theory the at-
tempt is to correct the text on the grounds oI some impossible esthetic Jor-
stellung about what kind oI style would be most appropriate to an author

34
In Roman stylistics elegantia is a translation oI the Greek term 0',-/1 23-(#43 'choice oI
words. The seminal Renaissance work here is Lorenzo Valla`s De linguae Latinae elegan-
tia libri sex (1444), which laid the rules, regulae, Ior a new understanding oI ancient Latin
stylistics, and was reworked into many oI the textbooks on composition and syntax later
used by the humanist schooling system. On the meaning oI elegantia in the Petronian con-
text, see Dell`Era 1970, 21I.
35
Ann. 16.18: dein revolutus ad vitia seu vitiorum imitatione inter paucos familiarium Neroni
adsumptus est, elegantiae arbiter, dum nihil amoenum et molle adfluentia putat, nisi quod
ei Petronius adprobavisset. unde invidia Tigellini quasi adversus aemulum et scientia vo-
luptatum potiorem. ('Then Ialling back into vice or the imitation oI vice, he was recruited
by Nero to be one oI his Iew intimate associates, as the arbiter oI elegance, while the em-
peror thought nothing charming or reIined in luxury unless Petronius had recommended it
to him. Hence the envy on the part oI Tigellinus, who saw him as a rival and some one
more expert in the science oI pleasure.) Tacitus only once uses the term elegantia in the
stylistic sense (Ann. 13.3) and then about the emperor Claudius; elsewhere it means some-
thing like 'reIined table manners (Ag. 21.3, conviviorum elegantiam), 'good habits (Hist.
3.39, elegantiam morum; Ann. 5.8, morum elegantia), 'tasteIulness in a woman (Ann.
13.46, elegantiam uxoris), or 'respectable liIe style (Ann. 14.19, elegantia vitae).
36
Sat 34.4, laudatus propter elegantias dominus ('the host was praised Ior the reIined ar-
rangements); cI. also Sat. 60.1, tam elegantes strophas ('such sophisticated turns).
37
Coccia 1973, 11, has a list oI reviews oI Mller`s 1961 edition. Coccia`s own book is en-
tirely a reaction to the interpolation hypothesis. On the topic, cI. also Smith 1975, xxiiiiv,
and Mller himselI, who has a survey in Mller 1983, 474I.
14 1 NARRATI VE
reputed to be 'elegant,
38
Fraenkel and Mller nevertheless struck out pas-
sages not only Ior stylistic reasons, but because they contained reIerences
which connected episodes to one another. The continuity oI the story in these
places was considered too artiIicial by the editors.
39
The outcome, oI course,
is a text that is more Iragmented than ever.
Taken together these unbalanced responses to the text oI the Satvrica,
ranging Irom creative and Iantastic restorations, rejected by serious editors,
to cumulative puncturing, reshuIIling oI Iragments, and pruning oI passages,
perpetrated by those same editors, have in common an unjustiIiably high
degree oI conIidence in the possibility oI restoring the textual tradition.
Clearly, we gain nothing by continuing this approach. It is not so much a
question oI being right as oI adopting a sensible working hypothesis. We
have no choice but to respect the text as it has been handed down by tradi-
tion, which means avoiding changes iI at all possible. Filling the gaps and
increasing their number is equally disrespectIul oI textual authority.
40
The
chances oI scoring in the game oI textual correction should be recognized as
slim, unless the context is a wholly predetermined one, as in the case oI the
zodiacal catalogue. We thereIore conclude that a new edition, according to
conservative principles, is still needed.
1.1.2 The Other Text(s)
The troubles besetting the text oI the Satvrica at the hands oI editors have
never deterred interpreters Irom devising critical strategies to solve the per-

38
See Parker 1994 Ior a telling example oI how a simple phrase in standard classical Latin
was misunderstood by three centuries oI editors and translators, aIter its misinterpretation
was Iirst introduced by a seventeenth-century commentator. The case raises a legitimate
concern over whether the sensitivity oI scholars to Petronian Latin is at all up to the task oI
correcting on stylistic grounds the Iragments oI the almost two-thousand-year-old Satvrica.
39
Vital passages in the Quartilla episode which link it to the preceding market scene and the
Iollowing dinner at Trimalchio`s were bracketed. In the 1983 and 1995 editions, 16.3 and
25.2 are still treated as 'interpolations, although 26.7 has been restored to the text.
40
Mller and Ehlers 1983 in two instances curiously enough both identiIy a lacuna, in 108.1
and in 123 v. 236, and then Iill it 'exempli gratia, as they say, by inventing wholesale the
supposedly missing phrases (Mller`s addition runs like this: prae pudore vix eram mentis
compos, praebebat enim foedissimum adspectum'Irom shame I was hardly in control oI
my thoughts; Ior it resulted in a disgusting look; Ehlers adds a whole hexameter: haud
secus hic acuit Martem, formidine victus'in this way he prepared Ior war, overcome by
Iear). One cannot help but wonder whether Mller and Ehlers hoped that their Latin com-
positions would meet with such critical appraisal among scholars as to be incorporated into
the text, making them co-authors oI the Satvrica?
1. 1 TEXT, CONTEXT AND I DENTITY 15
ceived enigma oI this ancient work oI literature. BeIore we introduce Iurther
varieties into the already complex secondary scholarship, a method must be
sought to give a brieI overview oI the modern schools oI thought. In his nar-
ratological study oI Apuleius` Asinus Aureus or Metamorphoses,
41
Jack
Winkler made the pragmatic suggestion that, in philological interpretation,
another text, deemed by the scholar to be oI primary importance Ior the cor-
rect understanding oI the text under studya sort oI Rosetta stone to decode
the main textmight in Iact be the real determinant oI the resulting inter-
pretation. Such a 'comparison text can constitute everything Irom a part oI
a work, to a whole work, or even a cluster oI texts associated with one an-
other, according to some criterion, so as to Iorm a single context. In Iact, a
speciIically highlighted part oI the main text under consideration, a Iragment
Ior example, is oIten used as the 'comparison text. Far Irom being secon-
dary, this supposedly lateral and auxiliary text becomes the primary context
in which the main body oI text is read, at once a master-reIerent and a sup-
plement. As Winkler persuasively demonstrated, distinct primary contexts in
which the work oI Apuleius had traditionally been read could readily ac-
count Ior the Iive interpretations that he listed as main contenders.
This clever idea is a handy tool to account Ior the three readings oI the
Satvrica which have long dominated scholarship on the work. A brieI survey
oI the thrust oI the main arguments associated with these three primary con-
texts will Iacilitate the plotting oI the scholarly landscape. For reasons on
which I shall Iurther elaborate towards the end oI this book, the readings
show a rather obvious correlation to a triad oI trends in the humanities at
large: historicism, Iormalism (or 'pure philology), and the study oI national
literatures. The Iollowing, despite the matter-oI-Iact presentation, are in-
tended as interpretive stances, rather than stated Iacts.

(i) Testimonia. The author oI the Satvrica is commonly identiIied as the Ro-
man consular Petronius who was a close associate oI the emperor Nero, and
whose coerced suicide Tacitus describes in a memorable passage oI the An-
nales (16.1719). Allegedly, at the hour oI death, the elegantiae arbiter en-
tertained himselI with 'light poems and easy verses (levia carmina et
faciles versus) and made it one oI his last tasks to 'write, sign and send to
Nero a catalogue oI his disgraceIul passions, where under the names oI de-
bauched boys and women the novelty oI each sexual act was described (fla-
gitia principis sub nominibus exoletorum feminarumque et nouitatem
cuiusque stupri perscripsit atque obsignata misit Neroni). Tacitus and others

41
Winkler 1985, 7.
16 1 NARRATI VE
also inIorm us oI Nero`s disposition Ior secret explorations oI the brothels
and riotous nightliIe oI the capital. The Satvrica contains light poetry, it has
a brothel scene, albeit Iragmentary, and it is rich in explicit descriptions oI
orgies and riotous parties in Campaniawhere Petronius incidentally died in
his villawhich are vivid enough to give the impression oI Iirst-hand ex-
perience. 'One oI its chieI aims was to advance the author`s standing in the
court circle by appealing to the emperor`s literary tastes,
42
while parodying
the literary products oI personages oI the court circle, notably Lucan and
Seneca, who were Ialling out oI Iavor with Nero.
43


(ii) Fragmenta. The Satvrica contains a copious sample oI poetry composed
in numerous meters, some oI which has been detached Irom the work and is
now Iound in other sources as loose Petronian poems. From a separate
manuscript tradition comes also the Cena Trimalchionis, a long description
oI a dinner party reminiscent oI symposium literature. In addition, we Iind in
the work several rhetorical set pieces; two short erotic narratives, so-called
Milesian tales;
44
and many Iragmentary Iarcical scenes, which in the text are
explicitly compared to mime and comic theater. The Iact that we can take
such Iragments oI the Satvrica and break them out by genre indicates that the
sum oI its parts is no more than a 'large Iramework, or container, into which
|Petronius| could pour . all the wealth oI literary, philosophical, and artistic
expression that was welling up within his Iertile genius.
45


(iii) Asinus Aureus or Metamorphoses. The Satvrica shares with the work oI
Apuleius a predilection Ior seemingly realistic and satirical descriptions oI
low liIe in regions oI the Roman Empire. It so happens that much oI the ex-
tant Satvrica is set in Campania, where two modern archeological sites,

42
Sullivan 1985b, 1666.
43
I have listed this context Iirst, because it is still the dominant one. In the English language
scholarship, J. P. Sullivan`s inIluential work, The Satyricon of Petronius (1968), is the best
representative oI this essentially historicist approach. Sullivan also prepared Ior publication
posthumously the other Iundamental text Ior this reading, K. F. C. Rose`s 1971, The Date
and Author of the Satyricon.
44
According to a persistent misunderstanding oI the term Milesia |sc. historia sive fabula|,
which does not, in Iact, denote 'short stories or 'novelle; see Iurther section 3.2.3.
45
Ben Edwin Perry, The Ancient Romances (1967), 205. This reading was perhaps originated
by Albert Collignon in his Etude sur Petron. La Critique litteraire, limitation et la parodie
dans le Satiricon (1892), but it was given its present and more dogmatic Iorm by Martin
Rosenblth in his doctoral dissertation, Beitrge :ur Quellenkunde von Petrons Satiren
(1909). Rosenblth treats the Satvrica only in pieces (Stcken) which he compares with
various genres to support the claim that Petronius was an 'original artist and his work a
'synthetic composition.
1. 1 TEXT, CONTEXT AND I DENTITY 17
Pompeii and Herculaneum, have greatly improved our knowledge oI the
material and linguistic (through graffiti) aspects oI daily liIe in Roman It-
aly.
46
The two Roman novels are the only extant comic and satiric prose
Iictions Irom antiquity, and they can be seen to Iorm a contrast with the ex-
tant sentimental Greek romance. Moreover, the peculiar mixture oI prose and
poetry in the Satvrica is the same as that used by Seneca in the Apocolocvn-
tosis, a Iorm introduced to Roman literature by Varro in his lost Saturae
Menippeae. The genre oI the Satvrica was thereIore, as was said by Quintil-
ian about hexameter satire, tota nostra Ior the Romans, and displayed a dis-
tinctly 'native or 'national geniusMller calls it 'der italische Wille zu
individueller Gestaltung
47
which can be seen to parody, amongst other
things, a Greek penchant Ior the Iabulous.
48
As a genre the Roman Comic
Novel has Iound its Nachleben in the Spanish picaresque novel.
49


The three contexts above, all oI which have a long-standing and well estab-
lished relationship with the Satvrica, do not oI course constitute a complete
list; nor are they necessarily incompatible, since one regularly Iinds two or
more used by the same interpreter, although with varying priority. All three
are possible readings oI the work. (I grant that my presentation oI them may
not show them in their best possible light.) What I wish to emphasize, how-
ever, is that whatever context we choose to put the Satvrica in as we read
itand Ior whatever reasonthis context will inIluence, iI not determine,
our interpretation. This is especially true with respect to the vexed problem
oI Iinding a suitable generic label Ior our work.
A particularly interesting variation on the use oI master contexts is Iound
in two important book-length studies, The Satyricon of Petronius by J. P.
Sullivan (1968) and Reading Petronius by Niall W. Slater (1990). In both

46
The archeological context (Pompeian graIIiti) was made use oI Iirst by Arminius von
Guericke, De linguae vulgaris reliquiis apud Petronium et in inscriptionibus parietariis
Pompeianis (1875), and more widely by Amedeo Maiuri, La Cena di Trimalchione (1945),
who was the director oI the archeological site at Pompeii.
47
Mller 1983, 496. Theodor Mommsen 1878 is certainly the ideologue behind the crudest
Iorm oI the nationalist interpretation oI the Satvrica, which incidentally is only one aspect
oI his general attempt to read ancient Roman literature as 'national literature. See section
3.2.2.
48
A by now classic example oI this reading is P. G. Walsh`s The Roman Novel (1971, repr.
1995). Other works oI major importance Ior this reading include Albert Collignon`s Etude
sur Petrone (1892), and Ettore Paratore`s two volume edition Il Satvricon di Petronio
(1933).
49
A recent development in Spanish studies, however, is the recognition oI the positive critical
reception oI Heliodorus in the Renaissance, and his importance as a model Ior picaresque
Iiction in Golden Age Spain (Forcione 1970).
18 1 NARRATI VE
studies a programmatic statement is extracted Irom a short selected passage
oI the work itselI. It is assumed that in this passage the author bypasses the
'unreliable persona oI Encolpius to explain his true purposes in writing the
work. Suddenly we are no longer in the proper context oI the work, but have
been transported into the historical context. As such the chosen passage is
granted a categorically diIIerent status Irom the rest oI the text and gives, so
to speak, a green light Irom the author to the scholars to go ahead with a
certain interpretation. II we consider closely what it is about these small bits
oI the whole workin both cases we are dealing with a couple oI elegiac
distichsthat makes them susceptible to this reading, the reason appears to
be the occurrence oI the words opus (132.15), 'deed or 'work, in Sulli-
van`s passage,
50
and!pagina (80.9), 'page, in Slater`s.
51
Each term is under-
stood by the respective scholar to designate the Satvrica itselI. However, as
the reader oI the work will recall, the word opus (132.15) does not necessar-
ily stand Ior the work as a whole, but is more naturally taken in the immedi-
ate context as reIerence to a comic oration addressed by Encolpius to his
penis (132.8I.).
52
The same is true oI the word pagina (80.9), which does not
necessarily designate a page in the Satvrica, but more obviously Iorms part
oI what we may loosely term 'poetic circumlocution. The phrase ubi riden-
das inclusit pagina partes'when the page closes upon the laughable roles
would be another way oI saying 'when the role-playing recitation ends. The
immediate context is Encolpius` betrayal by his two Iriends and lovers, Gi-
ton and Ascyltos, leading him to a state oI disillusionment which makes their
Iriendship seem, retrospectively, a Iarce that has ended abruptly. Thus there
is in both cases another possible reading, which does not require the peculiar
allegorical hermeneutics used by these scholars. However, their two quite
diIIerent readings are not really dependent on being correctly 'authorized
by Petronius.
53
The present analysis merely transIers responsibility Irom
Petronius to the real authors.
To escape the methodological dilemma oI excessive reliance on privi-
leged passages and context, I propose in principle to treat the whole text oI
the Satvrica itselI as its own privileged context. Instead oI searching Ior the

50
Sullivan 1968, 9899, is here under the inIluence oI Heinz Stubbe 1933, 15053, who like
him considered the phrase 'a work oI simplicity simplicitatis opus (132.15) a direct reIer-
ence to the Satvrica and sought to support his argument by pointing out that opus could also
mean 'genre, according to Quint. Inst. 10.1.67II.
51
Slater 1990, 13.
52
This was pointed out by Beck 1973, 51.
53
Sullivan`s reading is oI the historicist type, whereas Slater`s is oI the philologico-Iormalist
variety with a strong emphasis on drama as the dominant genre in the 'literary medley that
is, supposedly, the Satvrica.
1. 1 TEXT, CONTEXT AND I DENTITY 19
'basic idea oI the work in another text or generic system oI texts (history,
epic, drama, satire, realistic novel, picaresque novel etc.), we look Ior it in
the text itselI. In the light oI general claims about the uniqueness oI our
work, this ought to allow us to discover something which is not easily Iound
elsewhere. Nevertheless, I do not intend to deprive my reading oI external
contexts, so long as they are kept secondary to the main text; nor can I guar-
antee that I shall give equal weight to every passage in the work. No reading
can really take place in a contextual vacuum. The issue is not whether we
read in context, but our choice oI context, the quality oI our intertextual
analysis, what we do with the context and whether we properly respect tex-
tual boundaries. Hence the three traditional contexts and their implications
Ior the reading oI the Satvrica will oIten be reIerred to in this study, while
new 'comparison texts will also be introduced to provide Iurther contextual
support to allow a revision oI the traditional intertextual relationships, and
thus to Iully enable a new reading oI the work.
1.1.3 Masks and Faces
The Iocus oI my reading will inevitably be on how the person oI the narrator
tells his story, and how this story Iorms a coherent whole, meanwhile al-
lowing the mysterious Iigure oI the historical author to retire into the back-
ground, behind the mask where he originally chose to stay. Many scholars,
on the contrary, seem to sense the authorial presence oI Petronius behind the
words oI Encolpius, who is regularly described in terms such as 'doublure
(Veyne), 'puppet-actor (Perry), 'ambiguous (Sullivan), or 'chameleon
(Walsh). Slater even announces his discovery oI 'the absent presence oI a
narrator which according to him is 'the essence oI Petronian comedy.
54

What the paradoxical language seems to be telling us is not to recognize
Encolpius as a person in his own right, but rather to watch in his every move
Ior the signs oI Petronius lurking behind a transparent mask. It may be ques-
tioned, however, whether a 'transparent mask is properly a disguise at all,
Ior such a mask does not conceal the identiIying signature oI the Iace and
substitute new Ieatures. In discussing artistic representation we conIlate an
important distinctionwe risk losing sight oI the important Iunction oI mi-
mesisiI we allow degrees and transparency oI identity, and speak oI a per-
son as 'partly identical to another. Although we are admittedly in proximity
with perhaps the most diIIicult philosophical problem oI modern times, the
problem oI the subject, Irom an anthropological point oI view an explanation

54
Slater 1990, 173.
20 1 NARRATI VE
Ior the ability to eIIect a change in identity through artistic representation
rests on a complex convention agreed on by perIormers and audience alike.
To avoid digressing into the anthropological aspect oI the problem, we may
simply describe the general cognitive act oI identiIying the representing
'this with the represented 'that through the compact Iormulation provided
by Aristotle: -5-3 6#& -7#-. 0'893-.'so that this one is that one
(1448b17).
55

Cast in pragmatic and textual terms, identities and proper names can be
described as transitional signiIiers, which carry us Irom one context to an-
other. Suetonius relates the Iollowing anecdote about Nero, the perIormer,
which illustrates well the semantic and contextual power oI identity.

|Nero| sang tragedies wearing the masks oI heroes and gods, and even oI
heroines and goddesses, having the masks Iashioned in the likeness oI
his own Ieatures or those oI the women oI whom he chanced to be en-
amored. Among other themes he sang Canace in Labor`, Orestes the
Matricide,` The Blinding oI Oedipus` and The Irenzy oI Hercules.` At
the last named perIormance they say that a Ireshman recruit, placed as
soldier on guard at the entrance, upon seeing the emperor in mean attire
and bound with chains, as the plot demanded, rushed Iorward to lend him
aid.
56


Critics who reIuse to take Encolpius at Iace value risk making the mistake oI
the young recruit. They break the spell oI Iiction and ruin the eIIort oI the
artist. It is true that, in Suetonius` anecdote above, Nero is deliberately play-
ing a game with identities by having the male masks made in his own like-
ness and the Iemale masks in the likeness oI a Iavorite mistress, but Nero`s
conviction that his exceptional status in the world put him on par with heroes
and deities did not alter the Iact that his unhappy identiIication could not be
accommodated within the context oI the myth oI Hercules.
57
It is all very

55
See Nagy 1989, 47. A useIul discussion oI masks as identities in perIormance is provided
by Calame 1995, 97115, who also supplies ample reIerences to the relevant anthropologi-
cal and semiotic literature.
56
Suet. Nero 21, Tragoedias quoque cantavit personatus heroum deorumque, item heroidum
ac dearum, personis effectis ad similitudinem oris sui et feminae, prout quamque diligeret.
inter cetera cantavit Canacen parturientem, Oresten matricidam, Oedipodem excaecatum,
Herculem insanum. in qua fabula fama est tirunculum militem positum ad custodiam aditus,
cum eum ornari ac vinciri catenis, sicut argumentum postulabat, videret, accurrisse feren-
dae opis gratia.
57
Nero`s playIulness in the perIormance oI Hercules says more about the perIormer than it
does about the story oI Hercules itselI, just as Curiatius Maternus` passionate impersonation
1. 1 TEXT, CONTEXT AND I DENTITY 21
well to have the Greco-Roman hero dressed in mean attire and bound with
chains 'as the plot demanded (sicut argumentum postulabat), but that is no
way to treat the man who bears the supreme identity oI Caesar. The young
military recruit, by virtue oI his youth and the values oI state which he had
imbibed through his military training, possessed the kind oI simplicitas that
was valued at Rome. So simplex was he that he read everything in the same
context, the oIIicial context oI Rome. His inability to restrain himselI Irom
rushing to Caesar`s aid has the eIIect oI valorizing the context oI Nero as
Caesar, and making unreadable the context oI Nero as Hercules. This is the
point oI Suetonius` anecdote, Ior the imperial biographer`s agenda is also to
privilege the oIIicial context, and to condemn the man Nero Ior his unstately
theatrical escapades while in the oIIice oI the Roman emperor.
58

Classical authors and storytellers have always known how to exploit the
semantic magic oI personal identities. Recognition scenes in Greek literature
are based on the principle that identity is key to the meaning oI words said
and deeds done, as Sophocles` play, Oedipus Rex, so cleverly dramatizes. On
a smaller and less artistic scale, we have seen earlier how the words opus
(132.15) and pagina (80.9) in the Satvrica are given an entirely diIIerent
range oI reIerence, depending on who is 'recognized as their source. The
unhappy anagnorismos oI the historical author as the speaker oI those words
is enough to annihilate their immediate context and replace them in another
context appropriate to the person oI Petronius. Reading the Satvrica as a
statement by Petronius is disruptive oI the story, even iI some oI 'the masks
were Iashioned in the likeness oI historical personages.
What historicists accomplish by reading Petronius and his emperor
Iriend into the Satvrica is to read the text in the primary context oI the testi-
monia, without respecting the boundary between text and context. What
remains oI these historical personages is made oI written 'text, the same
material as the Satvrica. By 'recognizing Petronius in Encolpius we indis-
criminately Iuse the Annales with the Satvrica. Such contextual Iusion does
admittedly have an illustrious pedigree. Ancient Greco-Roman grammarians,
it seems, invented the allegorical hermeneutics oI the historico-biographical
reading. Servius` readings oI Virgil`s Ecloguesshort personal narratives
some oI themrecognized and identiIied the communiques oI Virgil on the
subject oI his own liIe and times in the words oI shepherds and goatherds

oI Cato, in reciting his historical tragedy, revealed something about his own political stance,
i.e. that he had adopted that oI Cato; the ideas oI Cato himselI were well known beIore
(Tac. Dial. 2.1). We should distinguish between speciIic interpretive perIormances and the
general requirements oI the story that its own logic be respected.
58
CI. also D.C. 63.9.4II.
22 1 NARRATI VE
with names like Meliboeus and Corydon.
59
We will encounter Iurther sam-
ples oI this style oI criticism as this study progresses, Ior most oI the Sa-
tvrica`s external Iragments come in the awkward wrappings oI the grammar
school.
Paul Veyne was perhaps the Iirst late modern scholar to address directly
the problem oI identity in the Satvrica. He summarizes his conclusions thus:
'dans la Cena, Encolpe est le porte-parole de Petrone, dans le reste du ro-
man, il est son alibi et lui permet de prendre ses distances sur le genre
mineur qu`il pratique.
60
It is hard to resist the thought that Veyne`s arbitrary
splitting oI Encolpius` person might be motivated by some institutional
sense oI propriety. Petronius` ventriloquial narration oI Trimalchio`s dinner
partythrough the dummy Encolpiusis tolerable, because in this, the least
obscene part oI the story, the protagonist is represented as alooI, while the
narrator increasingly expresses a critical and mocking attitude towards the
tasteless parvenus. However, in the rest oI the story, where shameless ob-
scenities Ily thick and Iast Irom the mouth oI the narrator, the Roman con-
sular Petronius must be seen to take appropriate distance, and moral respon-
sibility is leIt with the dummy.
The most recent attempt to salvage the Petronian presence in the words
oI Encolpius is Gian Biagio Conte`s 1995 Sather Lectures published under
the title The Hidden Author. An Interpretation of Petroniuss Satyricon.
Conte`s lectures contain an excellent description oI the strange literary and
cultural pathology inIecting the world oI the Satvrica but they Iail, in my
view, to accomplish their titular project oI elucidating the nexus between
narrator Encolpius and his absent or 'hidden author. In Romantic and post-
Romantic literary criticism, subscribed to by Conte in the preIace, a work oI
literature is 'intended as the expression oI an author`s subjective sensibility.
However, iI the author is not there in the text, as Conte argues Ior the Satv-
ricaa premise which has been gaining acceptance since it was Iirst pro-
posed by Roger Beck in 1973
61
and iI getting at the author`s intention nev-
ertheless 'is the only way oI retrieving the overall meaning oI the text,
62
it
Iollows that the meaning oI the text is in Iact irretrievable and just as 'hid-
den as the author. But Conte does not give up this easily and instead over-

59
On autobiographical Iallacy in criticism oI bucolic poetry, see Reed 1997, 49. For some-
thing analogous in modern Petronian scholarship, see Rose`s 1971, 55, suggestion that
Petronius was educated and spent his youth in Massalia. Slater 1990, 10, accepts the sug-
gestion.
60
Veyne 1961, 301.
61
Beck 1973, 4261.
62
Conte 1996, viii.
1. 1 TEXT, CONTEXT AND I DENTITY 23
comes the aporia with metaphysics oI intentionality that could be termed
transcendental. It happens in a Iew subtle 'moves that are taken gradually
and must be tracked down in order to be Iully comprehended. First, Conte
interprets the author`s absence as his passive aggressive intention, '|a|s a
deliberate strategy, the author reIuses to play a direct role in the account,
preIerring to make himselI a detached external observer, like the reader
whose complicity he is seeking.
63
Thus '|a|lerted by Petronius, the reader
interprets the text in open conIlict with the way in which the narrating char-
acter perceives and reports it,
64
and then 'assumes Ior himselI the ironizing
attitude oI the author, until '|t|he reader`s smile . makes explicit the au-
thor`s implicit voice, a voice that would otherwise be bound to silence in a
text in which the narrator`s 'I ostensibly conducts the entire narration.
65

Finally, the reader (Conte himselI) is empowered to speak on behalI oI the
author, revealing the 'hidden intention behind the Satvrica.
66
Thus, it
doesn`t really matter that the author is 'hidden, absent Irom the text, or
even dead, as the poststructuralists would say; his intention has been trans-
mitted to the next oI kin, his authoritative reader. Predictably, Petronius
redivivus, alias Conte the reader, is stuck in the Iamiliar time warp oI
Petronian scholarship and endorses most oI the theories about him by the
German philologists who invented him over a century ago (see section
3.2.2).
As Philippe Lejeune, the French theorist oI autobiography, concedes
with respect to his concept oI le pacte autobiographique,
67
there exists no
method to distinguish between autobiography and the autobiographical novel
on the basis oI internal textual evidence: '|A|ll narrative in the Iirst person
implies that the protagonist, even iI some distant adventures about him are
being told, is at the same time the real person who produces the narration.
68

What deIines the status oI the personal narrative with respect to the external
world is merely the identity or lack thereoI oI the writer`s proper name, on
the title page, and the name oI the narrator and protagonist, in the narrative.
An inIamous case involving the learned Byzantine scholar Photios pro-
vides an excellent illustration oI Lejeune`s point.
69
Photios had two Greek
Ass-Stories in Iront oI him, a :8#"-%);*8&. in several volumes, and a

63
Conte 1996, 267.
64
Conte 1996, 29.
65
Conte 1996, 734.
66
Conte 1996, 150.
67
Le Pacte autobiographique` (orig. edn. 1975); translated in Lejeune 1989, 330.
68
Lejeune 1989, 25.
69
Phot. Bibl. Cod. 129. My discussion oI this passage is inspired by Winkler`s 1985, 253-255.
24 1 NARRATI VE
<-='&-. > ?@3-. in a single book. They were both the narratives oI Loukios
oI Patrai, and /A8& BC D E'"#A%-$ ,F/-. G,"*(#43 C3 $H&'I3, J%%K#--
G-&L". BC "M*N%O.'each was stuIIed with Iabulous stories and shameless
obscenity. Judging Irom internal textual evidence the only notable diIIer-
ence was the length. Photios seemed hesitant whether to declare the shorter
an abbreviation oI the longer or the longer an expansion oI the shorter (the
shorter happens to be still extant, but the longer is lost). However, he was
convinced that Lucian had written the shorter version and was equally cer-
tain that Loukios oI PatraiPhotios calls him P,,-. <-$'&"3F., 'another
Lucianhad written the longer version. This inIormation may have been
derived Irom the title pages oI the MSS to which Photios had access.
70
Bear-
ing in mind the two authorial identities Ior virtually the same story (whether
the shorter was an abbreviation oI the longer or the longer an expansion oI
the shorter, the diIIerence was only in detail and length), Photios interprets
the intention oI each writer thus:

D C3 <-$'&"3Q. *';G#43 '"R B&"*=%43 #13 B8&*&B"&-3+"3, S*G8% 'J3
#-9. P,,-&., '"R #-T#-3 *$3A#"##83. D BC <-='&-. *G-$B(U43 #8 '"R
G&*#(. 3-+U43 #V. 0W J3H%;G43 8M. J,,X,-$. 8#"-%);*8&. #(. #8 0W
J,F/43 8M. J3H%;G-$. '"R J3(G",&3 '"R #Q3 P,,-3 #I3 G","&I3 =H43
YH,-3 '"R ),X3")-3

Lucian designed his work to mock and ridicule Greek superstitions, as he
does in his other writing, whereas Loukios in all seriousness believed in
transIormations oI one human being into another, and into animals and
back again, and the other nonsensical babble oI ancient myths.

The same story means two diIIerent things according to whose name is on
the title page, i.e., according to whether the nave persona oI Loukios is
taken at his word, or whether there is a notorious jester hiding behind the
mask, making the words oI the simple persona acquire a double meaning.
The signiIicant moment is when Photios points to what Lucian 'does in
his other works. This contextual reIerence transports the ass-story into the
context oI Lucian`s other works and gives it a similar meaning. Without that
context, and without the proper name oI Lucian on the title page, Loukios is
king oI his story. In Iact, despite many brave attempts, modern scholars have
not been able to determine with certainty whether Loukios oI Patrai, the
author oI the :8#"-%);*8&., is also a character in the master-context oI res

70
For a possible title-page with the name oI Loukios oI Patrai, see Mason 1994, 1669, n.16.
1. 1 TEXT, CONTEXT AND I DENTITY 25
gestae, the historical context, or merely existed in the text oI his personal
story.
71
Likewise, iI we had no inIormation about who wrote the Satvrica, we
might well be debating today whether Encolpius, a Latin speaking Graecu-
lus, actually existed, and iI so, what he intended by his autobiographical
narrative.
Fortunately, this is not the case and it seems, indeed, possible to deter-
mine with certainty that Encolpius is not a character in the historical context.
The reason is that another`s signature is Iound on the edge oI the work. This
edge, unlike the text itselI, overlaps with the historical context. Provided we
recognize the author designated by the name oI Petronius Arbiter, as the
Tacitean consular Caius Petronius, who played the role oI elegantiae arbiter
Ior Nero, we have a legitimate reason to dip the edge oI the Satvrica into a
vast ocean oI historical context. But let us remember, no more than the edge.
'Petronius Arbiter and 'Encolpius are two diIIerent proper names embed-
ded in two distinct contexts, each standing on the opposite side oI the
boundary between history and Iiction.
But what, then, somebody may protest, oI the oIt noted similarity be-
tween the historical Petronius and the protagonist Encolpius, e.g., their mu-
tual quality oI simplicitas in word and deedwhich Tacitus in Iact suspects
to be faux in his Petronius?
72
This quality has sometimes been used as a
ground to identiIy the two men. However, estimating the similarity between
two objects is only possible by measuring the extent to which they both
share the quality oI some third object or principle which links them, and so
the establishment oI similarity constitutes a diIIerent cognitive act Irom that
oI identiIication.
73
In this case the mutual element is simplicitas, a quality
shared by both Iigures, although not oI course exclusive to the pair oI them.
Scholars have sometimes recognized in Tacitus` thanatography oI Caius
Petronius a quotation Irom Sat. 132.15.
74
The thesis oI Tacitus 'quoting

71
For a survey and extensive bibliography oI the scholarship on the Greek and Roman Ass-
stories, see Mason 1994, 16651707. Apuleius clearly did not take Loukios oI Patrai to be a
historical person as is shown by his preserving Loukios/Lucius as a narrator oI his
Metamorphoses. He even changes the narrator`s biography, makes him get involved with
the cult oI Isis and Osiris and move to Rome to become an advocate in the courts there.
This is not the manner in which to treat some one`s actual biography.
72
Tac. Ann. 16.18.
73
The essential diIIerence between similarity and identity in the classical context is discussed
by Trimpi 1983, 16675.
74
132.15, 'quid me constricta spectatis fronte Catones / damnatisque novae simplicitatis
opus? / Sermonis puri non tristis gratia ridet, / quodque facit populus, candida lingua
refert ('Why look at me Irowning, you Catos, and condemn a deed oI new simplicity? It
is the grace oI pure not pompous language that smiles; whatever people do my candid
26 1 NARRATI VE
Petronius would be proven, iI all three oI the Iollowing conditions were Iully
satisIied: i) the identiIication oI the two Petronii is correct in the historical
context; ii) Tacitus knew and had read the Satvrica, at least as Iar as book
sixteen, assuming he started Irom the beginning; iii) Irom the vast text oI the
original, Tacitus picked out a single word Irom a single poem, which he
recognized as the direct communique and programmatic statement oI
Petronius, anticipating Heinz Stubbe (1933) by almost two millennia.
But let us grant, Ior the sake oI argument, that all three conditions have
been satisIied. The consequences are not encouraging Ior those who had
hoped to Iind the historical Petronius, since we would have to declare him
contaminated by context with the Satvrica`s wanton characters. For we must
ask to what extent the reading oI at least sixteen books oI the lascivious Sa-
tvrica inspired Tacitus` bizarre account oI Petronius, who had been dead Ior
over halI a century when the section was penned?
75
We must ask how much
oI Encolpius has Iound its way into the character delineation oI Petronius?
Instead oI historicizing Encolpius, we may end up Iictionalizing Petronius.
The counter-intuitive argument that I have made in this introductory
investigation can be summarized thus: The Satvrica means one thing iI En-
colpius is identiIied as the speaker, and it means something entirely diIIerent
iI we identiIy the speaker as Petronius, Ior the simple reason that diIIerent
contexts come into play in these distinct readings. At this moment in the
reception-history oI the work we have accumulated countless readings on the
authority oI Petronius, this having been the subject oI 'Petronian scholar-
ship.
76
Rarely has the person oI the narrator and protagonist been taken
seriously enough Ior there to have been an interpretation oI the Satvrica on
the authority oI Encolpius; this is thereIore what we shall try to accomplish
in the present study.

We will proceed Irom considerations oI the contextual links that have been
Iorged between the Satvrica and other worksin several places attempting
to revise the traditional readings by re-interpreting old 'comparison texts
and introducing new onesto analysis oI the proper context oI the work, the
whole extant text oI the Satvrica. By Iocusing on the Iigure oI the narrator at
various levels oI analysis ('perIormance, 'discourse, 'story, 'narrative
and 'genre) the re-integration oI the Iragments is brought about through

tongue relates.`) Bogner 1941, 2234; and Rankin 1971, 1068.
75
Caius Petronius died in 66, and according to Syme 1958, 473, 'nothing Iorbids the assump-
tion that Tacitus was writing as late as 120, or even 123. The chapter about Petronius
comes in the antepenultimate year oI the Annales, Tacitus` last work.
76
Beck 1973 and 1975 is an exception to the rule.
1. 1 TEXT, CONTEXT AND I DENTITY 27
recognizing the source oI the whole in a single voice. In this mimetic and
desultory voice, moreover, one encounters multiple impersonations and a
roster oI traditional moods and generic stances providing diverse impacts at
diIIerent moments in the story. All autobiographical narratives imply that the
person who produces the narration is as real as the protagonist himselI, and,
although the Satvrica has traditionally been read in historical context as the
narrative oI Petronius, the proper name and person oI Encolpius distin-
guishes him Irom the author and provides the dominant organizing principle
in the context oI the story.
Part One analyses the act oI narrating the Satvrica, as opposed to the
story told in the narrative, which will be treated in Part Two. The implica-
tions oI the three traditional contexts on the question oI perIormance are
discussed and a new model oI the perIormance is advanced. This new model
is used to account Ior such Ieatures oI the work as the subordinate narrators,
the colloquial language and the variety oI discourse types. Finally, we relate
the critical and esthetic questions raised by the characters in the story to the
plurality oI Iorm in the whole oI the Satvrica.
Part Two attempts to reconstruct the story and logic oI the 'recollec-
tions oI Encolpius on the basis oI the preserved text. As an aid to the recon-
struction, the linear progression oI Encolpius` travelogue will be used to map
allusions in the extant text and the Iragments onto the topographical grid oI
the story. In section 2.2, this reconstructive work will be continued with a
special emphasis on the utterances oI young Encolpius in soliloquies and
colloquia personarum, where much oI the inIormation about lost parts oI the
Satvrica occurs. Special emphasis will be placed on reIuting the attempts oI
scholars to trivialize the signiIicance oI this important material. Section 2.3
contains a detailed creative summary or argumentum oI my Iindings, to rep-
resent in as concise a manner as possible the proper context oI the Satvrica.
The summary, which itselI is in the narrative Iorm oI a personal 'recollec-
tion, draws on the conclusions oI the two preceding chapters, and aims at
showing the conventionality oI the plot compared to other ancient erotic
Iictions.
Section 3.1 provides a new deIinition oI the genre oI the Satvrica based
on the narrative Iorm, and a critique oI the current idea oI the work as simul-
taneous narrative, by establishing a categorical diIIerence in the temporal
and cognitive status oI the protagonist, on the one hand, and narrator on the
other, as seen in the slow progress oI the protagonist Irom ignorance to
knowledge, in contrast to the narrator`s complete knowledge oI the story, as
demonstrated by his ability to narrate it. This section also analyses the
'moral message oI the genre, and Encolpius` social and moral status with
28 1 NARRATI VE
respect to his audience, which determines his comic narrative stance and puts
constraints on the manner in which he can deliver his satire.
Section 3.2 Iinally attempts to place the Satvrica in the literary-historical
context. By surveying the early modern scholarship on the work, we con-
clude that contemporary ideologies exerted an inIluence on the scholarship,
to the detriment oI our understanding, and that a revision oI the belieI in
Petronius` 'Italian characteristics and 'originality is called Ior, especially
in view oI recent discoveries oI Greek papyri which show close aIIinity oI
tone and Iorm with the Satvrica. Revisiting the camp oI the apparently de-
Ieated traditionalists among late nineteenth-century German scholars, we
restate their case, and supplement earlier arguments with new ones oI our
own making.
1.2 The Desultory Voice oI Encolpius
1.2.1 Narrative through Impersonation
The ancient philosophical and rhetorical theory oI narrative, though oIten
neglected by classicists and literary critics alike, arguably oIIers better tools
Ior the study oI ancient narration and narrators than does the modern disci-
pline oI narratology. The ultimate reason Ior its excellence lies in the diIIer-
ent goals and practices oI ancient literary production. While the ancient theo-
rists were attempting to explain a literature composed Ior vocal reading and
public delivery, modern narratologists have naturally seen their task as that
oI studying printed texts read silently by a solitary reader. These pragmatic
diIIerences are reIlected in the usage oI terminology and central paradigms.
77

For the ancient rhetoricians the person producing the narrative, and the cir-
cumstances surrounding that production, are always primary, whether it is
the author or a Iictional narrator, while the modern approach tends to privi-
lege the story, and attempts to work backwards Irom it to the narrator, build-
ing a complex typology oI narrators based on their connection with the char-
acters oI the story.
An early and problematic Iormulation in modern narratology regards the
use oI the grammatical person as the basis to establish a typology oI narra-
tors. The terms 'Iirst person narrator and 'third person narrator derive
Irom such early Iormulations. Petronian scholarship has made use oI the
modern terminology, apologetically, at least Irom the end oI the nineteenth
century.
78
However, as most narratologists have come to understand, the
grammatical Iirst person as such does not mark a narrator in any but the most
generic way, i.e., as 'the speaker. All speakers can reIer to themselves by
using the Iirst person, and its mere articulation does not change in the least
the status oI the discourse. Without Iurther inIormation the linguistic sign,


77
The ancient technical terms oIten derive Irom the theater (e.g., persona, 'mask); the mod-
ern are oIten borrowed Irom Iormal grammar (e.g., 'Iirst person narrative) or the criticism
oI painting and photography (e.g., 'point oI view).
78
Klebs 1889, 631I, '|i|n den erhaltenen Stcken erzhlt berall Encolpios in der ersten
Person |.| Wir drIen demnach mit Bestimmtheit sagen, da das Ganze, um den nicht
eben schnen, aber gebruchlichen modernen Ausdruck anzuwenden, in der Form des Ich-
Romans gehalten war |my underlining|.
30 1 NARRATI VE
'I, is devoid oI identity. The diIIerence between so-called 'Iirst person
narratives and 'third person narratives results Irom the identiIication, or
lack thereoI, oI the narrator as a principal character oI the story. In so-called
'Iirst person narratives, which usually take the Iorm oI recollections, we
can distinguish between two meanings oI the Iirst person: 'I the speaker,
and 'I the protagonist. Although terminology is not oI paramount impor-
tance (so long as we understand the crucial Ieatures oI the Iorm) a better way
to reIer to such narratives might be to call them, as I have already done
above, 'personal recollections.
79

The advantage oI the ancient rhetorical theory is perhaps most obvious in
the way it accounts Ior the narrative phenomenon oI utterance within utter-
ance. Modern narratology, because it privileges the story, must treat the sec-
ond discourse as 'quotation, i.e., as a practically autonomous speech act,
which existed prior to its later quotation by the narrator during the telling oI
the story. Rhetorical theory, on the other hand, since it uses paradigms Irom
perIormance literature, casts the problem as a question oI identity, persona.
And so the diIIerence between the main discourse and the discourse within
that discourse becomes a matter oI mimetic change in the identity oI the
speaker.
80
Mimesis as impersonation, or sermocinatio, unlike the modern
idea oI 'quotation, is mostly indiIIerent to concerns about historicity or
verbatim accuracy oI the utterance, which it tends to replace with qualitative
criteria such as the appropriateness and aptness oI the speech with respect to
the speaker and the circumstances when the words were uttered.
81
Another

79
This term bears a superIicial resemblance to the terminology oI Stanzel 1964, 'der person-
ale Roman, Iollowed by EIIe 1975 with some reservation. But Stanzel`s 'personal novel
is modern (late nineteenth century) and is characterized by the apparent absence oI a narra-
tor, i.e., an impersonal narrative stance. This is because Stanzel, like many modern narra-
tologists, sees only the characters as personae, behind whom the author in his 'personal
type somehow hides by means oI a scenic presentation, and Stanzel accordingly Iails to
consider the narrator as the central narrative persona.
80
Arist. Po. 1448a, &89*H"& Z*#&3 . JG"//A,,-3#" [#8%F3 #& /&/3F83-3 ('mimesis is .
when the narrator becomes some one else); Isid. Etvm. 2.21.32, ethopoeia est, cum ser-
monem ex aliena persona inducimus ('ethopoeia is when we introduce speech Irom a per-
son other than ourselves); Hermog. Prog. 9, \H-G-&+" Z*#& +K*&. ]H-$. ^G-'8&A3-$
G%-*;G-$, -5-3 #+3". _3 8`G-& ,F/-$. ab3B%-(NK 0GR cd'#-%& ('ethopoeia is an imitation
|mimesis| oI the manners oI an assumed person, Ior example words that Andromache might
say to Hector).
81
Prisc. Praeex. 9, ubique autem servanda est proprietas personarum et temporum. alia sunt
enim verba iuvenis, alia senis, alia gaudentis, alia dolentis . habeat autem stilum supposi-
tis aptum personis ('One must everywhere maintain the suitability oI persons and times: the
words oI a young man are diIIerent Irom those oI an old man, those oI a happy man still
something else, and those oI a grieving man . let the style suit the assumed person); Isid.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 31
modern concept, 'embedded narrative, likewise gives the impression that
the secondary discourse had a prior existence as an autonomous statement,
and was only later incorporated into the main discourse. Since this is neither
true oI Iictional narratives, nor oI ancient historiography, the advantages oI
the ancient model are obvious.
According to extant rhetorical manuals, sermocinatio, or speech within
speech, is a stylistic device related to 'vividness, evidentia or 03(%/8&", and
may be classiIied among Iigures oI speech, which contribute to the emo-
tional impact oI narrative.
82
Although it is at times possible to provide such
detailed verbal descriptions oI characters and events that the reader or lis-
tener may experience the illusion oI almost 'seeing what in Iact is being
read or listened to, in truth, the only thing that language is capable oI Iully
representing is language. Accordingly, the greatest sense oI presence in nar-
rative is not created through the abundant description oI details, and transla-
tio temporum (the use oI the present Ior the past tense), but eIIected by direct
speech in the person oI a character. This Iigure also aims to Iuse past and
present, or represent utterances Irom the past as present utterances. Because
such statements eIIectively exist simultaneously in the past and in the pre-
sent, they act out the past and stand in Ior it, as it were, in the present.
For the reasons explained above, ancient theorists had little need Ior
developing comprehensive typologies oI narrative based on the narrator`s
relationship with the characters oI the story. However, a Iew crucial distinc-
tions were made, which could serve as the basis Ior a limited system oI theo-
retical classiIication. Especially interesting to us is a type oI narrative which
is expressly said to be a separate Iunctional class Irom narratives used in
public speeches.
83
This literary narratio is subdivided into two types, one oI
which is marked by the occurrence oI alterations in speaker identity: eius
narrationis duo sunt genera. unum quod in negotiis, alterum quod in per-

Etvm. 2.14.2, in quo genere dictionis illa sunt maxime cogitanda, quis loquatur et apud
quem, de quo et ubi, et quo tempore, quid egerit, quid acturus sit, aut quid pati possit, si
haec consulta neglexerit ('in this type oI discourse one should primarily keep in mind, who
is speaking and to whom, about what and where, and at what time; what he has done, what
he is about to do, or what could happen to him, iI he should disregard the deliberations).
82
Quint. Inst. 9.2.58, imitatio morum alienorum, quae \H-G-&+" vel, ut alii malunt, +K*&.
dicitur, iam inter leniores affectus numerari potest ('the imitation oI the manners oI others,
which is called ethopoeia or, as some preIer, mimesis, can be counted among the milder aI-
Iects oI discourse).
83
Cic. de inv. 1.19.17, tertium genus [sc. narrationis] remotum a civilibus causis ('a third
type oI narration is unconnected with public issues); Rhet. Her. 1.8.12, tertium genus est
id, quod a causa civili remotum est ('a third type is that which is unconnected with public
issues).
32 1 NARRATI VE
sonis positum est ('there are two types oI this narrative: one which consists
in events, another which consists in persons).
84
The distinction may be more
theoretical than practical, since most narratives involve at least some imper-
sonation or direct speech, oratio recta, but since the Ieature is adduced as the
distinct characteristic oI the class one may at least assume that what is meant
is narrative which makes extensive use oI the device. The type is Iurther
deIined as illa narratio quae versatur in personis'that type oI narrative
which employs persons/masks.
85
According to Cicero, who so describes it,
one can recognize therein, in addition to the 'matter, res, oI the story, the
'utterances, sermones, and through them the 'minds, animi, oI the perso-
nae. The idea oI narrative in rhetorical theory, it goes without saying, is
based on the paradigm oI a single speaker, a single speaking voice.
In order to show the presence oI the same basic structure in the narrative
oI the Satvrica, we need to demonstrate how exactly Encolpius, the teller oI
the story, manages to be the only speaker throughout the work, and how he
impersonates all the other characters. In doing this, our aim is to read the
Satvrica not as modern narratologists but as ancient rhetorical theorists.
There seems to be no better method to demonstrate this than to go brieIly
through the extant text to mark the points at which Encolpius speaks as iI he
were someone else. To avoid making this an excessively long sequence
(since the point oI the demonstration is only to establish a model oI the Sa-
tvrica`s narrative structure according to ancient principles), we include only
major passages oI impersonation while many shorter utterances are leIt out.
In the Iollowing sequence the proper names may be thought oI as verbal
equivalents oI masks (the narrator is marked by caps., ENCOLPIUS, even in
minimal bridges crossing Irom one impersonation to another, but the imper-
sonated masks by quotation marks, e.g., 'Quartilla). By accident, the extant
Satvrica opens in the middle oI a passage where the narrator is impersonat-
ing his younger selI, a character in the story, as he spoke at that moment in
the past, aIter which the central identity resurIaces and so on and so Iorth:

|.| 'Encolpius ENCOLPIUS (3.1) 'Agamemnon ENCOL-
PIUS (6.117.3) 'Quartilla ENCOLPIUS (18.137.1) 'Her-
meros ENCOLPIUS (39.1) 'Trimalchio ENCOLPIUS (40.1
42.1) 'Seleucus ENCOLPIUS (43.1) 'Phileros ENCOL-
PIUS (44.1) 'Ganymedes ENCOLPIUS (45.1) 'Echion
ENCOLPIUS (47.1) 'Trimalchio ENCOLPIUS (47.750.4)


84
Rhet. Her. 1.8.13
85
Cic. Inv. 1.27.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 33
'Trimalchio ENCOLPIUS (52.455.4) 'Trimalchio EN-
COLPIUS (56.757.1) 'Hermeros ENCOLPIUS (59.161.5)
'Niceros ENCOLPIUS (63.1) 'Trimalchio ENCOLPIUS
(64.165.9) 'Habinnas ENCOLPIUS (67.271.5) 'Trimal-
chio ENCOLPIUS (72.174.13) 'Trimalchio ENCOLPIUS
(78.181.2) 'Encolpius ENCOLPIUS (82.183.7) 'Eu-
molpus ENCOLPIUS (90.1101.9) 'Eumolpus ENCOLPIUS
(102.10) 'Encolpius ENCOLPIUS (102.14) 'Giton EN-
COLPIUS (103.3106.4) 'Eumolpus ENCOLPIUS (107.7)
'Lichas ENCOLPIUS (107.12108.14) 'Tryphaena EN-
COLPIUS (109.1110.8) 'Eumolpus ENCOLPIUS (113.1
115.9) 'Encolpius ENCOLPIUS (115.20116.3) 'vilicus
quidam ENCOLPIUS (117.113) 'Eumolpus ENCOLPIUS
(124.2125.4) 'Chrysis ENCOLPIUS (126.8129.3) 'Circe
ENCOLPIUS (129.1012) 'Encolpius ENCOLPIUS (130.7
132.9) 'Encolpius ENCOLPIUS (132.1112) 'Encolpius
ENCOLPIUS (133.12) 'Encolpius ENCOLPIUS (133.4
134.10) 'Oenothea ENCOLPIUS (135.1138.5) 'Encolpius
ENCOLPIUS (139.3141.1) 'Eumolpus |.|.

The identity oI the main speaker (ENCOLPIUS in the schema above) is not
on the same Iooting as the subordinate masks that he dons. From the modern
perspective, the diIIerence lies in the Iact that the speakers oI the shorter
discourses within the main discourse also Ieature as narrated characters in
the main story, whereas the reverse is not true. The actions oI these charac-
ters are related by the narrator and speech is assigned to them, either by
oblique reIerence or by his speaking on their behalI, each in turn. From the
ancient perspective, these utterances in personis are thereIore not autono-
mous units, or in any way primary to the narrative itselI, but can be inter-
preted only with respect to the context in which they appear. We only need
to translate them into indirect speech, oratio obliqua, which does not involve
impersonation, to understand how integral they are to the narrative oI the
Satvrica as a whole.
It may seem strange to argue that Encolpius, himselI a persona, can
impersonate other personae, but such is the regular ancient understanding oI
the term. Cicero illustrates the type oI narrative that operates through imper-
sonation by quoting a dramatis persona Irom the Adelphi (6064) oI Terence
impersonating another dramatis persona oI that same play.
86
Here the use-

86
Cic. Inv. 1.27; cI. Quint. Inst. 9.2.58. Cicero`s example can be a little conIusing, because he
has taken the expositional speech oI Micio (over 80 lines long), which opens the Iirst act oI
34 1 NARRATI VE
Iulness oI the concept oI mask is obvious, iI correctly understood, Ior it em-
phasizes the change in identity, but does not seek to keep track schematically
oI all the possible identities involved in order to establish a complete theo-
retical hierarchy oI identities; nor does it insist on a distinction between real
and Iictive personae. Thus the term persona allows us to treat historical au-
thors and Iictional narrators in the same way (the term persona was com-
monly used oI people outside the theatrical or rhetorical context). As we
showed above (in our example oI Photios reading the two Greek Ass-
Stories) such a distinction would not have a basis in any internal textual evi-
dence. This simple method oI accounting Ior discourse within discourse
makes unproblematic Iurther impersonations by impersonated characters, oI
which there is plenty in the Satvrica; just as Encolpius can speak in the per-
son oI himselI as youth, he can also make his youthIul selI utter phrases in
the person oI a bombastic declaimer (1.1).
The central speaker identity oI Encolpius is the basis oI the Satvrica`s
thematic and Iormal unity. II the voices oI the subordinate personae were
truly the voices oI others, and not simply the voice oI Encolpius imitating
the characters oI his story, there wouldn`t be any organizing intelligence
behind the Satvrica. The reason we Ior example get a Cynic rejection oI
literature in the speech oI the semi-literate Echion and a Cynic rejection oI
philosophy in the epitaph oI the culturally snobbish, albeit ignorant, Trimal-
chio, is that this is what the disillusioned litterateur Encolpius has on his
mind.
87
As to Iormal unity, iI we Iollow the example oI some modern narra-
tologists and treat as primary the utterances oI the characters and the events
oI the story, we cannot, Ior instance, account Ior such phenomena as the
strange poetic utterance oI Tryphaena (108.13109.1), which at Iirst appears
to have been spoken by her in hexameters, but is then introduced by the nar-
rator in the middle oI the Iirst line with the verbum dicendi, 'exclamat.

|Tryphaena| protendit ramum oleae a tutela navigii raptum, atque in col-
loquium venire ausa
quis Iuror` exclamat pacem convertit in arma?
quid nostrae meruere manus? non Troius heros
hac in classe vehit decepti pignus Atridae,
nec Medea Iurens Iraterno sanguine pugnat.

the play, as an example oI narratio, although by deIinition a play as a whole is not narra-
tive.
87
46.7, 'nam litteris satis inquinatus est ('The boy is tainted enough as it is with litera-
ture`); 71.12, 'nec umquam philosophum audivit ('He never heard a philosopher lec-
ture`).
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 35
sed contemptus amor vires habet. ei mihi, Iata
hos inter Iluctus quis raptis evocat armis?
cui non est mors una satis? ne vincite pontum
gurgitibusque Ieris alios immittite Iluctus.`
haec ut turbato clamore mulier eIIudit, haesit paulisper acies, revo-
cataeque ad pacem manus intermisere bellum.

'Tryphaena held out an olive-branch Irom the ship`s Iigurehead, and
dared to come up and talk to us:
'What madness, she shouted, 'is converting peace into war?
What have we done to deserve this? No Trojan hero
carries away in this Ileet the spouse oI the cuckold son oI Atreus.
No Medea Iights here driven mad by her brother`s blood.
But rejected love has violent impulses. Ah! Who was it then
that took up arms and summoned the Iates on these waves?
For whom is one death not enough? Do not outdo the sea
and heap waves oI blood upon its savage Iloods.
The woman poured out these words in a loud excited voice, the battle
line hesitated Ior a while; summoned back to peace we dropped the war.

The peculiarity oI this particular passage was noted by Michael CoIIey,
88

and not long ago remarked upon extensively by Slater, who acknowledges
that Irom his modern standpoint the speech oI Tryphaena 'raises a unique
diIIiculty with regard to its status as utterance. AIter several attempts to
account Ior it within the narrative oI Encolpius as verbatim quotation, Slater
eventually considers the possibility that it may not aIter all be a quotation.
However, he immediately rejects this as unacceptable since '|i|I Encolpius is
so unreliable a narrator that we Iind him altering direct speech he is suppos-
edly reporting, how shall we Iind our way out oI the resulting solipsism?
89

This anxiety at discovering the power oI the story-teller to give verbal Iorm
to the story seems quite modern, since Ior the ancient theorist oI narrative the
principal paradigms oI rhetorical delivery and theatrical perIormance made it
less problematic to assume that the actual words oI any narrative, including
utterances in personis, could well originate entirely with the narrator. The
ancient theory is sounder since it is more generally applicable.
90
The labyrin-

88
CoIIey 1989, 189.
89
This and the Iormer citation are Irom Slater 1990, 172.
90
Consider, Ior example, what happens in translation when every single word oI the 'direct
speech oI a character is altered to another language and nevertheless retains its status as the
direct speech oI that same character.
36 1 NARRATI VE
thine solipsism, which Slater abhors, is more oIten and less IearIully termed
narrative subjectivity.
Tryphaena`s speech makes Encolpius` activity in creating the story un-
usually obtrusive Ior the reason that he changes his discourse type Irom
prose to hexameter at the very moment he begins to speak as iI he were Try-
phaena. This makes us associate the prose with his voice and the hexameter
with her voice. But then, when we read (or hear) the word, exclamat, in the
middle oI the Iirst line oI the meter, we realize that things are not so simple.
Exclamat is, oI course, not spoken in Tryphaena`s persona, but in the narra-
tor`s own voice. We are thereIore Iorced to reject the metrical Iorm oI the
language as a mark oI her voice, and recognize that the narrator has merely
switched to narrating in meter instead oI his customary prose. But what a
place to make that switch! Our reconsideration and Iinal conclusion that
Tryphaena cannot have spoken these words as such is based on the observa-
tion that the 'unit oI meter, which in hexameter is the single verse, does not
square with the 'unit oI utterance attributed to Tryphaena. In order to 're-
store her actual speech, we would have to remove the word exclamat, and
leave a silent hole in the hexameter line.
91

The present case is the only instance oI this exact type oI jarring in the
extant text oI the Satvrica, but a similar eIIect occurs indeed in all hexameter
verses where the 'unit oI utterance does not square with the 'unit oI me-
ter.
92
Although in epic the main narrative is conducted in hexameters, such
passages oI direct speech interrupted in the middle to add the declarative
statement (as is common practice in Latin prose) are Iormal absurdities that
disprove themselves as historical utterances, since the declarative statement,
when removed, there too leaves a silent hole in the verse. In the Satvrica we
are, however, made more sensitive to the problem due to the jarring transi-

91
Or add a metrically equivalent phrase, such as the words that we Iind instead oI exclamat in
the verse as reported by Isidore (Etvm. 2.21.19, quis furor, o cives, pacem convertit in
arma) who Iails to mention the poet`s name.
92
E.g., the 'quotation oI Entellus in Virgil`s narrative (Aen. 5.47376):
hic victor, superans animis tauroque superbus,
'nate dea vosque haec,` inquit, 'cognoscite, Teucri,
et mihi quae fuerint iuvenali in corpore vires,
et qua servetis revocatum a morte Dareta.`
dixit et .
('At this the victor cries, triumphant in spirit and glorying in the bull:
You goddess-born,` he said, and you Trojans, learn
what strength I had in my youthIul body,
and Irom what impending death you recall and rescue Dares.`
So he spoke .)
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 37
tion Irom prose to poetry at the very beginning oI the impersonated speech.
For this reason it is clear that the word 'exclamat has not been thought-
lessly inserted here although this seems to be the case in Virgil (Aen. 5.473
76). The eIIect oI the anomaly is simply too goodthat is iI it is noticed and
understood Ior what it isand exceedingly appropriate to Encolpius` con-
stant play with diIIerent discourse types, Ior it leads to a sudden realization
in the reader/hearer that conventional epic heroes address each other, ab-
surdly, in verses, and that no historical individuals can have spoken thus! As
we shall discuss in more detail below, this is very much in keeping with the
general tendency in the Satvrica to represent poetry comically as a sort oI
speech aberration.
1.2.2 The Identities oI Voices
The ancient theory oI discourse within discourse is best considered in the
context oI delivery and with respect to the capacity oI the living voice to
alter its identiIying Ieatures. The paradigm oI the human voice is better
suited Ior the problematics oI narrative and story telling than is the static and
silent Iormalist vision oI the printed page, which knows no other method oI
distinguishing voices than typographical signs. For marking the change in
speaker-identity, the natural modulations oI the human voice ideally make
unnecessary even the paraphernalia oI the theater. As Quintilian explains,
voices are no less constitutive oI identities than are Iaces:

ut Iacies, quanquam ex paucissimis constat, inIinitam habet diIIerentiam,
ita vox, etsi paucas, quae nominari possint, continent species, propria
cuique est, et non haec minus auribus quam oculis illa dinoscitur. (Inst.
11.3.18)

As the Iace, although consisting oI very Iew Ieatures, is Iound in inIinite
variety, so the voice, although there are only a Iew kinds to which we
can give a name, is proper to each; in Iact the voice is no less recogniz-
able by the ear than the Iace is by the eye.

Ut facies, ita vox is a succinct and appropriate Iormulation oI the simple
principle which makes possible narrative impersonation or speech in the
voice oI another. In practice this is accomplished by using the voice in a
particular manner. The Rhetorica ad Herennium contains the Iollowing ad-
38 1 NARRATI VE
vice to the would-be orator as to how he should deliver the utterances oI
others concerning the case being expounded:

si qua inciderint in narrationem dicta, rogata, responsa, si quae admira-
tiones de quibus nos narrabimus, diligenter animum advertemus ut om-
nium personarum sensus atque animos voce exprimamus. (Rhet. Her.
3.14.24)

II it so happens that in the narration there occur statements, questions, or
answers, or some comments about the case we are relating, we shall give
careIul attention to express through the voice the Ieelings and thoughts
oI all the personae.

The semantic logic oI identity makes it necessary Ior the old identity oI the
voice to be momentarily suppressed Ior the new one to come out, and so
there is no keeping track oI the layers oI identities, although the actual mo-
ment oI change is highly signiIicant and allows us to partly retain the mem-
ory oI the suppressed identity. This underlying identity, however, does not
remain Iunctional on the surIace level, but is kept in suspension until the
speaker is recognized again Ior what he was beIore.
1.2.3 The Apotheosis oI Encolpius
This brings us to the methodological quandary oI the currently popular theat-
rical interpretation oI the Satvrica, which I presented earlier as one oI several
readings enabled by the use oI the Fragmenta as 'comparison text.
93
The
essential quality oI classical narrative is that it is conducted through one
persona (the underlying speaker identity which reappears aIter occasional
subordinate impersonations) and thus the whole work Ialls into a single un-
broken context, whereas the dramatic Iorm makes use oI many basic drama-
tis personae and thereIore many textual partes, each oI which Iorms a dis-
tinct context within the play (although in perIormance the action connects
them). As beIore, it is the recognition oI the speaker identity which places
the utterance within its proper context and thus determines how it should be
understood. A written dramatic text is practically unintelligible iI the proper

93
Panayotakis 1994a, xx, 'The surviving Satvrica |.| should be regarded as a sophisticated
synthesis oI many diIIerent literary genres, including oratory, historiography, epic, elegy,
satire, Greek romance, and drama |.| the element oI theatricality is a dominant Ieature in
the Satvrica.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 39
names oI the characters (not pronounced on stage since the speakers are
identiIiable by their visible masks) are not included in the margin. On the
page, thereIore, the proper names Iunction as masks do on stage, i.e., as con-
textual markers. The Satvrica can be thought oI as theatrical only in so Iar as
we recognize that in staging the work we would need only one actor Ior the
basic voice, or persona, oI Encolpius. The term 'actor, however, may be
misleading, since the Satvrica is no drama and lacks all action in perIorm-
ance apart Irom the gesticulation which would accompany a lively delivery.
It is as iI the promoters oI the theatrical interpretation oI the Satvrica
wanted to translate Encolpius` narrative into so many modern-style stage
directions (ancient stage directions were carried in the text: not 'enter X,
but 'here comes X). As sometimes seems to happen in the writings oI mod-
ern narratologists, the characters oI the story are treated as primary and the
narrator is sublimated into a 'voice Irom heaven.
94
Costas Panayotakis
makes much oI 'the Iact that the Satvrica can be read theatrically as iI it
were the narrative equivalent oI a Iarcical staged piece with the dramatic
structure oI a play produced beIore an audience.
95
It is quite possible that
there is some 'theatrical quality to the work, but there is certainly no 'dra-
matic structure in the narrative oI the Satvrica. In order to enhance the the-
atrical qualities oI the Satvrica Panayotakis thereIore simply rewrites the
Quartilla episode and the quarrel oI Encolpius and Ascyltos (79.11-80.6) as
dialogue and stage directions. As we intend to show, the Satvrica is written
to be recited in a lively manner. As such it is perIormance literature, as are
plays. I would like to stress, however, the rather obvious Iact that not all
perIormance literature is drama. Indeed, a paradoxical requirement oI the
current theatrical reading, at least in its most exaggerated Iorm, is to virtually
obliterate the only persona oI the work that could, without rewriting the text,
be presented on stage, Encolpius the narrator.

94
The interpretation oI Scripture as the word oI God must necessarily dispense with the rhe-
torical emphasis on the speaking person, time and circumstances oI utterance (e.g. since the
persona oI God is omnipresent and eternal). In the Iamous sixteenth-century quarrel oI Lu-
ther and Erasmus over the Ireedom oI the will, the biblical exegete Luther criticizes the
more rhetorically minded Erasmus Ior applying the criteria oI time, place and person to the
interpretations oI Scripture. According to Luther 'God did not give us a word which gives a
hoot Ior places, persons and times (Nec dedit nobis Deus uerbum, quod locorum, persona-
rum, temporum delectum habeatLuther, De servo arbitrio |1525|, 628 |W|).
95
Panayotakis 1994a, 320; the Iormulation oI the phrase 'narrative equivalent oI a stage
Iarce belongs to Walsh 1974, 186, Panayotakis` supervisor at the University oI Glasgow.
This seems to be the main position oI Slater 1990, 14, as well: '|R|eading is Iar Irom a pas-
sive process. It is just as participatory, though in diIIerent ways, as watching a perIormance
in the theatre.
40 1 NARRATI VE
Certain accidents oI preservation have contributed to making apparently
viable the dramatic interpretation. It so happens that the extant text oI the
Satvrica opens not in a passage oI the narrative, but in the middle oI an im-
personated speech, in the voice oI the character oI young Encolpius. Like-
wise, by accident, the text breaks oII in the middle oI a speech oI another
character, the poet Eumolpus. But perhaps even more conIusing than the
mutilated condition oI the beginning and the end oI the manuscript tradition
is the Iact that in several places, where the text is very Iragmentary, modern
editors have added in the margin the names oI characters who are thought to
be the speakers oI such isolated Iragments (in some places the name oI the
supposed addressee and subject oI the speech is also included).
96
These addi-
tions derive originally Irom the scribes oI the Longer Fragments; they are
obvious attempts at reconstructing the internal context and oI course were
not Iound in the original text. Bcheler leIt them out oI his edition, while
later editors such as Ernout and Mller have included them. By virtue oI
these marginal glosses, parts oI the Iragmentary modern text have taken on
the appearance oI a dramatic text.
Another contributing Iactor is Iaulty method. By reason oI a certain un-
happy use oI terminology, the protagonist Encolpius is commonly reIerred to
as the 'narrator, although he is, Irom the point oI view oI narrative Iorm,
not the teller oI the story, but a character represented in it.
97
It is, oI course,
Encolpius the narrator (who exists in the present oI the narrating act) who
represents this younger selI (existing in the narrated past), sometimes by
providing inIormation about him and sometimes by speaking on his behalI.
In Iact, the narrator`s representation oI his younger selI is eIIected by more
or less the same technique as is used Ior his treatment oI other characters.
Although the narrator is the same individual as the young man Encolpius,
both the time and the circumstances oI speech are diIIerent, and these com-
prise deIining qualities oI speaker personae.
II the Satvrica is not a dramatic text, and not even the 'narrative equiva-
lent (to me an incomprehensible phrase) oI such a text, what then can we
make oI the handIul oI places where the narrated events oI the story are ex-
plicitly described as mime-like and theatrical? To begin with we are cer-
tainly not required to interpret reIerences oI this sort as Encolpius` attempt to
settle the typological problem raised by his narrative in Iavor oI theatrical
mime.
98
II we consider closely the context in which he uses terms such as

96
85.1, 94.1, 96.7, 99.1, 104.1, 107.1, 113.11, 126.1, 128.1, 128.3, 128.7, 129.1, 132.1, 134.1,
134.8.
97
On this 'Iallacy in modern narratology, see Genette 1988, 223.
98
Walsh`s 1974, 189, assumption that '|t|he central action oI the Satvricon |.| is pointed up
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 41
mimicus or scaena, we note that it is to underline a certain strange and alien
quality oI narrated situations and acts. Quartilla`s laughter is mimicus be-
cause it expresses her sinister joy at having the boys absolutely in her power
(19.1). Trimalchio`s dining room resembles the perIormance oI a mime be-
cause, as Encolpius notes with amazement, both the slaves and the host him-
selI are continuously singing cantica Irom literary mimes (31.6, 35.6, 55.5
6). The reason is that the Ireedman host, because oI his lack oI aristocratic
education, when entertaining guests in his dining room looks Ior his models
in public entertainment such as the mime, and is quite ignorant oI the proper
etiquette oI the Roman triclinium.
99
It is thereIore the mime which in the
narrator`s mind seems misplaced. Furthermore, the Iriendship oI Ascyltos
and Giton is compared to a Iarce, because Encolpius sees it as having been
deceitIul in the light oI their later betrayal (80.9). The mock-suicide oI Giton
is associated with the mime because it is blatant Scheintod (94.15, mimicam
mortem); the slave-disguise on the ship is a mime because it is designed as
deception (106.1, mimicis artibus); and Iinally the conIidence trick oI Eu-
molpus is given this name Ior the reason that Eumolpus and company take
on roles to deceive Ior Iinancial gain (117.4). Though certain incidents that
supposedly happened to the narrator in his past strike him in hindsight, as he
delivers the narrative, as iI he had been trapped in a low and tasteless Iarce,
this certainly does not justiIy the claim that mime is a 'source Ior the Sa-
tvrica, or that the work has been 'inIluenced or shaped by the genre oI the-
atrical Iarces.
100

Such comparisons may come naturally to Encolpius, because in the Iirst
century the mime was by Iar the most popular Iorm oI theater.
101
It is ques-
tionable whether it was even possible to depict common people without re-
sorting to re-presenting them according to the convention oI contemporary
Iarce. Certain similarities, thereIore, between the Satvrica and the contempo-
rary theater may well be commonplaces oI the culture. Theater in general

by so many reIerences to the mime that the novel proclaims itselI the narrative equivalent oI
a stage Iarce |my underlining| is superIicial.
99
It is true that dinner-theater and other entertainments seem to have been Irequent in the
houses oI the wealthy (Plin. Ep. 9.17.1; Quint. Inst. 1.2.8), but it is also true that moralists
(Plut. Mor. 711A II.) wanted such activities to be reserved Ior public entertainment on the
stage, and Encolpius` words show that he sides here with that austere Iaction (Sat. 31.7).
The blurring oI public and private in dinner theater and theater-dinners is discussed by
Jones 1991, 18598.
100
Early protesters against Rosenblth`s original 1909 thesis (that theatrical Iarce was an
important 'source Ior Petronius when he composed the Satvrica) include Mring 1915 and
Paratore 1933, 1:99104.
101
For a survey oI the documentary evidence Ior the ancient mime, see Maxwell 1993.
42 1 NARRATI VE
and the mime in particular, oI course, provide potent metaphors in the Sa-
tvrica, and there is no question that the rhetorical and philosophical vocabu-
lary relied heavily on the institutional practices oI the theater. However,
where mime is explicitly mentioned in the Satvrica, this seems to include a
rejection oI this Iorm as 'diIIerent and no less unappealing to our narrator
than, let us say, declamation, or verbose epic. As I have argued above, the
theatrical reading oI the work seems to be inIluenced by a vision oI the text
as Iragments broken out by genre. To select one oI these Iragments as 'the
genre oI the Satvrica is unlikely to settle the question.
By concentrating on qualities that he rejects, Encolpius increases their
importance as a 'generic other (and thereIore as an indirect means to arrive
at his 'generic selI). It is more IruitIul to see such allusions in the Satvrica
as attempts to negotiate a 'diIIerence and thus to establish the proper quali-
ties oI this narrative by positioning it with respect to other somewhat related
Iorms. In this sense the work at hand conIorms to the regular practice oI
ancient literature, since in the Greco-Roman generic system deIinitions oI
individual works are more oIten negative than positive, and arrived at by
reIerence to what the work is not, rather than what it is (not 'this is elegy
but 'this is not epic or tragedy). We should resist the tendency to view an
ancient genre as an absolute category. II this were the case, the constant re-
negotiations within the system would not be necessary.
102
In Iact, the laws oI
the genre are oIten violated in order to be re-established (a cliche oI comic
narratives runs something like this: 'dear reader, know that we are ascending
to the tragic buskin). The diIIerence between drama and narrative, however,
is not a question oI genre, and thereIore does not depend on generic negotia-
tion. This distinction is more basic, and regards the number and interrela-
tions oI the masks. Neither does the discourse-type, oI which we Iind a great
variety in the Satvrica, aIIect this distinction. Although generic categories
are usually open to renegotiation, prosimetry is not properly a genre, in my
view, any more than prose or poetry Ior that matter is a genre.
103

1.2.4 Remembering is Telling
We now come to the mode oI presentation implied by the narrative oI En-
colpius. At three points the narrator explicitly states that he is recalling Irom

102
On the messiness oI ancient genres, see Hinds 1998.
103
This rule did not, however, seem to apply to the discourse type oI hexameter which was
pretty much equated with the genre oI 'epic in the tradition oI ancient rhetorical criticism,
see Koster 1970. For the opposite view oI prosimetry as a genre, see Relihan 1993.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 43
memory what he is relating. In two oI the three cases it is an apparent at-
tempt to remember the exact wording oI a written text, which prompts the
reIerence to memory. The Iirst instance occurs in the context oI a detailed
description oI certain tablets in Trimalchio`s house, one oI which has an
inscription which Encolpius attempts to cite verbatim with the Iollowing
reservation: si bene memini'iI I remember correctly (30.3). Again he
relates an incident when a cup was passed around with inscribed pittacia
'strips oI cloth, which were recited while countless presents with punning
associations were oIIered. AIter recounting eight oI these jokes, he says:
sexcenta huiusmodi fuerunt, quae iam exciderunt memoriae meae'there
were countless |labels| oI this kind, but they have now slipped my memory
(56.10). We note that it is implied by these reIerences to memory that
enough time has passed Irom the events related until the narrating instance
itselI (the word iam clearly marks the present oI narration with respect to the
past oI the event) to make accurate recollection problematic. Measuring this
time span may be impossible, but this is less important than understanding
the basic premise oI the Satvrica`s narrative, viz., that it is presented as En-
colpius` recollections oI events experienced by him in the past. This is oI
course the classical constitution oI autobiographical narratives, also called
'recollections, or memoirs, to highlight the role played by memory in this
narrative Iorm.
104

That reIerences to the limitations oI the narrator`s memory should be
prompted especially by the attempt to cite written documents verbatim im-
plies a certain dichotomy with respect to how written and non-written words
are treated when reported. Only the Iormer, when cited, are subject to the
veriIication which the medium oI writing enables, whereas the latter are
regarded as unveriIiable and Ileeting, and their citation, thereIore, necessar-
ily Iree oI similar constraints. In this respect the Satvrica does not diIIer
much Irom ancient historiographical texts, which represent written docu-
ments diIIerently Irom the occasional speech attributed to a historical per-
sonage, which could be Ireely invented iI due respect was paid to what was
known about the individual in question and the situation in which the state-
ment was made.
105
It was with good reason that rhetorical theorists consid-

104
The important role oI memory in the ancient Iormulation oI the concept is emphasized in
such terms as ^G-3X"#", JG-3K-38="#", monumenta, commentarius. Once 'memo-
ries have been committed to writing the written text can in turn be viewed as an aid to
reminding. Plato calls the art oI writing a drug oI reminding: 'You have discovered a drug
not oI memory, but oI reminding-e 3XK. J,,` ^G-3X*84. )(%"'-3 (Phaedr.
275A).
105
E.g., Sallust limits himselI to reIerring to Cicero`s written and published speech against
44 1 NARRATI VE
ered it a useIul preparation Ior historiography to declaim in the persons oI
historical Iigures.
106

Now, iI we press Iurther this dichotomy oI the written and the spoken,
we oI course notice that the Satvrica is itselI a written text, although it is not
entirely clear whether Encolpius is supposed to be aware oI this. There
would have been many ways Ior him to acknowledge as much. A simple
mention oI the act oI writing or reading, understood to apply to his narrative,
would have suIIiced, or he might have reIerred explicitly to his recollections
as a book. OI the great historians, Thucydides is at pains to clariIy that his
GF,8-. between the Peloponnesians and Athenians is a written document,
'#f( #8 0. "M8R O,,-3 > J/;3&*" 0. #Q G"%"N%f" J'-=8&3'and a
property Ior the Iuture rather than a contest Ior the occasional listener
(2.22.4). Herodotus, however, does not present his g*#-%+K. JGFB8W&. as writ-
ten any more than spoken. OI the ancient novelists Chariton begins by deIin-
ing himselI as the scribe, ^G-/%")8=., oI the rhetor Athenagoros, and in
book 8 predicts that the ending oI his written composition, #-T#- *=//%""
(8.1.4), will be the most pleasant to his readers, #-9. J3"/&/3;*'-$*&3;
Xenophon oI Ephesus adds at the end that a written version oI his tale oI
Anthia and Habrocomes was dedicated in the Iamous temple oI Artemis in
Ephesus (5.15.2); Achilles Tatius is silent about whether his love story, told
in the persona oI Clitophon, is to be thought oI as a written document or
merely as 'spoken; Longus is deIinitely engaged in the act oI /%()8&3
(praef. 4); Heliodorus` deIinition oI his story at the end as *=3#"/" makes
it into a book; and Apuleius` Latin version oI the Ass-Story is a papvrus to
be read (1.1), and its narrator is destined to become several 'books (2.12).
UnIortunately, we do not have the whole text oI the Satvrica and so the
possibility cannot be excluded entirely that such a reIerence did in Iact oc-
cur, Ior example, at the beginning. However, as Iar as it goes, our text oI the
Satvrica does not present its narrator as betraying any knowledge oI the Iact
that his words constitute a written text. Tentatively, then, we can say that
Encolpius` narrative is not just presented as a recollection Irom memory, but
also seeks to hide its own textuality, leaving the impression oI a living voice
telling the story.
The idea oI the 'orality oI Encolpius` narrative is strengthened by his
third reIerence to memory, prompted by the recollection oI certain items oI
Iood, quarum etiam recordatio me, si qua est dicenti fides, offendit'the
mere recollection oI which, take this speaker`s word Ior it, disgusts me

Catilina (orationem . quam postea scriptam edidit) without even relating its contents (Cat.
31.6), but Ireely recreates unpublished speeches.
106
Quint. Inst. 3.8.49.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 45
(65.1). What is being alluded to here is not something Encolpius claims he
said or heard in the past oI the story; this is Encolpius presenting himselI in
the act oI 'speaking the Satvrica. Later in the story he again reIers to his act
oI speaking the text when trying to describe the beauty oI Circe, quidquid
dixero minus erit'whatever I say will Iall short (126.14). These phrases
go beyond a passive omission oI a reIerence to the Iact that the narrative
exists in the Iorm oI a written document; they actively attempt to create the
impression that the narrator is 'speaking his story as the memories Ilood his
mind. SigniIicantly, the Iormula si qua est fides (literally: 'iI by any chance
there is Iaith |in the speaker|) occurs predominantly in the emphatic plead-
ing oI declamatory speeches, a genre oI literature that must be classiIied as
perIormative and 'spoken.
107
Encolpius` particular use oI the declamatory
Iormula even includes Ior emphasis the rare additional Ieature oI the present
participle, dicens, reIerring to himselI. There is no weightier testimony as to
the manner in which he tells his story.
By the most natural reasoning, 'remembering equals 'telling, as is
made plain in the phrase vivorum meminerimus, 'let`s remember/talk about
the living, which occurs twice in the extant Satvrica (43.1, 75.7). The
memory oI past events necessarily triggers emotional responses in the pre-
sent as the story is told.
108
The recollecting done by Encolpius is not just an
activity oI the intellect, but involves a vivid taste memory experienced as

107
Sen. Con. 1.1.18, 7.1.7, 7.5.1, 7.6.9, 9.4.5, 9.6.19; excerpta 5.1.1, 7.5.1, 8.3.1; |Quint.| Decl.
1.5, 3.3, 4.2, 9.10, 12.8, 16.4 (si qua dicentis fides est). Apul. Apol. 43. Virgil and Ovid use
the Iormula (sometimes modiIying it) especially in the character speeches oI epic (A. 2.142,
3.433; Met. 9.55, 371, 15.361) and in elegy when the poet pleads with his puella (e.g. Am.
1.3.16); such rhetorical Iormulae in Ovid`s language are usually explained as inIluence
Irom declamation.
108
10.4, rurusus in memoriam revocatus iniuriae ('I was again reminded oI his insult); 54.3,
pessime mihi erat, ne his precibus periculum aliquid catastropha quaeritur. nec enim adhuc
exciderat cocus ille qui oblitus fuerat porcum exinterare ('I was very much aIraid that this
petition was leading up to some comic trick. The cook who had Iorgotten to gut the pig had
not yet Iaded Irom my recollection); 81.2, redeunte in animum solitudine atque contemptu
('as my solitude and rejection again came to mind); 113.3, non Lichas risit, sed iratum
commovens caput . non dubie redierat in animum Hedvle expilatumque libidinosa migra-
tione navigium ('Lichas didn`t laugh but shook his head angrily . no doubt it was Hedyle
who came to mind and how his ship had been pillaged on her libidinous elopement);
106.2, Lichas memor adhuc uxoris corruptae iniuriarumque, quas in Herculis porticu ac-
ceperat, turbato vehementius vultu proclamat ('Lichas, still remembering the seduction oI
his wiIe and the insults he took in the Portico oI Hercules, cried out with an even more vio-
lent look on his Iace); 131.10, ut me vidit paululum erubuit, hesternae sicilicet iniuriae
memor ('So when she saw me, she blushed a little, no doubt remembering yesterday`s in-
sult).
46 1 NARRATI VE
nausea in the present. The strong disgust oI the previous passage is Iurther
underscored in phrases such as pudet referre quae secuntur'One is
ashamed to tell what Iollows (70.8), where the present tense oI the verb,
secuntur (Irequent alternative spelling oI sequuntur, see OLD s.v.), shows
that what Iollows is the narrative oI the subsequent events oI the story, not
just the events themselves. It is interesting to note that the English translators
Heseltine 1913, Sullivan 1986, and Walsh 1997 all change the tense oI quae
secuntur and translate 'what happened or 'Iollowed. But it is the telling in
the present that is shameIul. As the demand Ior literal accuracy prompted the
Iirst two reIerences to memory, the necessity to induce belieI in the outra-
geous and incredible elicits an appeal to trust, which ultimately depends on
the authority and character oI the speaker. At a subordinate level, Encolpius
also impersonates Ascyltos as uttering the above mentioned declamatory
Iormula oI appeal, while questioning his own and Encolpius` credibility in
the urbs Graeca where they have but newly arrived and are strangers to the
locals: 'Who in this place knows us, or who will take our words Ior any-
thing (14.1, quis . hoc loco nos novit, aut quis habebit dicentibus fi-
dem?). He also uses the same Iigure, when impersonating Trimalchio as he
tries to induce belieI in the superstitious werewolI-story oI Niceros: 'I most
certainly believe you, take my word Ior it that my hair stood on end. I know
Ior a Iact that Niceros doesn`t tell triIling stories (63.1, salvo . tuo ser-
mone . si qua fides est, ut mihi pili inhorruerunt, quia scio Niceronem nihil
nugarum narrare).
Once more he uses the Iigure, now again at the primary level, when he
relates his own reaction that night aIter Ascyltos had abducted Giton Irom
his bed while he was Iast asleep in drunken slumber:

itaque ego ut experrectus pertrectaui gaudio despoliatum torum, si qua
est amantibus Iides, ego dubitaui, an utrumque traicerem gladio som-
numque morti iungerem. tutius dein secutus consilium Gitona quidem
verberibus excitavi, Ascylton autem truci intuens vultu. (Sat. 79.10)

So as I woke up and ran my hands over the bed robbed oI its joy, take
my word Ior it, that`s how we lovers are, I wondered whether to run the
pair oI them through with my sword and prolong their sleep in death. But
Iollowing a saIer plan, I woke Giton with blows and glared at Ascyltos
with a savage look on my Iace.

The idea oI killing Giton, his love, and Ascyltos, his Iriend and Iormer lover,
is oIIered by Encolpius as a measure oI his intense suIIering. At the same
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 47
time this outrageous event oI his story is accounted Ior by reminding who-
ever is listening that this is a tale told by a lover, one oI many such tales
(hence the plural in amantibus), and thereIore normal and credible oI its
kind. It is not just that Encolpius was a lover in the narrated past when he
supposedly experienced this unhappy incident, he still presents himselI as a
lover to his audience as he relates the story oI that incident, Ior he uses the
present tense: si qua est amantibus fides.

The main narrative oI the Satvrica is Iree oI any immediate context (the title
and name oI the author in themselves do not provide a context) and so it is
not introduced by an external speaker who then impersonates the narrator.
This is the case, on the other hand, with Clitophon`s erotic recollections in
the work oI Achilles Tatius. Here we get a Iull description oI the young man
beIore he starts telling his story. In the Satvrica, however, only the subordi-
nate speeches are introduced by Encolpius beIore he undertakes the change
in speaker identity. Lacking an ulterior narrator who tells us how Encolpius
tells his story, we must try to Iigure this out Irom the story itselI. This is not
as complicated as it might seem at Iirst and there is, indeed, a simple method
Ior catching him in the act, as it were. II we careIully observe how he de-
scribes the manner oI speech oI the subordinate personae, beIore and aIter he
impersonates them, we may capture at least the style oI these parts oI his
narrative. By working our way Irom there, we can get an adequate picture oI
his over-all manner oI telling the story.
As it turns out, Encolpius regularly brings the subordinate personae into
his impersonation clapping their hands (11.2, 18.719.3, 24.2, 34.7,
137.1),
109
shouting (9.78, 30.5, 64.12, 97.1, 106.2), laughing or smiling
(11.2, 18.719.3, 57.1, 64.12, 65.5, 69.2, 78.2, 126.89), crying (81.2), with
some expression on their Iaces (30.10, 48.1, 49.8, 53.8, 79.11, 90.5, 91.2,
105.1, 106.3, 115.4, 128.2), and gesticulating in various ways (7.2, 8.1, 11.3,
13. 2, 17.3, 18.4, 24.7, 33.1, 39.2, 47.1, 49.7, 64.8, 67.5, 69.5, 74.2, 78.3,
78.5, 80.1, 91.8, 92.3, 99.5, 101.1, 101.2, 105.9, 113.2, 114.5, 114.8, 114.10,
115.12, 127.1, 131.10, 132.13, 134.10, 139.4), or otherwise indicating their
mood beIore they begin to speak (8.2, 9.7, 24.1, 52.8, 71.1, 98.2, 100.4,
100.6, 104.2, 106.1, 131.7). Finally, there are descriptions oI the tone oI
voice in which statements are uttered, oI the type: ingerebat nihilo minus
Trimalchio lentissima voce'nevertheless Trimalchio kept repeating in the
most drawn-out voice possible.
110
On the whole, in Iact, the Satvrica gives

109
Quartilla also wrings her hands till the joints crack (17.3).
110
36.7. See also 126.8, vox clara 47.11; vox canora 68.4; rabiosa barbaraque voce 96.5; vox
eiusmodi congemuit 100.3.
48 1 NARRATI VE
the impression oI being an extremely 'clamorous text, judging Irom its
uniquely rich vocabulary in this department.
111

Based on this internal evidence it is clear that Encolpius` style oI speak-
ing is not at all restrained. Every time he gives a description oI the expres-
sion on the Iace oI a persona, or some simple gesture like clapping the hands,
or the tone oI voice in which he or she utters the phrases, that description
also becomes descriptive oI how his own impersonation is conducted, and is
thus an important direction as to how the narrative itselI should be delivered
in perIormance. When the reciter or perIormer comes upon a phrase oI the
type, 'and he shouted, Iollowed by a character statement, the most natural
manner oI delivering that statement is by mimicking a shouting tone oI
voice. The same is true oI gesticulation. These built-in stage directions, in
Iact, are similar to those used in ancient drama in general, where characters
are regularly made to describe the manner oI imminent action on stage in
their actual dialogues, as opposed to such directions being external to the
text and added in the margin in the manner oI modern theatrical texts.
112

Another quality oI the Satvrica, which is owed to its presentational
mode, is the curiously disruptive style oI its narrative discourse. Although
perhaps in danger oI being conIused with it, this characteristic oI the work is
very diIIerent Irom textual Iragmentation. By virtue oI the Iorm, Encolpius
the narrator has only one voice and can impersonate no more than one
speaker at a time. But it is as iI the many invoked identities oI the Satvrica
were competing to possess his Iaculty oI speech. In the opening passages oI
the extant text, Agamemnon interrupts Encolpius declaiming in the portico,
just as the rhetor in a lost passage was interrupted 'sweating out a suasoria

111
clamare 1.1, 9.7, 14.5, 40.1, 46.8, 52.6, 58.5, 60.8, 68.3, 95.5, 107.10, 108.5, 108.7, 115.3,
127.9 v. 6, 137.4, 138.3; clamitare 92.7; conclamare 50.1, 54.1, 102.9; declamare 3.1, 48.4,
137.9 v. 5; exclamare 21.1, 22.4, 30.5, 49.5, 74.7 and 11, 83.4, 94.13, 95.3, 103.1, 105.5,
108.14, 135.2; proclamare 9.5, 14.6, 43.1, 54.2, 64.12, 67.13, 68.4, 80.4, 81.2, 97.1, 104.5,
105.7, 106.2, 115.9; clamor 14.7, 40.2, 59.6, 68.5, 73.4, 89.1 v. 20, 92.7, 94.9, 108.8, 109.1,
114.8, 136.13; declamatio 2.3, 6.1, 48.4, 107.12, 133.1; declamator 1.1.; vocare 24.6, 36.8,
47.11, 49.4, 67.3, 69.5, 117.10, 132.2, 133.1; vox 24.4, 34.1, 35.6, 36.7, 41.6, 44.9, 47.11,
59.3, 68.4, 70.7, 74.1, 91.5, 92.7, 96.5, 100.3, 100.5, 105.6, 105.7, 105.9, 108.4, 108.5,
113.8, 118.4, 120 v. 78, 121 v. 102, 122 v. 155, 122 v. 180, 124 v. 183, 127.1, 127.5, 134.6,
140.9; vociferatio 14.5; vociferare 114.6; strepitus 1.2; sonare 5 v. 18; exsonare 73.4,
109.6; personare 122 v. 177; sonus 2.2, 5 v. 16, 41.7, 68.5, 100.5, 105.6, 115.2, 127.5; soni-
tus 123 v. 225; canor 5 v. 19; cantare 28.5, 31.4, 31.5, 34.1, 53.13, 62.4, 64.3, 70.7; can-
turire 64.2; cantus 109.6, 120 v. 72, 131.8 v. 8; canticum 31.6, 35.6, 73.3; exsibilare 64.5.
112
E.g., Davos` words in Terence`s Phormio: 'Ah, is that Geta (coming there)? . All right:
here he is (Ter. Ph. 50I., DAJOS. sed videon Getam? . praestost.), as opposed to the
modern interlinear, 'ENTER Geta FROM Demiphoss HOUSE, which has been added in
the Loeb English translation.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 49
in the school (3.1; 6.1); and the rhetorician is again interrupted by the ingens
scholasticorum turba'the enormous mob oI scholastics as he extempo-
rizes a type oI Lucilian satire in the portico (6.12). Frequently, in order to
speak, a persona must interrupt another`s speech, who accordingly is not
allowed to Iinish. Trimalchio interrupts Hermeros (39.1); Seleucus Dama
(42.1); Phileros Seleucus (43.1); Echion Ganymedesin the middle oI a
sentence(45.1); Trimalchio his accountant (53.68); a slave reciting the
pittacia interrupts his master (56.7); Scintilla Habinnas (69.1); Encolpius
Eumolpus (93.3); Encolpius, Eumolpus and Giton each other (102.1103.1);
Hesus Lichas (104.5); Lichas Eumolpus (107.7); Eumolpus Lichas (107.12);
and Iinally Chrysis interrupts Encolpius (139.4). Occasionally, the events
related in the narrative either make possible some character`s speech (14.8),
or more oIten the narrated action cuts short someone who is speaking.
113

Young Encolpius is at times leIt speechless Irom shame aIter a verbal on-
slaught Irom another character.
114
Certain utterances also interrupt the events
oI the story.
115
It also happens that events are supposed to take place in the
background while a character is speaking.
116
And there are other speech-
related interruptions.
117
All these have to do not just with how things suppos-
edly happened in the past, but rather how Encolpius gives narrative Iorm to
his memories.
The way in which he organizes his narrative can also have signiIicance
Ior the over-all impact he wishes to create. At the dinner party oI Trimalchio,
the host tyrannizes the Iaculty oI speech and must, quite literally, be narrated
to the pot in order to enable the Iamous speeches oI the Ireedmen to take
place (41.9). Upon returning, he again becomes the dominant persona (47.1).
A most interesting paradox in the narrative oI the dinner-party is the silence
imposed upon the scholastici, who are Iormally trained speakers (46.1, 58.8),
while untrained speakers are allowed to ramble on Ireely (61.35). What
causes the silence oI the scholastics is not that they appreciate the discourse
oI the Ireedmen, but the Iact that they are Trimalchio`s parasites and have
only been invited to dinner to Ilatter him and allow him to pose as one who
appreciates the liberal arts and associates with educated men.
118
When, Ii-

113
52.4; 49.1, 54.1, 74.1, 90.1, 93.4, 99.5, 114.6, 110.1.
114
E.g. 10.3, Irom Ascyltos; 108.1, Irom Lichas.
115
26.8, 53.1, 97.1, 100.35, 106.1, 105.7.
116
7.5, 34.7, 41.6, 97.1, 124.2.
117
115.4, 108.3.
118
48.7, haec aliaque cum effussissimis prosequeremur laudationibus ('We Iollowed this and
other attempts oI his with the most proIuse shouts oI admiration); and 52.7, excepimus ur-
banitatem iocantis [sc. Trimalchionis], at ante omnes Agamemnon qui sciebat quibus meri-
tis revocaretur ad cenam ('We praised the urbanity oI Trimalchio`s joke, but none more
50 1 NARRATI VE
nally, the long suppressed laughter oI Ascyltos and Giton breaks out, it is
met with the aggressive and solecistic 'eloquence oI Hermeros (57.1
58.14).
119
Ascyltos` attempt to respond is quickly cut short by the delighted
host himselI (59.1). Even when invited to relate the day`s declamation,
Agamemnon is interrupted by the Ireedman`s need to drop an 'urbane re-
mark (48.5). Throughout this part oI the narrative we note the heavy irony oI
the narrator,
120
and the Ireedmen`s unappreciative attitude towards higher
learning,
121
which seeks to substitute liberal letters with litterae in
domusionem'letters Ior domestic uses (46.7, 48.4), which is the only
Iorm oI literacy they appreciate.
122

1.2.5 Scripted Memories
From the immediacy oI the Satvrica`s presentation stems another unique
Ieature oI the work, namely the deliberate attempt oI the narrator to treat all
literature as oral literature, or as plain speech, not just his own narrative, but
declamatory speeches, poetry oI various types, as well as the selI-contained
subordinate narratives. True, the Satvrica as a whole is a narrative and so
even the wall paintings in Trimalchio`s house (29) can only be seen through
the verbal commentary they provoke. However, the mode oI presentation
goes Iurther here than is necessary in a written document since, as a rule,
what is explicitly said to be written literature is not quoted directly Irom the
written document (and displayed in writing on a page oI the Satvrica), but
only enters the narrative or is alluded to indirectly, when someone actually

than Agamemnon who knew how to earn another dinner invitation).
119
CI. 32.1, Trimalchio . expressit imprudentibus risum ('the sight oI Trimalchio . pro-
voked laughter Irom those oI us who were imprudent).
120
E.g., 39.1, tam dulces fabulas ('such sweet tales) oI the vulgar gossip oI Hermeros; 59.1,
eloquentia ('eloquence) about the same Ireedman`s barbarous speech; 60.1, tam elegantes
strophas ('such sophisticated turns) about the histrionics oI a meat carver; 61.5, haec ubi
dicta dedit, talem fabulam exorsus est ('These were the words he uttered; then he embarked
upon this tale), the Iormer phrase being an epic cliche (cI. Verg. A. 2.790 and Austin 1980,
ad loc.) here used to introduce Niceros` badly told werewolI-story; 70.7, ingeniosus cocus
('genius cook); 74.5, doctissimo coco ('most learned cook).
121
71.12, 'nec umquam philosophum audivit ('He never heard a philosopher lecture`).
122
46.8, 'litterae thesaurum est ('letters are a treasure`), which reIers not to liberal letters
but to the practical use oI writing; cI. 46.7, 'emi ergo nunc puero aliquot libra rubricata,
quia volo illum ad domusionem aliquid de iure gustare. habet haec res panem. nam litteris
satis inquinatus est ('I bought the boy some books with red letters in them a little time
ago. I want him to have a taste oI law in order to manage the property. Law has bread in it.
He`s tainted enough as it is with literature`).
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 51
reads it aloud or recites Irom memory. To make this distinction is vital to our
reading, since we wish to argue that although a physical text, the Satvrica
pretends to be a 'spoken discourse, which cannot resort to 'displaying its
contents visually as a written text, but must restrict itselI to 'proclaiming
audibly whatever it has to say. This perIormance aspect oI Encolpius` narra-
tive is then worked into the description oI his past, where all speech (later to
be reported by him in the narrative) tends to come to his attention as sono-
rous utterances in the manner oI his own delivery, and not as texts read by
him and then reproduced. For the modern 'reader oI the Satvrica, the eIIect
created is strictly speaking an illusionthe work is just as much a written
text as any otherbut it is an illusion worth noting Ior the inIormation it
gives about the design oI this narrative.
Instances oI straightIorward reading include the recital oI the estate-diary
(53.19); poetry that is read aloud just aIter being composed on wax tablets
(55.3); written jokes or riddles which are read aloud (56.8); a Latin transla-
tion oI Homer which is recited in a sing-song manner (59.3); a testament
which is read aloud (71.4); and verses mumbled by the poet as he is compos-
ing and writing them down (115.12), which are Iinally recited to the captive
audience oI the poet`s traveling companions, either Irom memory or through
the reading oI the parchment on which they were written (118.1124.2). This
last composition, oI course, is the Bellum Civile, which is by Iar the longest
piece oI Iormal verse in the extant text.
Besides these regular readings there are several instances involving the
recital oI literature Irom memory. This can be seen as related to the process
oI reading because it requires the prior internalization oI a written document
in memory, and was conceptualized in antiquity as the ability to read without
the physical text. Such is the manner in which the Rhetorica ad Herennium
discusses the techniques oI memorization:

Quemadmodum igitur qui litteras sciunt possunt id quod dictatur eis
scribere, et recitare quod scripserunt, item qui mnemonica didicerunt
possunt quod audierunt in locis conlocare et ex his memoriter pronun-
tiare. nam loci cerae aut chartae simillimi sunt, imagines litteris, disposi-
tio et conlocatio imaginum scripturae, pronuntiatio lectioni. (Rhet. Her.
3.17)

Those who know their letters can thereby write out what is dictated to
them and read aloud what they have written. Likewise, those who have
learned the techniques oI memorization can put what they have heard in
'places, and Irom those places deliver it by memory. For the places are
52 1 NARRATI VE
very similar to wax tablets or papyrus, the images like letters, the ar-
rangement and placing oI the images like writing, and the delivery is like
reading.

Even natural and untrained memory is regularly compared to wax writing
tablets in the ancient philosophical and rhetorical contexts.
123
It is oI slight
importance to us whether this analogy is valid as a clinical description oI the
workings oI the human memory; it suIIices that we accept that the accurate
recital oI written texts Irom memory required the texts to be internalized by
some means or other. In the Satvrica, the Latin verb recitare is quite am-
biguous about the presence or absence oI the physical text, and in Iact it does
not make the distinctions that seem vital to us, between 'reading aloud,
'reciting Irom memory or 'extemporizing.
124

In the category oI 'reciting Irom memory we Iind the proIessional
Homeristae conversing in the original Greek oI the poet (59.3); the singing
oI verses Irom the Aeneid at a dinner party (68.4); the singing oI poetry at
work or as entertainment (23.3, 34.1, 64.25, 70.7); the singing oI verses
Irom literary mimes (31.6, 35.6, 53.13, 55.6, 73.3); singing while traveling
(28.5, 62.4); and even the recitation oI poetry in the baths (73.3, 91.3, 92.6),
and in the theater (90.5). The above seem to be instances oI recitation Irom
memory, although it is, understandably, diIIicult to determine in every case
whether we are dealing with recitation Irom memory or extempore recita-
tion.
The third class is that oI pure improvisation or extemporization oI
speeches, poetry and stories. Despite appearances, this activity is also closer
to reading than it is to the living voice. To the trained ancient speaker who
had perIectly mastered the techniques oI memorization, the idea oI reading
Irom memory was so Iamiliar, that even improvised Iormal speaking was
thought oI as the reading oI what one was about to say.
125
The voice oI the
extemporizer is variously restrained by written and memorized schemata,
sententiae and exempla, and as such it is properly a scripted voice. Consider
the astonishing capacity oI the Elder Senecaat least iI we believe the
rhetor`s own perhaps boastIul claims in the introductory words oI his
(re)collection oI controversiae and suasoriaeto recall innumerable senten-
tiae and many whole declamations. Such memory oI textual bits and pieces
no doubt Iormed a crucial part oI rhetorical training, without which speakers

123
Pl. Tht. 191cd; Arist. Mem. 450ab; Cic. Part. 6.26, de Orat. 2.88.360.
124
CI. 53.1, 53.9, 55.2, 56.8, 71.4, 91.3, 90.1, 92.6, 90.5.
125
Cic. Orat. 150, ut in legendo oculus sic animus in dicendo prospiciet quod sequatur ('as the
eye in reading, so the mind in speaking looks ahead to what comes next).
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 53
and poets were unable to extemporize declamations or verses. According to
Quintilian, speaking extempore is the 'greatest outcome (maximus fructus)
and 'most abundant reward (praemium amplissimum) oI the long labor oI
literary studies (Inst. 10. 7. 1).
In the Satvrica the importance oI building a vast literary memory beIore
attempting a Iormal composition is stressed both by the persona oI the rheto-
rician Agamemnon and by that oI the poet Eumolpus, who also happen to be
well versed in each other`s art.
126
In telling his story Encolpius alludes to or
represents completely several extemporal pieces: young Encolpius` declama-
tion in the school portico (1.12.9); Agamemnon`s declamation and subse-
quent extemporization oI a poem (3.15 v. 22); the extemporalis declamatio
oI the speaker who takes over Irom Agamemnon (6.1); Eumolpus` disserta-
tion on the liberal arts and the subsequent extemporization oI a poem on the
capture oI Troy (88.290.1); a shorter poem by Eumolpus (93.2); and more
declamations (101.7103.2, 107.112, 132.8133.1, 137.9 v. 5), both con-
troversiae (15.3, 48.46, 118.2) and suasoriae (6.1). The persona oI Eu-
molpus also delivers two short erotic narratives (85.187.10; 111.1112.8),
which are simply presented as his own recollections although they are
clearly satirical Iictions oI a certain generic texture. In contrast, we get in
Niceros` werewolI story (61.662.14), and in Trimalchio`s ghost story
(63.310), as well as in his poetry (34.10; 55.3), some Iaulty examples oI
literary composition by men who have not had the prerequisite training and
exposure to good literature.
Lastly there is a class oI 'poetry which is not marked as poetry at all but
appears as pure speech, which on occasions makes the transition Irom prose
to poetry and back again within the same syntactical unit (108.13I., 132.11).
Quite oIten the narrator (both while narrating in his own person and during
impersonation) switches Irom one discourse type to another without a warn-

126
4.3, 'ut studiosi iuveni lectione severa irrigarentur . ut quod vellent imitari diu audirent
. iam illa grandis oratio haberet maiestatis suae pondus ('so that bookish boys were
steeped in serious reading . would over a long period listen to pieces they wished to imi-
tate . then that grand style would gain its Iull Iorce and splendor`); 5 vv. 122, 'artis sev-
erae siquid ambit effectus / . / . det primos versibus annos / Maeoniumque bibat felici
pectore fontem. / . / . sic flumine largo / plenus Pierio defundes pectore verba ('II any
man seeks Ior success in the stern art / . / . let him give his Iirst years to verses / and let
his Iortunate heart drink oI the Homeric Iountain / . / . so shall one pour out words in
swelling torrent Irom a heart the Muses love`); 118.36, 'neque concipere aut edere par-
tum mens potest nisi ingenti flumine litterarum inundata . nisi plenus litteris sub onere la-
betur ('the mind cannot conceive or bring Iorth its oIIspring unless it is steeped in the
vast Ilood oI literature . the poet will sink under the burden unless he is Iull oI litera-
ture`).
54 1 NARRATI VE
ing and as iI nothing had happened (14.2, 15.9, 18.6, 79.8, 80.9, 82.5, 83.10,
93.2, 112.2, 126.18, 127.9, 128.6, 131.8, 132.8, 132.15, 133.3, 134.12135.8,
136.6, 137.9, 139.2). Most oI this metrical speech seems to be new composi-
tion, but twice the poetry is actually Virgilian pastiche (112.2, 132.11). This
accounts Ior all the verse in the extant Satvrica.
1.2.6 Human Language and Kunstsprache
What is the point oI the eccentric presentation oI the poetry oI the Satvrica?
Some glimpses oI the theory behind it are preserved in the oIten Irustratingly
limited Iragments oI the work. The exposition oI this theory comes as a part
oI the introduction oI Eumolpus, the compulsive poetaster.
127
AIter the poet
has been pelted out oI the pinacotheca Ior his recitation, young Encolpius is
made to ask him, Quid tibi vis cum isto morbo?'what do you think you are
up to with this disease? (90.3). He then proceeds to remark with amazement
that minus quam duabus horis mecum moraris, et saepius poetice quam hu-
mane locutus es'during the less than two hours that you have been in my
company you have more oIten spoken like a poet than like a man (90.3).
The distinction here drawn is an essential one Ior the Satvrica: just as de-
claimers are possessed by some alien madness (1.1, num alio genere furia-
rum declamatores inquietantur |'are declaimers driven mad by a new kind
oI Furies?|), poets are crazy. The mark by which we recognize their mad-
ness is the way they speak. They do not speak humane, but Iall into various
deIective and contrived speech mannerisms which make them seem pos-
sessed by alien Iuries, or to have taken leave oI their senses (90.4, ego quo-
que sinum meum saxis onerabo, ut quotiescumque coeperis a te exire, san-
guinem tibi a capite mittam |'I too shall load my pockets with stones, and
whenever you begin to go out oI yourselI I shall let blood Irom you head|).
Formal discourse is contrasted with 'human, i.e., spoken language. The
underlying conceit is to recognize no speech as sane other than conversa-
tional language and to measure all discourse types by that standard. Every-
thing which deviates Irom the norm oI urbane colloquialism is Iound to be
alienating, insane, or plain ridiculous. The Iamiliarity and spontaneity oI

127
It might be said that I am here invoking a 'comparison text by selecting a passage oI the
work to serve as a master-key to the interpretation oI the larger context. This is true to some
extent and probably unavoidable in any interpretation. However, my reading does not re-
quire any transIeral oI context because Encolpius` qualiIication oI Eumolpus` poetry as
'mad is already made in the context oI (some oI) the poetry in the Satvrica.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 55
conversational language is such that it seems uniquely 'transparent and
normal and therein lies its unassuming but persuasive authority.
To view poetry as inIerior to plain talk is oI course a reversal oI the clas-
sical order oI discourses. That order, in Iact, is brieIly promoted by the crazy
poet who makes the point that poetry requires that one Ilee the 'vile and
'plebeian common language (118.4, refugiendum est ab omni verborum, ut
ita dicam, vilitate et sumendae voces a plebe semotae). But the mad poet`s
ideas are immediately undercut by his own admission that poetry is, indeed,
madness (118.6, furentis animi vaticinatio). The over-all diIIerence in the
reception aIIorded poetry and prose in the Satvrica is also striking, and
shows how Encolpius himselI does not sustain the classical order. While
Eumolpus the poet is spontaneously stoned by bystanders Ior the recitation
oI his poetry, Eumolpus the teller oI short erotic narratives, which are con-
ducted in the same idiom as the main narrative, gets Iavorable reception
Irom the other characters oI the story.
128

It is important to diIIerentiate here between young Encolpius and the
older narrator,
129
who privileges a loose conversational language, and repre-
sents himselI in the past as being prone both to rhetorical pomposity and
lapses into poetic 'madness. The narrator is considerably more sober, in
this particular sense, and although he does occasionally lose his composure
and 'slip into verse (mostly in connection with such turbulent topics as
Giton or money), he usually corrects himselI immediately in balanced con-
versational prose (e.g., 79.89, sine causa gratulor mihi |'I congratulate
myselI without a reason|), and so is able to continue the narration in a
calmer tone. The occasions when the narrator speaks metrically in his own
voice are relatively Iew, but nevertheless signiIicant. In such 'lapses and
subsequent 'corrections the boundaries oI genres are deliberately crossed
Ior a variety oI reasons, but only in order to be re-conIirmed aIterwards
when the narrator`s voice re-settles into conversational rhythm and dic-
tion.
130
Although the purpose oI the verse is oIten to 'correct the characters
with a comment, such as a Cynic observation about liIe (e.g., 'money can
buy anything), the apparent 'insanity and lack oI seriousness oI the verse
inevitably subverts the message that the narrator (and his characters) are
attempting to deliver.
131
This is a useIul technique Ior many reasons, not the

128
See Beck 1979 Ior a comparison oI Eumolpus as poet and story teller.
129
The diIIerence between narrator and protagonist will be dealt with in greater detail in sec-
tion 3.1.
130
Beck 1973 alternatively proposed that the 'slips into verse by the narrator represented
reconstructions oI the protagonist`s Iantasies.
131
Relihan 1993 correctly emphasizes the selI-subverting quality oI prosimetry without, in my
56 1 NARRATI VE
least because it can be used by the narrator to articulate criticisms indirectly
which he otherwise has no moral authority to back.
The basic theory oI the mixed discourse seems to date at least to the third
century B.C.E. and as Iar as we can tell was invented by Cynic popular phi-
losophers. Though their own words are mostly lost, the type oI philosophical
discourse they practiced, and thus the logic oI their art, is still preserved.
132

There exists, Ior instance, a satiric dialogue by Lucian called Menippus. It
takes the Iorm oI a conversation between the Cynic wise man and an un-
named Iriend who acts the straight man to the clownish philosopher newly
arrived Irom a trip to the underworld. Menippus begins by addressing his
Iriend with two verses, a pastiche Irom a play by Euripides. The Iriend an-
swers with a greeting and a question. Menippus again quotes two lines Irom
another play oI Euripides. This goes on until Menippus has uttered nothing
but pastiches Irom Euripides Iour times in a row and the Iriend gets irritated,
though he is slightly bewildered as to what he should call the type oI lan-
guage that he wishes the philosopher to use, G"T*"&, "'(%&8, #%"/hBI3 '"R
,A/8 -^#4*+ G4. iG,I. '"#"jV. JGQ #I3 M"j8+43'Stop this tragic re-
cital! my dear man, and speak to me this way, somehow, plainly, stepping
oII the iambs! Menippus steps oII the iambs and answers in Homeric hex-
ameters (!), quoting two lines Irom the Odvssev, except that he replaces the
word 'mother in Homer with 'Iriend to suit his addressee. In exasperation
the Iriend retorts, -7#-. J,,` k G"%"G"+8&.l -e /V% _3 -Y#4. 0A#%4.
0%%"mnB8&. G%Q. P3B%". )+,-$.'Man, you are surely out oI your mind, or
you wouldn`t rhapsodize metrically in this way to your Iriends. Menippus
answers (now in conversational language), 1 H"$(*o., p E#"9%8l 384*#R
/V% de%&G+Bo '"R q@X%h *$//83F83-. -e' -rB` 6G4. J38G,X*HK3 #I3
0GI3 '"R "e#F"#( -& #V A#%" 0GR #Q *#F" Z%N8#"&'Don`t be surprised,
my Iriend, having just been together with Euripides and Homer |in the un-
derworld| somehowI don`t know howI`ve become Iull oI words/phrases
and spontaneous verses come to my lips.
We notice that, just as in the Satvrica, poetic speech is here treated
humorously as a compulsive linguistic aIIliction; in the case oI Menippus it
is caused by too much time spent with Iamous dead poets, which is clever

view, adequately explaining the logic oI the interplay between discourse types, and how this
logic undercuts opinions and ideas that are being expressed.
132
There exists no group oI Cynic texts comparable with the Stoic and Epicurean collections.
Editions oI the Iragments are rare and hard to come by and as a result 'Cynicism is oIten
described on the basis oI Musonius RuIus, Epictetus, and Dio Chrysostom, whose views are
to a large extent Stoic. Lucian is a good source at least with respect to the nature oI Cynic
literature. The material in Diogenes Laertius is mostly anecdotal, and Julian`s writings on
the topic are late and biased. The Cynic epistles are an important but neglected source oI in-
Iormation.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 57
caused by too much time spent with Iamous dead poets, which is clever
metonymy Ior excessive exposure to the Classics. Both Eumolpus and
Menippus, who essentially belong to the same type, lose control and begin to
speak in meter automatically, but return to humanity and sanity only when
they speak again like ordinarily people. Switching Irom poetry to conversa-
tional prose restores normality. Another important similarity is the idea oI
being 'Iull oI words/phrases and meters which must break out in ever new
recitals. Agamemnon, Eumolpus, and young Encolpius use the same meta-
phor to describe the desired eIIects oI literary studies, and the state oI literary
inspiration (5. v. 1322, 118.6, 126.8).
133
It is as iI the speaker had Iirst be-
come saturated or satur (Iood-metaphors come easily to the lips oI these
characters)
134
with literature and had subsequently gone out oI himselI or lost
his identity and normal manner oI speech to become possessed by poetic
personae Irom a gallery oI classical authors.
OI course, this eIIect upon the schoolman`s psyche is even more worri-
some iI the literature one is being exposed to is oI inIerior quality. In his
declamation Ior Agamemnon, young Encolpius Iirst laments the unreal sub-
ject matter and bombastic style oI declamation, which he calls mellitos ver-
borum globulos et omnia dicta factaque quasi papavere et sesamo sparsa
'honey-balls oI phrases, every word and act besprinkled with poppy-seed
and sesame (1.3), and then he says that qui inter haec nutriuntur non magis
sapere possunt quam bene olere qui in culina habitant'people who are Ied
on this diet can no more have taste (or be wise) than people who live in the
kitchen can smell good (2.1). One can sense here, in the Iigures oI the lan-
guage, a tone reminiscent oI an ancient philosophical polemic against soph-
ism and rhetoric, which had been likened to the (undervalued?) proIession oI

133
Sat. 5 vv. 1322, bibat felici pectore fontem. / mox et Socratico plenus grege [.] sic
flumine largo / plenus Pierio defundes pectore verba; 118.6, civilis ingens opus quisquis at-
tigerit nisi plenus litteris, sub onere labetur; 126.8, itaque oratione plandissima plenus [.]
inquam. The Iolaos (POxv 3010) Iragment provides a tantalizing parallel, which comes just
beIore the 'gallus begins addressing Iolaos in Sotadean verse, 910, |B&B"|Nfc BC G,X%|Kc
/8/-3;c| 0Gf,H8 G%Qc |#Q3 asF,"-3| 'having become Iull oI learning he went to Iolaos.
Although the text is lacunose, the letters G,X%|.| show that we have here the same meta-
phor oI being 'Iull oI learning (letters, phrases, words, poetry) beIore bursting into poetic
speech.
134
Juvenal speaks oI the farrago 'hodgepodge oI his little book (1.86). Consider also this
Iragment oI Varro`s Saturae Menippeae (Eumenides XXVIII): '. and we the other
schoolmen rise together, our ears saturated Irom the scholastic banquet and inebriated Irom
the interminable sophistry, but our eyes starving (et ceteri scholastici saturis auribus
scholica dape atque ebriis sophistice aperantologia consurgimus ieiunis oculis!). Satura
was also the name oI a dish oI mixed ingredients. For a general discussion oI Iood associa-
tions and satura, see Petersmann 1986, LeMoine 1991, and Gowers 1993.
58 1 NARRATI VE
cooking. This is the thesis that arranging words artIully is, like seasoning
Iood, basically a deception.
135
But apart Irom this there is the notion that
excessive exposure to artIul literature will Iragment the student`s subject and
control over Iuture compositions. Seneca, in a letter dedicated to the practice
oI literary studies (Ep. 84), shows awareness oI the possibly negative eIIects
oI ancient education through immersion, and strongly advises students to
'digest (Ep. 84.7, concoquere) the intellectual Iood they take in through
their reading so that it can truly be assimilated into the bloodstream, or the
seat oI genius (ingenium), not just into memory. Many voices singing in
harmony is the classical ideal oI generic oneness and consonance (Ep. 84.9).
The digestion is the process oI transIorming and adapting the miscellany that
one reads to the generic law oI the new compositionwhether that is one`s
speech or writingwhich should ideally level out any signs oI alien dis-
courses and make what came in many Iorms and shapes into a seamless
whole.
136

In the earliest prosimetry the ideal oI consonance is deliberately botched,
and we are allowed to see the oIten stark diIIerences between individual
components oI a work. It is this Ieature which gives the prosimetric narrator
his scatterbrained quality. By eschewing consistency and unity, the mixed
discourse not only deprives itselI oI an authoritative stance, it also subverts
conservative and contrived genres by playing with them and subjecting them
to jarring juxtapositions. The purpose oI this exercise is ultimately to make it
all the more obvious how strange and artiIicial and pompous literary diction
and specialized jargon are as compared with common language. This type oI
prosimetry, though seemingly anarchic in its shape-shiIting guise, is there-
Iore not without a center and does possess a casual or nonchalant authority
oI its own.
137
The conversational style is at its heart and Iunctions as a sort oI
non-literary center, which presumably is situated beyond Iormal language
and can thereIore serve as a measure oI the artiIiciality oI Kunstsprache.

135
See Conte 1992, 300312.
136
CI. the words oI the crazy poet Eumolpus in 118.5, curandum est ne sententiae emineant
extra corpus orationis expressae, sed intexto vestibus colore niteant'care should be taken
that the (rhetorical) sententiae not be obvious and stand out Irom the body oI the discourse,
but be interwoven and shine with the color oI the garment.
137
Whether PROSImetry, as opposed to prosiMETRY, is the original Iunction and purpose oI
the mixed discourse I dare not say with absolute certainty. Practically nothing is leIt oI the
works oI Menippus and Varro`s Menippean Satires are all in pieces. In order to determine
the Iunction oI the mixed discourse we need relatively complete sections oI text. Judging
Irom what is extant either Iully or partly (Menippus, Iolaos, Satvrica, Apocolocvntosis) the
basis oI the oldest prosimetry was simple prose.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 59
The one type oI discourse certainly privileged in this genre is prose, and
we should remember that the vast majority oI the extant Satvrica (and we
may assume oI the original work) is written in this idiom. The 'oral
presentation oI the various discourse types Iound in the Satvrica is also what
constitutes their uniIying quality and shows clearly the generic dominance oI
the conversational tone oI Encolpius` voice vis-a-vis other types oI dis-
course. His is the central speaker identity, underlying all other personae in
the work, and thereIore it should cause no surprise that all subordinate per-
sonae speak in Iorms oI conversational language, though the quality and
style oIten diIIer in accordance with their intended individual traits as char-
acters in the story.
1.2.7 Barbarian, Monstrous, Not Even Human .
Perhaps the most Irequently studied consequence oI the 'human language
mode oI presentation in the Satvrica is the well-known colloquial quality oI
its diction, morphology and syntax. We typically have two distinctions, one
internal, between what is spoken in the name oI the narrator and the other
educated characters and what is spoken in the personae oI the Ireedmen, and
one external, between the language oI the Satvrica as a whole and the style
oI Iormal literary Latin. II the linguistic analyses oI Bendz (1941), Mar-
morale (1948), Swanson (1963), Dell`Era (1970) and Petersmann (1977) are
to be trusted, not only the speeches oI the Ireedmen in the Cena, but also
Encolpius` main narrative and the conversations oI his peers are rich in col-
loquialisms. This over-all quality oI the language is oIten used to justiIy the
diIIerentiation Irom the more elevated literary idiom.
While we are able to determine with some certainty the mode oI presen-
tation in the Satvrica as 'spoken-to-be-heard, as opposed to 'written-to-be-
read, the classiIication oI the language as 'colloquial or 'literary is mean-
ingIul only in a relative sense, and depends on our choice oI 'comparison
text to serve as a typical example oI literary language. Because oI the anti-
quity oI the subject, all our evidence Ior spoken Latin comes Irom written
texts, and no written text is so 'literary as to be Iree oI the spoken language.
For obvious reasons these categories are never discrete. Moreover, as we
have clearly seen in our analysis oI the discourse types in the Satvrica, the
spoken mode oI presentation does not exclude literary language, since Ior the
composition oI ancient 'oral discourse, whether in prose or in metrical lan-
guage, a great repertoire oI literary bits and pieces was required.
60 1 NARRATI VE
The internal distinction, however, between the utterances oI the schooled
and un-schooled characters is arrived at by comparing the speech manner-
isms oI these two diIIerent groups in the Satvrica. In reading through the
work one senses a clear increase in the density oI barbarisms and solecisms
in the speeches oI the un-schooled Ireedmen. Representing this change by
statistical means is Iraught with diIIiculties, but Iortunately unnecessary
because Encolpius clearly acknowledges and intends this diIIerence. In the
persona oI Echion, who is perhaps the most conIident and the least apolo-
getic oI Trimalchio`s Ireedmen guests, the story-teller addresses Agamem-
non, the rhetor, in this Iashion (46.2):

'Videris mihi, Agamemnon, dicere: quid iste argutat molestus?` quia tu,
qui potes loquere, non loquis. non es nostrae Iasciae, et ideo pauperorum
verba derides. scimus te prae litteras Iatuum esse.

'Agamemnon, it seems to me that you are saying: Why is this tiring Iel-
low prattling on?` Well, it`s because you, who know how to speak, don`t
say anything. You`re not made oI our stuII, and so you make Iun oI how
we poor bastards talk. But we know that you`ve become silly Irom litera-
ture.

This obviously deIines the categories oI the Ireedmen`s speech versus that oI
the scholars in terms oI literary and non-literary, or schooled and un-
schooled. The same is expressed by the words oI another Ireedman in Tri-
malchio`s party, Niceros, beIore he begins his werewolI-story: timeo istos
scholasticos, ne me derideant 'I Iear those scholastics, that they will mock
me (61.4). No external context or comparison text is really necessary to
explain what characterizes this style with respect to the rest oI the narrative,
or why Encolpius has introduced this Ieature into his lively account oI Tri-
malchio`s dinner-party. Nevertheless, scholars have oIten attempted to use
external material (e.g., the apparently conversational language oI Cicero`s
and Seneca`s letters or the ungrammatical language oI Pompeian graIIiti) to
Iix a supposedly absolute and historical classiIication oI the two levels oI
spoken language in the Satvrica. The use oI such external contexts is oIten
accompanied by certain assumptions about the purpose oI this style, assump-
tions that sometimes ignore the indications in the Satvrica itselI as to its
intended signiIicance.
Ever since the rediscovery oI the Traguriensis manuscript by Marino
Statileo, in the middle oI the seventeenth century, scholars have debated the
signiIicance oI the strange language oI the Satvrica. Initially the question oI
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 61
language was raised in relation to the debate over the authenticity oI the
Cena manuscript.
138
Johan Christoph Wagenseil remarked that the barbara,
monstrosa, immo ne humana quidem'barbarian, monstrous, not even hu-
man vocabulary oI the Satvrica, especially in the speeches oI the Ireedmen,
could not possibly have come Irom the pen oI the elegantiae Arbiter, and
tried to show that the real author had to be Iamiliar with modern Italian. The
biographical bias oI the argument is obvious and Wagenseil makes no allow-
ance Ior the possibility that the author might have imitated linguistic man-
ners that were uncharacteristic Ior his own historical person. On the basis oI
this position, Wagenseil concluded that the new Iragment oI Petronius was
not by Petronius, and not even by a Roman, but an obvious Iorgery perpe-
trated by the Italian Statileo himselI.
139

It is here that we witness the invention oI the anachronistic Petronian
criteria oI style that have exerted so much inIluence on the literary and the
textual criticism oI the Satvrica. While Tacitus, in his necrology oI the Ro-
man consular C. Petronius (Ann. 16.18), reIers ironically to elegantiae arbi-
ter as his supposed Iancy title within Nero`s court and indicates that its true
meaning was 'expert at devising perverse pleasures, the language and style-
Iixated humanists had something other and more speciIic in mind when writ-
ing about Petronian elegantia. A key Iigure Ior understanding the meaning
oI the word elegantia in humanist Latin is the Italian humanist Lorenzo
Valla (d. 1457) whose seminal book De linguae Latinae elegantia libri sex
attempted Ior the Iirst time ever to give a description oI the usus or Latin
usage oI classical Roman authors. The work was enormously ambitious and
it was intended to replace all earlier (mostly medieval) grammatical treatises,
but even more importantly it laid the Iundaments Ior a new subject: historical
linguistics. An important part oI the treatise was historical in nature and con-
tained the theory oI the cyclic change oI Latin style Irom coarse beginnings
through reIinement and classical perIection in the time oI Cicero to degen-
eration and downIall at the end oI antiquity and the beginning oI the middle
ages. The development oI Latin, Valla claimed, kept pace with the growth
and decline oI the Roman Empire, and he thus substantiated the persistent
humanist notion oI a correlation between the quality oI the language (the
style) and the state oI the culture.
140
It is this work`s enormously inIluential

138
In the Iollowing exposition oI the scholarly literature on the language oI the Satvrica I owe
much to Bret Boyce`s survey oI the scholarship (Boyce 1991, 1434).
139
Wagenseil`s treatise is the Iormer oI the two Dissertationes de Cena Trimalchionis nuper
sub Petronii nomine vulgata (orig. edn. Paris 1666) that were reprinted in Burman 1743,
2:342350.
140
Lindberg 1993, 5455.
62 1 NARRATI VE
thesis about the rise and Iall oI Latin elegantia with all the connotations the
word had gathered through the humanist project which Iorms the background
oI Wagenseil`s denial oI the possibility that a text like the Cena, Iull oI bar-
barisms and solecisms, could possibly have been written by an ancient Ro-
man author.
Adrien de Valois, whose study appeared together with that oI Wagenseil
and under the same title,
141
argued along similar lines. Whoever wrote the
Cena Trimalchionis, he claimed, had to be a native speaker oIthis time
French, though Ior de Valois, the colloquial language oI the narrative and the
Grecisms oI the Longer Fragments indicated that Petronius lived under the
Antonines and was himselI indeed a Gaul Irom Massalia (see more on Mas-
salia below). Again the premise oI the argument is that the persona oI the
speaker in the narrative oI the Satvrica is throughout none other than the
historical author, Petronius Arbiter. This assumption then leads to certain
expectations about the style appropriate to the author`s historical persona,
expectations which are ultimately determined by a comparative reading oI
the Cena with Tacitus and Valla as comparison texts. A notable characteris-
tic oI their approach is the completely new attempt to classiIy the strange
language oI the work by introducing a non-Latin 'comparison text, in both
cases the written texts oI the Romance vernaculars. From the context we
may glean that although the Romance languages had by then come into their
own as written languages, they were still in the eyes oI these learned men
regarded as corrupt Latin.
A quick reply to these attacks on the authenticity oI the Traguriensis
manuscript was published the same year (1666) in Paris under the name oI
Marino Statileo himselI.
142
This treatise correctly stresses that the language
oI the Ireedmen deviates Irom the language oI the other characters because
oI the inIerior level oI their schooling and literacy, and claims moreover that
Petronius deliberately wrote thus to make the language suit his characters
(personae). The author oI this reply notes that the new Traguriensis manu-
script contains such peculiar language, while the previously known Longer
Fragments (L) don`t, Ior the reason that only in the episode oI the dinner-
party is conversation dominated by uneducated Ireedmen.
143
The author oI
the treatise shows a good understanding oI the Iact that the historical identity
oI Petronius as such is irrelevant to the language oI the Ireedmen, which
should instead be explained by reIerence to the ancient criteria oI appropri-

141
Reprinted in Burman 1743, 2:350358.
142
Reprinted in Burman 1743, 2:359379. The reply may not have been written by Statileo,
but by Pierre Petit oI the Pleade latine; according to Gaselee 1909, 171.
143
Burman 1743, 2:367.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 63
ateness in sermocinatio.
144
The assumption, that apart Irom the Ireedmen all
characters in the Satvrica are urbane and educated, may not be Iully accu-
rate, but it is certainly correct that oI the extant episodes only the Cena al-
lows uneducated characters to make longish statements.
145

Another important seventeenth-century dissertation under the name oI
Statileo, but actually composed by Giovanni Lucio,
146
introduced the ques-
tion oI Vulgar Latin into the scholarship on the Satvrica. By noting the simi-
larity oI the language oI the Satvrica to the Vulgate and to the Romance
languages, the scholar postulated that the colloquial language oI the work
was actually derived Irom the way Latin was spoken at the time by common
uneducated people. However, in Lucio`s mind, what prompted Petronius to
imitate this idiom was by no means the desire to document the language oI
the people, but an urge to ridicule and 'to laugh until his tears Ilowed, to-
gether with the educated and elegant men who were present at that dinner
party.
147
In the story, however, it is not Petronius who laughed at the Ireed-
men, but the character Ascyltos, as is clear Irom the Iact that Lucio`s lan-
guage imitates Sat. 57.1.
148
Besides, this interpretation obscures the Iact that
the jokes oI the Cena Trimalchionis are told no less at the expense oI the
parasitic and clownish scholastici. There is no question that the narrator En-
colpius agrees with the mordant wit oI Echion, when he states in his un-
schooled manner that the scholastics have lost their senses Irom excessive
study oI bad literature. Such is, as we have seen, the basic philosophy oI
prosimetry.
In general it is the character oI this seventeenth-century debate to treat
the language oI the Ireedmen not as an idiom on its own, which might be
worth studying, but merely as an aberration Irom the correct Latin oI Cicero
and other exemplary authorities on matters oI style. It took the intense ro-
mantic nationalism oI the nineteenth century to prod German philologists
into Iinding in vulgaris sermo not just low 'vulgarities but the Jolks-

144
The author asks, justly outraged: 'But what is this stupidity, not to notice what is appropri-
ate to the characters that Petronius portrays? (Quae autem haec inscitia est, ne animad-
vertere quidem, quid personis, quas Petronius exhibet, conveniat? Burman 1743, 2.373).
145
The character Bargates speaks 'in a raving and barbaric voice (96.5, rabiosa barbaraque
voce).
146
Apologia ad patres conscripti reip(ublicae) litterariae Marini Statilei Traguriensis, Iirst
published in Blaeu`s edition (Amsterdam 1670), and then reprinted in Burman 1743, 2:379
394.
147
Burman 1743, 2:387.
148
Sat. 57.1, ceterum Ascvltos, intemperantis licentiae, cum omnia sublatis manibus eluderet et
usque ad lacrimas rideret ('But Ascyltos, unruly and out oI control as usual, kept throwing
his hands up ridiculing everything and laughing until his tears Ilowed).
64 1 NARRATI VE
sprache, 'the language oI the people, and so to make this Iormerly disrepu-
table idiom worthy oI study in its own right.
149

The Iirst study to approach the question oI language in the Satvrica Irom
this modern perspective is an article oI G. Studer (1849) praising Petronius
as practically the only Roman author who 'has leIt behind Ior us a written
document (der uns ein schriftliches Document [.] hinterlassen hat) oI the
Jolkssprache that was spoken by the overwhelming majority oI uneducated
Romans, in the military and among the colonists oI the provinces, where it
later developed into the various Romance languages. The importance oI
Petronius as a document Ior this idiom is said by Studer to be even greater
because oI the loss oI all works which, like the mimes and Atellans, be-
longed to the 'base and comic national literature (der niedrigkomischen
Nationallitteratur). A new ideological agenda is clearly aIoot in this study
that reveals itselI in terms like Jolkssprache, Document and Nationallitte-
ratur. Studer`s article marks a radical shiIt in the discussion oI the language
oI the Ireedmen and involves a Iundamental departure Irom the humanist
premises oI the seventeenth century.
150

As Bret Boyce points out,
151
several major projects now inIluenced the
debate on the Satvrica`s language: G. Fiorelli`s archeological excavations oI
Pompeii (begun 1860); the publication oI Bcheler`s new edition oI the text
(1862); the publication oI the epigraphic evidence in the Corpus Inscrip-
tionum Latinarum (1863); and the publication oI Hugo Schuchardt`s three
volume study oI the sermo plebeius (18661868). As the result oI this devel-
opment we have, moreover, a doctoral dissertation by Arminius von
Guericke, De linguae vulgaris reliquiis apud Petronium et in inscriptionibus
parietariis Pompeianis (1875), which compares the language oI the Ireed-
men to the graIIiti on the walls oI ancient Pompeii. The conclusion oI this
study, however, is that the language oI the Satvrica, though interspersed with
vulgar Iorms, including the poetry, is not a IaithIul representation oI the con-
temporary Latin spoken by the lower classes. According to von Guericke this
is so because oI the constraints on the author to entertain an educated audi-
ence. This negative conclusion was, oI course, predictable, because no one in

149
The Iirst German ideologue to regard language as such as the most signiIicant expression oI
the Jolksgeist was Johann GottIried Herder (b. 1744, d.1803), but aIter the 1806 Prussian
deIeat at Jena, his ideas were disseminated in the more aggressive Iorm given them by Jo-
hann Gottlieb Fichte in his Reden an die deutsche Nation, which he delivered in Berlin in
the winter oI 18071808.
150
See section 3.2.2 Ior a Iuller discussion oI the inIluence oI nationalism on Petronian schol-
arship.
151
Boyce 1991, 1434.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 65
the Iirst century C.E. would have seen any point in giving a IaithIul repre-
sentation oI the language oI the common people. The rationale Ior doing so
did not exist until the nineteenth century.
The archeological evidence and interest in historical linguistics (and the
centrality oI language in the ideologies related to the contemporary consoli-
dation oI national states in Europe) tended to obscure rather than to clariIy
the speciIic logic oI the presentation oI linguistic characterization in the Sa-
tvrica. Wilhelm S was the Iirst modern scholar to draw attention to the
important diIIerence between the modern and the ancient approaches to Vul-
gar Latin as it related to the Satvrica. In two Latin treatises, De eo quem
dicunt inesse Trimalchionis cenae sermone vulgari (1926) and Petronii imi-
tatio sermonis plebei qua necessitate coniungatur cum grammatica illius
aetatis doctrina (1927), he tries to diIIerentiate the study oI Latin historical
linguistics Irom the study oI the speciIic reasons behind the imitation oI
speech characteristics in the Satvrica, which he emphasizes as being the
more important Ior understanding the work.
152

Barbaric and solecistic language in the Satvrica should be read and in-
terpreted in the context oI those passages in the work where the language oI
the Ireedmen is deIined as un-schooled and ridiculous in the eyes oI the
scholastics. According to that context the speeches oI the Ireedmen are there
to Iorm an antithesis to the correct Latin oI the schools.
153
Encolpius` attempt
to imitate the language oI the Ireedmen is surely called Ior because oI the
presence oI the scholastici at the dinner party oI Trimalchio. What tempts
him to undertake this exercise in mimetismo is the contrast between these
two groups. The episode is a sort oI Saturnalian reversal oI roles, where
scholars and orators are kept silent to allow those who violate the protocols
oI respectable public speech to ramble on in their 'shameIul and 'low
manner. S is also correct in arguing against the idea that Petronius tried to
imitate the speech mannerisms oI those who had Latin as their second lan-
guage.
154
Grecisms are widespread throughout the Satvrica and we may rest
assured that iI Encolpius had intended to ridicule the Ireedmen especially Ior
their use oI Grecisms he would somehow have indicated as much.
Wilhelm S makes it the starting point oI his latter thesis to investigate
the ancient grammatical theory oI barbarisms and solecisms. His conclusion
is that the language oI the Ireedmen displays many oI the Ieatures known to

152
S 1926, 3: . non modo grammaticis singula esse ex historia et necessitate orationis
latinae interpretanda, sed et philologis quaerendum esse, quo sit ductus consilio scriptor et
qua sit usus ratione in coloribus vulgaribus dicendi et eligendis et distribuendis.
153
For a discussion oI ancient schoolmen and society, see Kaster 1988.
154
This view was held, Ior example, by Salonius 1927.
66 1 NARRATI VE
grammarians as incorrect, although he avoids making this the only source
and says that in creating the language oI the Ireedmen the author 'employed
sharp wit together with learned practice and the design oI his art in order that
all might seem casual and haphazard.
155
S thus relies on the study oI the
external context oI Vulgar Latin only in so Iar as it supports the internal
mode oI presentation. This approach has the great advantage oI respecting
the Satvrica`s own explanation oI the signiIicance and purpose oI the lan-
guage oI the Ireedmen, i.e. oI reading those passages in the context oI the
rest oI the Iragments.
It is simply anachronistic to assign to the author oI the Satvrica an inter-
est in imitating the language oI the common people Ior the sake oI realism or
documentation, although this has been the unexamined assumption oI mod-
ern scholars who have studied Jolkssprache in the Satvrica.
156
In the section
above we accounted Ior this reading as determined by the context oI the
other extant ancient prose Iiction in Latin, the Metamorphoses oI Apuleius,
and pointed out that as an interpretive approach it shares many basic assump-
tions and premises with the general study oI national literatures. Although
modern scholars are obviously Iree to use any ancient text as a linguistic
document to support their theories about the vulgar idiom oI ancient Latin,
the assumption should be resisted that the peculiar language oI the un-
schooled characters oI the Satvrica is there Ior the reason that scholars wish
to study it.
II we are to revise the historical context oI the colloquial language oI the
Satvrica to dissociate the work Irom such texts as the Vulgate or the graIIiti
at Pompeii, the most obvious replacement Ior these traditional 'comparison
texts should be looked Ior in the appropriately general and theoretical an-
cient discussion oI styles. Here, indeed, we Iind in the lowest oI the three
stylistic characters, as deIined by the Rhetorica ad Herennium, a text which
shares many oI the linguistic peculiarities oI the Satvrica, but more impor-
tantly deIines a type oI style the wrong use oI which can accommodate a
density oI barbarisms and solecisms oI the kind we Iind in the Ireedmen`s
problematic discourse. This Latin treatise Irom the Iirst century B.C.E. pro-

155
S 1927, 5, miscuisse innata acumina ingenii cum usu eruditionis et consiliis artis ita, ut
omnia nescio quam simulationem temeritatis haberent.
156
Boyce 1991, 24, whose study has been useIul to me Ior its thoroughness in surveying the
scholarship, seems to believe that Petronius` purpose was to document the linguistic man-
nerisms oI the underclass, and he agrees with the negative and dismissive treatment oI
S`s thesis: In der Cena haben wir doch nichts anderes als ein von unerreichter Meister-
hand geschaffenes Abbild der damaligen Jolkssprache, wo:u ein drres Svstem von Jitia-
gruppen eines Grammatikers nichts beigetragen hat (quoted by Boyce).
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 67
vides us indeed with an example oI 'plain speech, sermo adtenuatus or
figura extenuata, which reads like a lost passage Irom the Satvrica:

Nam ut Iorte hic in balneas venit, coepit, postquam perIusus est, deIri-
cari; deinde, ubi visum est ut in alveum descenderet, ecce tibi iste de
traverso: 'Heus, inquit, 'adolescens, pueri tui modo me pulsarunt; satis
Iacias oportet. Hic, qui id aetatis ab ignoto praeter consuetudinem ap-
pellatus esset, erubuit. Iste clarius eadem et alia dicere coepit. Hic vix:
'Tamen, inquit, 'sine me considerare. Tum vero iste clamare voce ista
quae perIacile cuivis rubores eicere potest; ita petulans est atque acerba:
ne ad solarium quidem, ut mihi videtur, sed pone scaenam et in eiusmodi
locis exercitata. Conturbatus est adolescens; nec mirum, cui etiam nunc
pedagogi lites ad oriculas versarentur inperito huiusmodi conviciorum.
Ubi enim iste vidisset scurram exhausto rubore, qui se putaret nihil
habere quod de existimatione perderet, ut omnia sine Iamae detrimento
Iacere posset? (Rhet. Her. 4.10)

Now our Iriend happened to enter the baths, and, aIter washing, was be-
ginning to be rubbed down. Then, just as he decided to go down into the
pool, suddenly this Iellow turned up. 'Say, young chap, said he, 'your
slave boys have just beat me; you must make it good. The young man
grew red, Ior at his age he was not used to being hailed by a stranger.
This creature started to shout the same words, and more, in a louder
voice. With diIIiculty the youth replied: 'Well, but let me look into the
matter. Right then the Iellow cries out in that tone oI his that might well
Iorce blushes Irom any one; this is how aggressive and harsh it isa
tone certainly not practiced in the neighborhood oI the Sundial, I would
say, but backstage, and in places oI that kind. The young man was em-
barrassed. And no wonder, Ior his ears still rang with the scolding oI his
tutor, and he was not used to abusive language oI this kind. For where
would he have seen a buIIoon, with not a blush leIt, who thought oI him-
selI as having no good name to lose, so that he could do anything he
liked without damage to his reputation? (trans. Caplan)

In the example we note that the author oI the treatise marks the tone oI voice
and gesture appropriate to the stylistic category oI plain speech: The clamor-
ous shouting oI the scurra as well as the young man`s blushing. There is also
an appropriate linguistic setting in the baths, which the moralizing narrator
ut mihi videturclassiIies with the 'backstage and such places and con-
trasts with the language oI Iiner locations at the Sundial. The contrast be-
68 1 NARRATI VE
tween the young and noble youth and the shameless scurra deIines the style
in terms oI social class. The tone oI voice and gesture, the color oI the set-
ting, and the diIIerence in social status between the two characters, all oI this
contributes to marking the passage as 'low or 'thin (sermo adtenuatus)
and Iree oI ornament. Without these narratorial comments a Iormal stylistic
analysis is worth little, Ior the simple reason that the meaning oI linguistic
Ieatures in literature is primarily dependent on context, viz. who is speaking,
where and why.
157

Not even Caplan`s excellent translation above may do justice to the style
oI the passage, whose characteristics are, according to deIinition, the absence
oI rhetorical embellishment and the quick-witted and streetwise use oI lan-
guage (facetissima verborum adtenuatione).
158
Accordingly, the language oI
this style may Ireely sink to the lowest everyday usage (4.10, ad infimum et
cotidianum sermonem demissum). The middle style steers clear oI it by ab-
staining Irom the basest and thoroughly vulgar vocabulary (Rhet. Her. 4.8,
ex infima et pervulgatissima verborum dignitate). In the example given by
the rhetorical treatise the colloquial Iorms come Irom the narrator himselI
(the anonymous author oI the treatise) and not merely in his mimicking oI
the scurra. The whole point oI the base style is to lower the dignity oI the
discourse down to the street level. The 'thin style certainly works as a de-
scription oI the language Encolpius uses in his inIormal narrative and
speeches put in the mouth oI most oI the characters, but can it accommodate
the barbarisms and solecisms oI his language when he dons the masks oI the
Ireedmen in the Cena? The answer to this question seems to be negative,
because elsewhere the treatise deIines correct Latinity, Latinitas, as the basis
oI any appropriate and Iinished example oI each oI the three stylistic regis-

157
Harry Caplan, the editor and translator oI the Loeb text, conveniently summarizes how
scholars have tried to point to speciIic Iormal (lexical, morphological and syntactical) stylis-
tic Ieatures oI the passage to justiIy its classiIication by the author oI the Rhetorica ad He-
rennium. He mentions 'the Iorms oI colloquial usage (pedagogi, the diminutive oriculas),
idioms like de traverso, coepit with the passive, the vulgar use oI the archaism pone Ior
post, and oI the indicative potest in a characterizing clause, the expletive use as in conversa-
tion oI the ethical dative tibi with ecce, the Irequent use oI the demonstrative iste Ior hic or
is, the accusative oI quality in id aetatis, the asyndeton in satis facias oportet, and the type
oI parataxis characteristic oI comedy in ita petulans est . exercitata; Caplan also men-
tions the exclamations heus, eicere in the sense oI efferre, the phrase quod de existimatione
perderet, and the brevity oI Hic vix. Such Iormal analysis is interesting in itselI but cannot
ultimately explain the signiIicance oI the language in the passage (words and Iorms can sig-
niIy many things, e.g. archaic or vulgar) which is primarily deIined by the classiIication
provided by the context in which it occurs, i.e. the rhetorical, social, and ethical theory oI
the Rhetorica ad Herennium.
158
Rhet. Her. 4.11.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 69
ters. And barbarisms and solecisms happen to be the two Iaults in language
that lessen the purity oI the Latin language.
159
We must thereIore conclude
that the correct use oI the 'thin style would apparently not allow the lan-
guage oI the Ireedmen.
The Greco-Roman theory oI stylistics put Iorward in the Rhetorica ad
Herennium names grammatically correct Latinity as one oI the two corner-
stones oI literary taste, elegantia. Although the subject is Latinitas, the
terminology is nevertheless adapted Irom Greek stylistics, either directly, as
in the Grecisms barbarismus and soloecismus, by translation, as in the
calque elegantia (0',-/1 23-(#43), or by neologism, as in the term Lati-
nitas (corresponding to qt,,K3&*F.). It goes without saying that the thesis
oI Latinitas as a prerequisite Ior elegantia can potentially have devastating
consequences Ior Petronian elegantia. For, unless we regard the singularly
barbaric and solecistic language oI the Ireedmen merely as a demonstration
oI bad taste, we must conclude that according to the theory the author oI the
Satvrica has written a text that is the very antithesis oI 'elegant. This reali-
zation should, however, not come as a surprise to us, because Encolpius has
told us as much through his narrative, i.e. that the language oI the Ireedmen
is the very opposite oI the language oI the scholastici.
Fortunately Ior our ability to read the Satvrica in the context oI rhetorical
theory, the opposite oI what is desired and taught in schools is also articu-
lated in the theoretical literature. For each oI the three styles the Rhetorica
ad Herennium has an example oI the incorrect use oI that style. These exam-
ples are certainly not provided Ior imitation but rather Ior pedagogical rea-
sons, in order that the orator may know what to avoid. Nevertheless, the
unknown author oI the treatise has himselI composed these examples oI
oratio vitiosa not because he lapsed into vice (as he warns the reader not to
do) but because he, like Petronius, purposeIully broke the rules oI correct

159
Rhet. Her. 4.17, Elegantia est, quae facit, ut locus~ unus quisque pure et aperte dici
videatur. Haec tribuitur in Latinitatem et explanationem. Latinitas est, quae sermonem pu-
rum conservat, ab omni vitio remotum. Jitia in sermone, quo minus is Latinus sit, duo pos-
sunt esse. soloecismus et barbarismus. Soloecismus est, cum in verbis pluribus consequens
verbum superius non ad~commodatur. Barbarismus est, cum verbis aliquid~ vitiose ef-
feratur. Haec qua ratione vitare possimus in arte grammatica dilucide dicemus ('Elegantia
is what makes each and every passage seem spoken with purity and openness. Elegantia is
subdivided into Latinitas and clarity. Latinitas is what keeps the language pure, and Iree oI
any Iault. The Iaults in language which make the language less Latin, are two: solecism and
barbarism. Solecism is when the concord between a word and one beIore it in a group oI
words is Iaulty |i.e. ungrammatical phrases|. Barbarism is when the verbal expression is
incorrect |i.e. Iaulty word Iorms|. We shall explain clearly in the grammar how to avoid
these).
70 1 NARRATI VE
Latinity Ior the sake oI demonstration and art. The example oI the incorrect
use oI the 'thin style runs as Iollows:

Nam istic in balineis accessit ad hunc; postea dicit: 'Hic tuus servus me
pulsavit. Postea dicit hic illi: 'Considerabo. Post ille convicium Iecit et
magis magisque praesente multis clamavit. (Rhet. Her 4.16)

Now that guy came up to this one in the baths. AIter that he says: 'This
your slave boy beat me. AIter that this one says to the other one: 'I`ll
think about it. AIterwards the other one called him names and shouted
louder and louder, while a lot oI peoples were present.

I should state that my rendering (inspired by Caplan`s) is supposed to repre-
sent the bad Latin. We notice immediately that this text is intended as a
Iailed attempt, an incompetent blurt, at relating the same encounter oI the
young innocent with the scurra at the baths. In Iact, it is a successIul and
competent demonstration oI the erroneous style. The diction here shows that,
even iI the correct use oI the sermo adtenuatus doesn`t, its incorrect use cer-
tainly does allow the indignities inIlicted on the Latin language by the
Ireedmen at Trimalchio`s table. Despite the brevityit is much shorter than
the example oI correct sermo adtenuatus given earlierthis example oI
vitiosa oratio nevertheless Ieatures inadmissible inIractions against correct
Latinity. What is the violation oI the concord oI number in praesente multis
Ior praesentibus multis but a solecism? Apart Irom the obviously ungram-
matical it is diIIicult to determine exactly what else in this short passage the
author oI the treatise classiIied as barbaric or solecistic.
160
But we know Ior
certain that he saw here Iurther violations oI Latinitas, because he adds the
comment that his passage 'has not achieved the quality that belongs to the
'thin style, viz. a speech composed oI pure and chosen words (Rhet. Her.
4.16, non enim est adeptus id quod habet adtenuata figura, puris et electis
verbis conpositam orationem). II this interesting example were longer, it
would no doubt have presented a rich source Ior the students oI Latin Jolks-
sprache. The treatise even characterizes this way oI speaking as sermo inli-
beralis, 'the language oI the un-Iree. We have moved to the bottom oI the
social ladder, to the barbaric and solecistic language oI slaves and Iormer
slaves.

160
Caplan again points amongst other things to 'the vulgar locution convicium facere, the
abuse oI the demonstrative in istic, hunc, hic, hic, illi, ille, the monotonous transitions, the
awkward parataxis and short sentences, the employment thrice oI post or postea.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 71
The example oI vitiosa oratio or sermo vitiosus in the Rhetorica ad He-
rennium shows that the language oI the Ireedmen is a known category oI
verkehrte Sprache in the theoretical literature. Furthermore, the speeches oI
the Ireedmen at the Cena are not the only demonstrations Encolpius gives oI
the incorrect use oI a stylistic genus. In the opening oI the text oI the Sa-
tvrica as we have it, Encolpius gives us an example oI the reverse oI the
figura gravis or sermo gravis, the highest or most weighty oI the three ge-
nera according to the Rhetorica ad Herennium. Such is Encolpius` example:

haec vulnera pro libertate publica excepi, hunc oculum pro vobis im-
pendi; date mihi ducem qui me ducat ad liberos meos, nam succisi pop-
lites membra non sustinent. (Sat. 1.1)
!
These wounds have I sustained Ior people`s liberty, this eye have I Ior
thee IorIeited. Give me a leader to lead me to my children, Ior my ham-
strung legs cannot uphold my limbs.

The ideal that is deliberately botched here is accurately described by Encol-
pius in accordance with the theory oI styles as grandis oratio, and with equal
scholarly acumen he calls its vicious reverse turgida oratio (Sat. 2.6; cI. 1.2.
tumore), which is the very same terminology that is used by the Rhetorica ad
Herennium (4.15, turget; tumor). Such theoretical precision is surely a sign
that in the Satvrica we are dealing with a text that can proIitably be read in
the context oI the ancient schools oI rhetoric and their theory oI styles.
It is tempting to put our observation about the reverse elegance oI the
prose style oI the Satvrica in context with the exceedingly problematic poet-
ics oI our text. Catherine Connors has recently devoted a book to the study
oI Petronius as a poet and her Iindings throughout the study conIirm the
Ilaws oI the Satvrica`s poetic compositions: 'As everyone will agree, the
short poems on moralizing or erotic themes perIormed by Trimalchio, and
Eumolpus, and Encolpius (both as character and as narrator) represent utterly
conventional habits oI thought.
161
Likewise, 'the obsessive display within
the Troiae Halosis oI repetition, likeness, and imperIect re-enactment signi-
Iies . Eumolpus` lack oI literary control.
162
The same is true oI the longest
verse section oI the Satvrica, the so-called Bellum Civile (Sat. 11924). It
compares poorly with other examples oI the genre: 'Virgil, Lucan or Statius

161
Connors 1998, 50.
162
Connors 1998, 93.
72 1 NARRATI VE
can brilliantly rework inherited motiIs: so Iar as I can tell Eumolpus` poem
oIIers dim, overly studied transIormations oI tradition.
163

Connors persuasively argues that the restraint and elegance oI Calli-
machean esthetics, traditionally expressed in the metaphors oI the untrodden
path and the narrow stream, as opposed to the well-trodden one and a Ilood
oI water, are explicitly rejected by Eumolpus in the theoretical preIace to his
poem on the civil war. This, along with other evidence advanced by Con-
nors, indicates that the pretentious Mr. Sing-Well (the Greek verb 8e-,G893
Irom which the name Eumolpus is Iormed means to 'sing well) is meant to
expose himselI as a bad poet, according to the best contemporary standard,
neo-Callimachean poetics. Now, Connors also recognizes Eumolpus as 'a
Iigure oI metaliterary dimensions, reIlecting Petronius` own enterprise in
craIting the novel, and so she logically concludes by saying: 'Over and
over again, in becoming a poet Petronius acknowledges the limits oI the
poetry he leaves behind.
164

II we try to enter this debate on Connors` premise as iI the aim oI our
inquiry were to tease out oI the text the intentions oI the author in writing it,
we must ask why Petronius would want to write deliberatelv Ilawed poetry?
Surely, we can assume that he was capable oI writing good poetry. Connor
writes: 'To choose a genre, even one as loosely deIined as prose Iiction, is to
reject all the others . by producing verse within his Iictional prose,
Petronius sets his novel in a selI-consciously agonistic relationship with the
literary genres which he has repudiated.
165
As we have seen above the
prosimetric/Menippean Satvricon indeed not only sets prose 'in a selI-
consciously agonistic relationship with the poetry, it uses sermo ad-
tenuatusat last we have a name Ior it!this urban and educated idiom in
which most oI the Satvrica is written, as an ideal medium. And it puts every-
thing else in an 'agonistic relationship with it, also the various examples oI
verkehrte Sprache that we Iind in this text. While the students oI the schools
oI declamation, like those who are trained to extemporize poetry, have been
Ied on a bad diet, and are worse oII than the oldies Sophocles, Euripides,
Pindar, Plato and Demosthenes who never practiced declamation (Sat. 3.3
6), the Ireedmen who have no education are just as badly oII. Neither group
has mastery over the ideal medium oI spoken language, which is the prere-
quisite Ior making sense. It is as iI this whole philosophy oI colloquial lan-
guage was based on the Aristotelian theory oI vicious excesses and a golden

163
Connors 1998, 102.
164
Connors 1998, 143146.
165
Connors 1998, 147.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 73
mean, coupled with a notably conservative preIerence Ior the Greek, the old,
the established and the classical.
At the same time, however, Encolpius uses his demonstrations oI vitiosa
oratio as a vehicle to express critical ideas that are as subversive in thought
as they are mangled in Iorm. Though it may sound slightly absurd, the mis-
shapen and ugly creatures oI artistic processes are sometimes more revealing
oI the nature oI those processes, and the underlying assumptions involved,
than the most successIul and complete works oI art. This is evidently the
reason behind the ancient Iascination with messy and destructive parody. By
creating deliberately bad art, in the manner and style oI a recognizable indi-
vidual or school, the truth hiding behind the Iacade oI beauty with which
polished art is varnished is better exposed. The target can also be merely
hypothetical, but this does not preclude the sense oI recognition which
makes its reception such an interesting and enjoyable experience.
166


Where impersonation in the Satvrica involves inIractions against Latinitas,
at least in comparison with the proper sermo adtenuatus oI the general narra-
tive, this relative diIIerence does not allow us to assign the main narrative to
the category oI Iormal literary style. At the same time we should be wary not
to separate literary language entirely Irom spoken language, since the two
obviously overlap and are Iar Irom being discrete classes. Certain types oI
Iormal 'spoken discourses existed in antiquity, both in prose and meter,
which demanded a vast literary repertoire. In contrast to Encolpius, Lucius in
the Metamorphoses uses a more elevated literary stylenotwithstanding his
pretense to speak as a Ioreigner who Iirst acquired Latin as an adultwhich
puts him above colloquialisms and ungrammatical Latinity. This is why
Lucius and his subordinate personae all speak in an artiIicial and peculiarly
literary manner.
167
The diIIerence seems to correlate with what we have
noted previously, that there is a Iundamental distinction in the mode oI pres-
entation in the Satvrica and the Metamorphoses; one is perIormance litera-
ture while the other presents itselI as a written text to be read aloud by a
solitary reader to himselI.

166
Eumolpus` ecphrastic Troiae Halosis (Sat. 89) has oIten been compared to the tragedies oI
Seneca, and is sometimes analyzed in the context oI Neronian aesthetics. The same charac-
ter`s Bellum Civile (Sat. 11924) is, as a rule, compared to Lucan`s Pharsalia. Agamem-
non`s verses in the Lucilian style, at Sat. 5, have also been related to the satires oI Persius.
For speciIic reIerences to scholarship on this much discussed topic, consult e.g. Connors`
1998 vast bibliography (pp. 14961).
167
For a model oI the narrative speaker-personae in the Metamorphoses oI Apuleius oI the
kind we showed Ior Encolpius and his 'masks, see section 3.1.4 below and Jensson 2002.
74 1 NARRATI VE
It is pertinent to this whole question to consider the common denominator oI
those ancient works oI literature that include linguistic mimicry, whether oI
regional dialects, Ioreign languages or vulgar idioms. The relevant material
was collected by Boyce,
168
who lists such examples as the 'Persian oI the
Persians oI Aeschylus; the crude patois oI the Phrygian or Persian captive in
the Iragment oI the Persians oI Timotheus (dated ca. 419416 B.C.E.); the
imitation oI the Boeotian and Megarian dialects in Aristophanes`
Acharnians; the Laconian in the Lvsistrata; the Thracian gibberish oI the
Birds; the solecistic Greek oI the Scythian archer in the Thesmophoria:usae;
the 'Indian and the ranting monologue oI the vengeIul woman in a Irag-
ment oI subliterary mime (POxv 413, No. 76 |Page|); the general imitation
oI colloquial language in Plautus` plays; the 'Punic oI the Poenulus; and
Iinally the 'vulgar diction oI the Atellan Iarces oI L. Pomponius and No-
vius, as well as the mimes or fabulae riciniatae oI D. Laberius, Publilius
Syrus, and Cn. Matius.
Boyce ends his survey by concluding, 'realistic literary depiction oI the
speech oI commoners (and, we may add, oI barbarians) is conIined to the
comic mode. Some oI his own examples, however, are obviously not com-
edy, but come Irom tragic plays (Aeschylus, Timotheus) and to Boyce`s list
one could add the 'Phrygian in Euripides` Orestes, and the 'Egyptian in
Aeschylus` Suppliants.
169
Instead oI the 'comic mode we preIer to see the
common denominator in all oI these examples as 'theatricality, since in
every instance we are dealing with plays, written Ior perIormance in theaters
and Ior an audience. The exclusive use oI linguistic mimetismo in perIorm-
ance literature would thereIore seem to support our contention that the Sa-
tvrica was composed Ior perIormance.
170

Although direct evidence oI the perIormance oI novelistic narratives is
scarce,
171
some indications are Iound in Lucian`s Pseudologista. The outspo-

168
Boyce 1991, 314.
169
From comedy one could add the 'Ionian in Aristophanes` Peace and the Apulian calyx
crater by the Tarporley painter (New York Metropolitan Museum oI Art, No. 24.97.104). A
reproduction can be Iound in Taplin 1993, plate 10.2.
170
Much the same could be said oI the Iolaos (POxv 3010), and the Tinouphis (PHaun. inv.
400) papyrus Iragments which like the Satvrica contain 'a number oI vulgarisms and uncor-
rected errors in both the prose and the verse sections oI the text (Stephens and Winkler
1995, 367).
171
There is, however, the mention oI aretalogi, tellers oI marvelous survival tales in the style
oI Odysseus` lying Phaeacian tales (cI. Juv. 15.16; Manetho 4.447, 03 B` J%8#",-/+o
$H8="#" G-&'+,` ZN-3#".; and Lucian Merc. Cond. 1I. and J.H. 1.3), as regular entertain-
ers at Augustus` dinner parties (Suet. Aug. 74), who are probably the same as the lectores
and fabulatores oI the household to whom the emperor listened iI he could not sleep (ibid.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 75
ken sophist here impersonates the Iigure oI 'Reproach (?d,8/N-.) to blast
his enemy Ior delivering in Olympia, at a New Year`s Iestival, an 'improvi-
sation which was supposedly plagiarized Irom others and completely
memorized beIorehand. (What seems to have maddened Lucian especially
was that the perIormer managed to turn his heckling against him and roused
the laughter oI the audienceunIairly, perhaps, but successIully.) The im-
provisation in question was a mixture oI prose and poetry (D C3 #13 )4313
03#%Am". 0. A,-. |'his voice switching to song|),
172
and Lucian`s enemy,
who according to him was a great sexual pervert, had Iirst gained his reputa-
tion as a theatrical perIormer by impersonating Ninos and Metiochus.
173
The
juxtaposition oI these two names oI the male protagonists Irom the very ear-
liest Greek novels (probably Iirst cent. B.C.E.)
174
may provide some evi-
dence oI where, how and by whom the ancient novels were recited. Although
neither oI these two works seems to have been prosimetric or to have in-
cluded linguistic mimicry, evidence that even Iictional narratives without
such Ieatures were recited in the theater may indirectly support our belieI
that a work such as the Satvrica, which is theatrical in its narrative style, was
in Iact composed Ior perIormance.
175

It may seem at Iirst that all written texts are equally suited Ior literary
recital in Iront oI an audience, but it is not that simple. Certain texts include
Ieatures that would make perIormance awkward. Consider, Ior example, how
Thucydides` statement that his work was composed Ior posterity and not Ior
a public recital might sound iI read in Iront oI an audience. Another example
oI perIormance-unIriendly Iigures is Lucius` many personal addresses to a
single lector in the Metamorphoses. These would certainly sound awkward iI
they came Irom the mouth oI a public reciter oI the work; the reader, imper-
sonating Lucius, would appear to be talking to another aspect oI himselI (as
reader), while at the same time reciting Ior his audience! Although it is cer-
tainly impossible to write a text which can only be perIormed or only read by
a solitary reader, there are ways to make works either more or less suitable
Ior particular uses. As we have seen, nothing in the Satvrica makes it averse
to perIormance, while we have pointed to several Ieatures that would make it

78). We also have a reIerence to some one telling some 'history or 'Iable in the circus
(D.Chr. or. 20.10, #Q3 BC g*#-%+"3 #&3V > TH-3 B&K/-=83-3), but this last instance may not
have been more than casual entertainment.
172
Lucianus Pseudol. 7.
173
Lucianus Pseudol. 25.
174
See Dihle 1978, 5455.
175
For a discussion oI the 'aural quality oI early Greek novels and their intended 'readership
as not only the educated reader, but also a semi-literate or illiterate 'hearer, see Hgg 1994.
76 1 NARRATI VE
highly suitable Ior a lively recitation, a sort oI one-man act in the theater.
The Satvrica shares some oI the generic premises oI theatrical texts, which
unless making Iun oI their own conventiondo not allude to the Iact that the
words which are spoken on stage have been written previously, or at least
composed and memorized, but seek to create the illusion that the dialogues
on stage are conceived on the spur oI the moment. In likening the Satvrica to
theatrical literature, however, we should be careIul to distinguish between
regular plays, which Ieature many interacting personae, and Encolpius` one-
man recital.
1.2.8 Actor & Auditores
Our conclusion makes it necessary to reconsider a pre-existing hypothesis
about the Satvrica as perIormance literature which was oIIered by the his-
toricist school oI Petronian scholarship. The historicist school, however, did
not base its hypothesis on a close reading oI the Satvrica, but assumed on the
basis oI the historical context that Petronius must have written and recited
the Satvrica Ior an intimate circle oI Iriends at the court oI the Roman em-
peror Nero. Now, we must ask whether our conclusion that the Satvrica was
written Ior perIormance is perhaps a vindication oI the historicist assump-
tion.
The lively gesticulation which is implicit in the text oI the Satvrica; the
mimicry oI the speech mannerisms oI the un-schooled Ireedmen; and the
mixed Iorm oI the discourse, which abruptly switches Irom one discursive
type to another and Iorces the reciting voice to repeatedly change its rhythm
and tone; all this makes it clear that perIorming the Satvrica is a very diIIi-
cult task, iI it is to be done properly, and probably would require the training
oI a proIessional actor. It may thus be argued that the Satvrica is unlikely to
have ever been perIormed by the author.
176
Historically, in the early empire
the writer himselI or a proIessional lector was variously used to perIorm a
text.
177
The practice oI using lectores in aristocratic circles was common.
178

Incidentally, a lector oI Pliny the younger was named Encolpius.
179
In Plu-

176
Arrowsmith 1987, x I., also argued Ior a proIessional perIormer: 'For unless I am badly
mistaken, the Satvricon was clearly written |.| to be recited aloud by a trained artist with a
voice and virtuosity capable oI registering the enormous variety oI the work.
177
Starr 1991, 337I.
178
Starr 1991, 338.
179
Plin. Ep. 8.1.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 77
tarch`s Table-Talk, a Stoic sophist describes a Iorm oI entertainment accord-
ing to him recently brought in at parties in Rome: !

'`*#8 /(%, 8rG83, '6#& #I3 u,(#43-. B&",F/43 B&K/K"#&'-+ #&3A. 8M*&3
-g BC B%""#&'-+l #-=#43 -v3 #I3 B%""#&'I3 #-w. 0,")%-#(#-$.
0'B&B(*'-3#"& G"9B8. S*#` JGQ *#F"#-. ,A/8&3l G%F*8*#& B` ^GF'%&*&.
G%AG-$*" #x ]H8& #I3 ^G-'8&A343 G%-*;G43 '"R )43f. G,(*" '"R
*Nf" '"R B&"HA*8&. EGF83"& #-9. ,8/-A3-&. .

'You are aware, he said, 'that oI the dialogues oI Plato, some are narra-
tive and others dramatic. Slaves are taught the liveliest oI these dramatic
dialogues, so as to recite them Irom the mouth. They use a type oI pres-
entation appropriate to the personalities oI the characters in the text, with
modulation oI voice and gestures and delivery suited to the meaning .

Nothing stops a livelier narrative than Plato`s philosophical narrativeswe
note the careIul distinction between the basic Iorms oI narrative (monologue)
and drama (dialogue) in the passagesuch as the Satvrica Irom being re-
cited by a single slave.
180
The enjoyment oI the perIormance oI the Satvrica
would naturally increase considerably Irom using a lector with the looks and
qualities oI Encolpius to act the narrator as he skillIully rides through his
variegated narrative as a desultory horseman jumps between horses.
The merit oI the Neronian perIormance hypothesis is that it (uncon-
sciously?) recognizes that the Satvrica is perIormance literature. Since we
have made up our minds to identiIy Petronius Arbiter as the consular Iriend
oI Nero, it is conceivable that, like Nero, the author oI the Satvrica may have
had ambitions as an actor. Besides, ancient rhetorical training, which the
author oI the Satvrica obviously underwent, especially the training in deliv-
ery, was easily and consistently conIused with acting. Hence, Petronius
might have perIormed the work himselI in the person oI Encolpius. How-
ever, this historical possibilityIor it is no more than thatiI explored,
merely lands us in a situation similar to the one discussed in section 1.1 in
relation to the Suetonian anecdote about Nero playing Hercules, and the
reaction oI the simple-minded young recruit: iI we recognize the speaker as
Petronius, we rupture the Iabric oI the story oI Encolpius. The identity oI


180
Ancient recitals oI literary texts were certainly not as Iormal as the modern academic lec-
ture or readings oI Iiction or poetry by authors themselves. Jones 1991, 193, traces to
Friedlnder the incorrect view oIten expressed in commentaries and handbooks that private
literary perIormances included only excerpts, or recitals by a single author.
78 1 NARRATI VE
Petronius Arbiter, whoever he was, cannot be accommodated in the trave-
logue and erotic recollections oI Encolpius. In other words, it ultimately
makes no diIIerence to us whether it was Petronius who recited the Satvrica
or an actor, because unlike the author oI the Apocolocvntosis, who speaks in
his own voice,
181
Petronius would not have been identiIiable as such while
acting the persona oI Encolpius. The Iact is that no knowledge oI the Iirst
historical perIormance is available, and the less we assume about this matter
the less likely it is that we inadvertently conIuse the personal attributes oI
Encolpius with those oI the Neronian consular Petronius, a conIusion with
which we are all too Iamiliar in the scholarship. The answer to the question
that I posed above is thereIore purely Iormal: The perIormance model does
not imply any particular Iirst perIormance context, and so does not validate
or invalidate the hypothesis that Petronius wrote and recited the Satvrica Ior
Nero. From a theoretical point oI view the audience oI perIormative litera-
ture is as undetermined as are the actors, since by design the work is 'repeat-
able and can be perIormed again at least as long as it retains the power to
please audiences.
182
The importance oI the Iirst historical perIormance oI the
Satvrica has thereIore been greatly exaggerated.

It is Iar more IruitIul to pay attention to the audience whom Encolpius pre-
tends to be talking to in the text oI the Satvrica. Who are his implied listen-
ers? Nothing in the text indicates that they are a particular audience; rather it
is relatively easy to show that his intended audience is a highly generic one.
As the voice oI Encolpius is scripted in the text, so is an imaginary 'you,
the so-called Ideal Second Person. Whenever Encolpius evokes this imagi-
nary other (7.4, 22.5, 23.5, 31.7, 36.6, 83.2, 127.5, 136.13), he is trying to
establish a community oI opinion between himselI and his projected audi-
ence.
183
II we can make explicit this implicit presence oI the act oI narrating
the Satvrica we can establish the audience on the grounds oI the words oI the
only reliable witness to this scripted setting, Encolpius himselI. By collect-

181
It is not an implausible interpretation oI Dio Cassius` (60.35.3) mention oI the Apocolo-
cvntosis that Seneca recited it to Nero, Agrippina, and his brother Lucius Junius Gallio
about the time oI Claudius` Iuneral.
182
Vogt-Spira 1990, 18390, discusses historical perIormances oI the work, both in antiquity
and among nineteenth-century German philologists. Cocchia 1893, 455, describes the pro-
duction oI the Cena at the Court oI Hannover in 1702 with members oI the royal Iamily
participating as actors.
183
This indeIinite imaginary person takes a subjunctive where usually a deIinite person would
have an indicative. The singular is conventional and should not be taken literally (it was
used in public speeches, e.g., Liv. 31.7.11).
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 79
ing and studying the most signiIicant oI these reIerences we can Iorm a
rough idea oI what kind oI audience the narrator is constructing as he speaks.
At one point he tells us that while hurrying aIter Ascyltos to the room-
ing-house he got lost in the Greek city and ran in circles until, exhausted and
bathed in sweat, he approached an old street vendor whom he somewhat
idiotically asked, mater, numquid scis ubi ego habitem'mother, you
wouldn`t happen to know where I`m lodged? (7.1). To his surprise this
anus urbana responded that she did, but then led him straight to a whore
house and said hic debes habitare'here is where you should live (7.2).
Unexpectedly, he says, he Iound Ascyltos there in the lupanar in a similar
state oI exhaustion and adds, putares ab eadem anicula esse deductum
'you would have thought that the same little old woman had led him there
(7.4). As the narrative progresses Encolpius will soon tell oI how he called
Ascyltos to his Iace muliebris patientiae scortum, cuius ne spiritus purus
est'you whore, submissive as a woman, whose breath is not even clean
(9.6). The implicit audience has already been told everything about how
Ascyltos submitted to Encolpius` sexual advances in 'the garden (9.10,
viridarium). For Encolpius to assume, then, that his audience would have
given the penniless prostitute, Ascyltos, the beneIit oI the doubt regarding
his motives Ior being in the lupanarunlike the anicula and the pater fa-
milias, who took Ascyltos straight to the brothel to have sex with him (8.2
4), both having assumed exactly the contrary about the boysis to assume a
well-bred innocence in the audience, somewhat like that oI the young man in
the passage Irom the Rhetorica ad Herennium quoted above. We may there-
Iore conclude that Encolpius` ideal audience is trustIul and presumably lacks
insight into the machinations oI rascals and swindlers such as Ascyltos.
Again, in relating his discovery that Trimalchio`s slaves all sang while
serving at table, Encolpius adds pantomimi chorum, non patris familae tri-
clinium crederes'you would have thought that it was a chorus oI a panto-
mime, and not the dining-room oI a Iather oI a household (31.7). The audi-
ence oI Encolpius is so thoroughly Iamiliar with the protocol oI behavior in
Roman triclinia, and so unused to a household like the one oI Trimalchio,
that they would not recognize it as such, but think instead that Encolpius was
describing a public theater. The same assumption underlies Encolpius` next
appeal to the imaginary other in regard to a slave who served Iood at the
party, processit statim scissor et ad svmphoniam gesticulatus ita laceravit
obsonium, ut putares essedarium hvdraule cantante pugnare'Iorth came a
carver and moved so perIectly in tune with the music that you would have
thought he was a charioteer Iighting in the arena to the sound oI the water-
80 1 NARRATI VE
organ (36.6).
184
In the mind oI Encolpius` ideal addressees there is a sense
oI decorum that draws a strict line between what is appropriate behavior in a
respectable Roman domestic setting and what is acceptable only as public
entertainment. His audience is not ideally in Iavor oI conIusing these two
contexts.
Encolpius` generic audience is not just morally superior but can also be
relied on to appreciate illusive naturalism in painting and is Iamiliar with the
best known masters oI this old and lost Greek art:
185
The IiIth-century Zeuxis
oI the Attic school, who was said to have painted a boy holding grapes so
realistic that birds Ilew to peck at them, which didn`t please him, however,
Ior he said the boy should have been liIe-like enough to scare the birds away
(Plin. Nat. 35.66); Protogenes, a contemporary oI Aristotle and a Iriend oI
Apelles, who covered his painting oI the Rhodian hero Ialysus with Iour
layers oI paint so as to give it a longer liIe and included a liIe-like dog whose
highly convincing exhalation he was said to have Iinally made perIectly
natural when in Irustration and aIter many tries with the brush he chanced on
the right eIIect with his sponge while removing paint (Plin. Nat. 35.102
3);
186
Apelles, the most celebrated oI all Greek painters, whom Alexander
the Great, through an edict, made his sole portrait painter; Iamous Ior his
invisibly Iine lines, he could paint horses so liIe-like that they were neighed
at by the real animal (Plin. Nat. 35.95). It is in expressing his adoration Ior a
painting oI Apelles, titled Monocnemon ('The Single-Greaved), that En-
colpius adds, tanta enim subtilitate extremitates imaginum erant ad similitu-
dinem praecisae, ut crederes etiam animorum esse picturam'Ior the out-
lines oI the images were cut with such Iinesse, that you would even have
thought that the painting was endowed with the Iorce oI liIe (83.2).
187


184
CI. also 60.1, where Encolpius calls tam elegantes strophas the antics oI a Iood carver who
mimics the madness oI Ajax slaughtering the cattle.
185
88.1, consulere prudentiorem coepi aetates tabularum . simulque causam desidiae prae-
sentis excutere, cum pulcherrimae artes perissent, inter quas pictura ne minimum quidem
sui vestigium reliquisset ('I began to consult the better inIormed man about the age oI the
pictures . and at the same time to inquire into the reason Ior our present decadence, when
the most beautiIul arts have died out ).
186
Sextus Empiricus (Ph. 1.28) recounts a very similar anecdote involving, this time, Apelles`
Irustration and accidental success with the representation in a painting oI a horse`s Ioam.
187
I take animus here to be the Iorce oI liIe, or that quality which distinguishes between the
apparently similar dead eIIigies, including corpses, and actual living people. CI. Trimal-
chio`s words, at Sat. 52.1, et pueri mortui iacent sic ut vivere putes ('and the boys are lying
there dead so |well depicted| that you would think they were alive). The joke is oI course
on Trimalchio, who doesn`t master the language oI contemporary art criticism; his appeal to
the ideal second person indicates that he is emulating the language oI others.
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 81
Encolpius can saIely assume in his ideal audience a knowledge oI these
the most celebrated oI old Greek masters, since in the Iirst century B.C.E.
their works could especially be seen in the temples and squares oI Rome.
Fulvius Nobilior brought the Muses by Zeuxis to Rome. In the Iirst century
there was a Helen by him in the Porticus Philippi, and a Marsyas in the De-
lubrum Concordiae (Plin. Nat. 35.66). At the same time Protogenes` Ialysus
was Iound in the Templum Pacis (Plin. Nat. 35.102). Pliny thinks that a Ia-
mous painting by both Protogenes and Apelles with lines almost invisible to
the eye was destroyed by Iire in the imperial palace on the Palatine. Augu-
stus dedicated Apelles` Venus Anadyomene in the Delubrum Caesaris.
When it deteriorated with age, nobody could repair it, and Nero had it re-
placed with a painting by Dorotheus. Augustus also dedicated two paintings
oI Alexander the Great (one oI which had the king triumphantly riding in a
chariot and the image oI War with hands tied behind its back) in a prominent
location in his Forum Augusti. Later, Claudius had Alexander`s Iace cut out
and replaced by the image oI Augustus.
The ideal audience oI the Satvrica exists in the margin oI the historical
context oI which the Naturalis Historia Iorms a part. The encyclopedia oI
Pliny (2379 C.E.) is appropriately generic, like the rhetorical treatises we
have used above, to provide supplementation Ior the cultural context pro-
jected by Encolpius. Zeuxis, Protogenes and Apelles are properly characters
in the historical context, although they are known to the Iictional world oI
the Satvrica. The conservative Roman taste Ior these ancient masters like-
wise Iorms a part oI the historico-cultural context oI Iirst-century Rome. In
this case we are thereIore directed by Encolpius himselI towards the crossing
oI the boundary between Iiction and history. The simple Iact that the Greek
Encolpius addresses his audience in Latin also shows that his generic audi-
ence is not a universal one, but strictly Roman, and either upper-class or at
least subscribing to the value-system oI the nobler Iamilies. On the whole the
implied audience is supposed to be Iamiliar with literature oI all genres
especially literature in Latin, which Iurther shows this to be a speciIically
Roman audienceand able to discriminate between good and bad examples.
One important Ieature oI Encolpius` audience is the Iact that they are
supposed to be well versed in the Homeric tale oI the wandering Odysseus.
Unlike Trimalchio, they can be relied on to receive correctly esoteric and
parodic allusions to this Iundamental work oI Greco-Roman literature. In the
humorous anagnorismos scene on the ship, Lichas recognizes Encolpius by
his genitals alone and addresses them, not the man, with the words Salve
82 1 NARRATI VE
Encolpi.
188
Immediately aIterwards the narrator adds this philological obser-
vation: miretur nunc aliquis Ulixis nutricem post vicesimum annum cica-
tricem invenisse originis indicem, cum homo prudentissimus confusis omni-
bus corporis indiciorumque lineamentis ad unicum fugitivi argumentum tam
docte pervenerit'let no one be surprised any more that Odysseus` nurse
discovered the scar which revealed his identity aIter twenty years, when a
clever man hit upon the one test oI a runaway so brilliantly, though every
Ieature oI his Iace and body was blurred (105.10). The name oI Encolpius,
which could be rendered into English as Mr. Incrotch ('F,G-. means
bosom`, crotch`, Iold oI garment`; Enkolpios thereIore means he who is
in the 'F,G-.`),
189
appears to be a signiIicant mark oI his identity, and the
narrator oI the Satvrica is thereIore, in some sense, a speaking phallus, or, in
more pedestrian language, a talking prick. His audience, however, who are
expected to know enough Greek to understand the implications oI the name,
is no less respectable Ior that.
Another Homeric allusion comes in the narrative oI his inIatuation with
the young Crotoniate beauty signiIicantly named Circe. At the time, Encol-
pius says he hid under the name oI Polyaenos, an epithet oI Odysseus, and
had Iallen head over heels in love with a prosaic version oI the Iamous en-
chantress, as iI Iollowing in the Iootsteps oI the epic hero. AIter having im-
personated the person oI Circe speaking, Encolpius adds in his own voice,
haec ipsa cum diceret, tanta gratia conciliabat vocem loquentis, tam dulcis
sonus pertemptatum mulcebat ara, ut putares inter auras canere Sirenum
concordiam'even as this woman spoke grace made her words so attrac-
tive, the sweet noise Iell so soItly upon the listening air, that you would have
thought that the symphony oI the Sirens was ringing in the breeze
(127.5).
190
We note that the paintings adored by Encolpiuswhich all had a
similar motiI: an older lover abducting or chasing aIter a young boy, who is
the object oI desirewere assumed to be adorable to his audience as well.
And here this same audience is assumed to be susceptible to seduction by a
Iabulously beautiIul young woman. The audience is accordingly allowed
what seem to have been considered 'normal desires in ancient gentlemen
Ior boys as well as pretty young women.
191


188
The passage makes one wonder what gestures would have accompanied this part oI the
narration.
189
Walsh 1970, 81. On the name, see Maass 1925, 447. Martial uses a similar name, Encolpos,
Ior a boy slave, the sexual partner oI his master (1.31; 5.48).
190
The narrator/perIormer can thus retrospectively be seen as setting a high standard Ior his
own delivery oI 127.12 and 4and prospectively Ior 127.67.
191
CI. Sat. 127.12, Mox digitis gubernantibus vocem 'si non fastidis` inquit |sc. Circe|
1. 2 THE DESULTORY VOI CE OF ENCOLPI US 83
What can we make oI these explicit reIerences to characters Irom the
Odvssev, and the Iact that Encolpius is on more than one occasion associated
directly with Odysseus, who has been called 'the very Iirst explorer-narrator
in the literary record?
192
To an audience assumed to be so Iamiliar with the
text oI Homer and the whole tradition derived Irom the Homeric poem, a
travelogue and narrative oI erotic intrigues told by an unreliable but enter-
taining vagabond would present itselI as a Phaeacian tale, a type oI story so
called aIter the yarn spun by Odysseus at the court oI the noble iI gullible
king Alkinous.
193
In the Iollowing part we shall adopt this ubiquitous cultural
model (I make no speciIic claims that the Satvrica is a parody oI epic or
indeed has any direct intertextual relationship with the Odvssev) as the work-
ing hypothesis in our reconstruction oI the central Iable oI the Satvrica,
which is the story told by Encolpius to his audience oI old-Iashioned Ro-
mans.

'feminam ornatam et hoc primum anno virum expertam, concilio tibi, o iuvenis, sororem.
habes tu quidem et fratrem, neque enim me piguit inquirere, sed quid prohibet et sororem
adoptare? ('She let her Iingers guide her words and said: Young man, iI you do not dislike
a rich woman who has this year known her Iirst man, I shall give you a sister. True, you
have also a brother, Ior I made bold to inquire, but what stops you Irom adopting a sister
too?`)
192
Romm 1992, 183.
193
The ancestry oI this type oI literary Iabrication is obvious, e.g., to Juvenal (15.16) and
Lucian (JH 1.2).
Part 2
Story

2.1 Sorting the Fragments
2.1.1 The Logic oI the Story
The coherence oI the plot can be assumed to be the most important quality oI
the Satvrica, iI it is to be read as an extended Iictional narrative. Because oI
the limited interest shown lately by scholars in this larger aspect oI the work,
little consensus exists as to what was told by Encolpius in the lost early part
oI the story. In my view, the work as we have it cannot well be read without
some idea about the earlier context. Throughout the extant text and in the
Iragments are scattered reIerences to the lost earlier parts, which need to be
elucidated, and not just explained away as meaningless embellishments oI
language. Anyone who has Iaced the Irustration oI students who approach
this work Ior the Iirst time would agree. Not so long ago an American pro-
Iessor teaching a course on the Ancient Novel reported the Iollowing experi-
ence to The Petronian Societv Newsletter: 'A discovery particularly distress-
ing |.| is that my students Iound Petronius to be the hardest text to deal
with, not because oI its matter (though quite a Iew were rather surprised by
what they Iound) but because oI its Iragmentary nature and its generic pecu-
liarities.
194
It is thereIore clearly not mere pedantry to resuscitate the seem-
ingly dead debate on the reconstruction oI the Satvrica`s plot. It seems, in-
deed, Irom the above that we have little to lose, in terms oI readability, iI we
endeavor to alter the current perception oI the work as isolated episodes and
loose Iragments to that oI a more coherent Iictional narrative.
2.1.2 The Erotic Travelogue
Let us begin then. There is good evidence (I shall discuss it below) in the
Iragments to locate the origin oI the story and Encolpius` home city in Mas-
salia. Our hero is a young man and typically needs to leave his hometown in
order Ior his adventures to commence. Such is the invariable beginning oI
ancient Iictions that involve the adventure stories oI young people and teen-
agers. Ninos must leave his home and 'traverse so much land (Fr. A.II) and

194
Relihan 1992, 8.
2 STORY 88
prove himselI as king and leader oI armies beIore he can propose to young
Semiramis. According to Thomas Hgg`s reconstruction oI the lost Metio-
chus and Parthenope,
195
Metiochus leaves the Chersonese and arrives in
Samos where Parthenope is the daughter oI the tyrant Polycrates. From there
on the story becomes an adventure story with wandering, separation and
eventual reunion. In Chariton, Chaereas and Callirhoe Iirst meet at the public
Iestival oI Aphrodite and then get married. Soon aIter he kicks her out oI
jealousy and apparently kills her. The plot starts like a squalid story oI do-
mestic violence, but the adventure Iirst begins when she, still alive in her
tomb, is removed by tomb-robbers and taken on a pirate ship away Irom
Syracuse. The Wonders bevond Thule also opened with Dinias wandering
Irom his Arcadian homeland in search oI inIormation. In Iamblichus` Babv-
loniaca, Sinonis and Rhodanes must Ilee Babylon to escape Irom the men oI
King Garmus, who has Iallen in love with Sinonis. In Xenophon, Habro-
comes and Anthia meet at the local Iestival oI Artemis and Iall desperately in
love. Despite a gloomy oracle they are married and sent abroad on a ship,
which is later captured by Phoenician pirates. In Achilles Tatius, Clitophon
is betrothed to Calligone and would have married her had she not been kid-
napped, and had he not met Leucippe and eloped with her Irom Tyre to Bei-
rut, where they went on board a ship that soon aIter was wrecked in a storm.
Daphnis and Chloe, although they never leave Lesbos, are exposed by their
parents Irom Mytilene, and their adventures involve his kidnapping by pi-
rates and her kidnapping by the Methymnian Ileet. Heliodorus` Aethiopica
tells, in its own elaborate manner, the story oI Theagenes, a young Theban,
and Charikleia, apparently a Delphian maiden, who meet at the Pythian
games in her adopted home-city and Iall in love. They decide to Ilee by sea,
but not unpredictably Iall into the hands oI pirates. Even the Greek Ass-
Story begins with Loukios having just leIt his hometown Patrai Ior Thessaly
on business Ior his Iather. Lucian`s parodic Jera historia, likewise, begins
when the hero has set out one day Irom the Pillars oI Hercules, i.e., he has
leIt the oikoumene or the inhabited world. Alexander, oI course, must leave
Pella so that his bloody adventures in the world can earn him the epithet
'Great. Apollonius king oI Tyre also leaves Tyre and Antioch, and aIter he
has married the daughter oI the king oI Cyrene, they travel by sea and she
gives birth to a daughter and 'dies and he travels in distant parts beIore the
Iamily is reunited and goes back home.
It is thereIore not surprising that Encolpius, too, must leave his home
city. As we have seen, the protagonists oI ancient Iiction leave their home

195
Hgg 1984, 1985.
2. 1 SORTING THE FRAGMENTS 89
Ior various reasons: Ior love, power and knowledge. Encolpius Ialls into the
Iirst category, but his love is not Ior a girl, but Ior a boy. Does this make any
diIIerence? Scholars sometimes assume, apparently Iollowing the German
philologist Richard Heinze (1899), that as a 'homosexual love story the
Satvrica is unique in ancient literature and must somehow be parodying the
'heterosexual so-called Ideal Greek Romances. The evidence does not war-
rant this assumption. A similar plot to that oI the Satvrica, involving two
young men as the loving couple, is Iound in the brieI narrative oI Hippo-
thous to Habrocomes in the Ephesiaca (3.12). It is worth quoting in Iull,
because, as we shall see, it adheres to an established narrative paradigm,
which it has in common with the Satvrica. As a 'comparison text this one
oIIers the closest available parallel with respect to the identities oI the pro-
tagonists and the Iigures oI the plot. We should especially pay attention to
the highly generic circumstances and manner oI introducing the narration oI
this 'novel within the novel:

y"R B1 8e4N-$A343 "e#I3 0*#A3"W83 D qsGGFH--. '"R 0G8B('%$*83l D
BC qbj%-'FK. ]%8#- "e#Q3 #+. z "M#+" #I3 B"'%=43. y"R {. '8/(,"
Z)K '#JV B&K/X"#" '"R G-,,13 ZN-3#" #%"/hB+"3. adBA8#-
qbj%-'FK. 8MG893, ^G&*N3-=83-. '"R #V '"H` "^#Q3 B&K/X*"*H"&. q@ BC
J3","j|3 P34H83 (F3-& BC 0#=/N"3-3 }3#8.) 0WK/89#"& #V '"H` "^#F3.
'ad/| Z)K '8MR #Q /A3-. GF,84. u8%+3H-$ (G,K*+-3 BC #f. ~%'K.
z GF,&.) #I3 #V G%I#" 0'89 B$3"A343l J'-=8&. BC '"R #13 uA%&3H-3 .
Z3B-W-., '"R #-w. P3B%". . 8eB"+-38. 03#"TH". ad'89 3A-. 3 \%(*HK3
8&%"'+-$ '",-Tl k3 BC #Q 8&%('&-3 #I3 0G&N4%+43l }3-" qG8%(3HK.
k3 "e#x. at%(*HK3 BC #V G%I#" 03 /$3"*+-&. B&"G","+-3#" MB|3 '"R
-e' 0'"%#A%K*". qd-%#f. J/-A3K. 0G&N4%+-$ '"R G"33$N+B-. 0G` "e#f.
G%F*8&& #x qG8%(3Ho '"R g'8#8=4 '"#-&'#89%"&l J'-T*"3 BC #Q
8&%('&-3 G(3#" ^G&*N389#"& '"#8,8f*(3 8. y"R #V G%I#( /8 #-T
Z%4#-. DB-&G-%89 )&,X"#" '"R m"=*"#" '"R G-,,V G"%` 0-T B('%$"l
#A,-. BC \B$3XHK83 '"&%-T ,"jF83-& /83A*H"& 8#` J,,X,43 F3-&
'"R #Q #f. z,&'+". |J,,X,-&.| J3=G-G#-3 k3. y"R N%F3h *$3f83 G-,,x,
*#A%/-3#8. J,,X,-$. B&")8%F3#4., [4. B"+43 #&. z93 038A*K*8. y"R
Z%N8#"+ #&. JGQ $U"3#+-$ (G,K*+-3 BC #Q $U(3#&-3 # u8%+3Hh) J31%
#I3 #V G%I#" 0'89 B$3"A343, |{.| 0GR G,-=#h '"R G8%&-$*+ A/"
)%-3I3l ab%&*#F"N-. 0'",89#-. @7#-. 0G&jV. 8eHw. # u8%+3Hh, .
^GF #&3-. JG8*#",A3-. '"#` 0-T H8-T, D% #Q3 qG8%(3HK3 *w3 0-R
'"R 8eHA4. i,+*'8#"&, #-T 8&%"'+-$ H"$(*". #Q '(,,-., G(3#"
D3#&3-T3 0G(/8*H"& B$3(83-3. ad%"*H8R. BC -e'A#& 8#%+4. '"#89N8 #Q3
Z%4#", J,,V #V C3 G%I#" #x 8&%"'+h G%-*AG8G83l . BC J3X3$#-3
2 STORY 90
k3 "e#x (D /V% qG8%(3HK. B&V #13 G%Q. 0C 83-&"3 -eBA3" G%-*+8#-),
G8+H8& #Q3 G"#A%" "e#-T, G-3K%Q3 P3B%" '"R 0,(##-3" N%K(#43l D BC
"e#x B+B4*& #Q3 qG8%(3HK3 G%-)(*8& B&B"*'",+".l Z,8/8 /V% 8r3"&
,F/43 #8N3+#K.. u"%","j|3 BC "e#Q3 #V C3 G%I#" '"#(',8&*#-3 8rN8,
8#V #-T#- BC JGf%83 0. $U(3#&-3. dgGFK3 'J/;, G(3#43
'"#")%-3X*". #I3 0"$#-T, '"R 6*" 0B$3(K3 *$3XK3 #x 8&%"'+hl
0B$3(K3 BC 2,+/", '"+ -& )+,K" *G(3&-3 0/+38#- '"R ,",&V B$*N8%X.l
0)%-$%-=K3 BC ^GQ G-,,I3. 8,8$#"9-3 -e'A#& '"%#8%I3, 0"$#Q3
G"%-W=3". 0G(38&& 8M. uA%&3H-3 '"R G(3#" 6*" k3 -& '#X"#"
JG-BF83-., *$,,AW". P%/$%-3 8M. $U(3#&-3 Z%N-"& '"R ,"j|3
W&)+B&-3 (*$3B-'-T3 #-T#- '"R #x qG8%(3Ho) 8`*8&& 3='#4% 8M. #13
-M'+"3 #-T ab%&*#-(N-$ '"R 8^%+*'4 *$/'"#"'8+83-3 #x G"&BR '"R
2%/f. G,K*H8R. G"+4 #Q3 ab%&*#F"N-3 '"&%+"3. qt*$N+". BC -*K. '"R
G(3#43 J3"G"$-A343 ZW8&& . 8rN-3 ,"H;3, 0G"/F83-. '"R #Q3
qG8%(3HK3, '"R B&` 6,K. 3$'#Q. DB8=*". 8M. uA%&3H-3, 8eHw. 38|.
0G&jV. -eB83Q. 8MBF#-. ZG,8-3 8M. ab*+"3. y"R AN%& A3 #&3-. B&X3$*#-
8e#$NI. D G,-T.l #8,8$#"9-3 BC '"#V <A*j-3 z93 /83-A3-&. 0G+G#8&
G38T" *)-B%Q3 '"R J3"#%AG8& #13 3"T3. yJ/| C3 #x qG8%(3Ho
*$383KNFK3 ^G&|3 "e#x '"R '-$)-#A%"3 #13 3fW&3 0G-&-=K3l 3$'#Q.
BC /83-A3K. -e'A#& 038/'Q3 #Q 8&%('&-3 G"%8+HK #x '-,=jh '"R
JG-H3*'8&. ad/| BC #-*-T#-3 \B$3XHK3 #Q *I" B&"*I*"& 0GR #13 /f3
'"R H(m"&l '"R G-,,V B"'%=*". '"R *#83(W"., J)8,|3 ,8+m"3" '"R
B$3KH8R. 8eG-%f*"+ G-$ E3Q. 0G&#KB8+-$ ,+H-$ *#X,K3 0GA*#K*" #x
#()h '"R 0GA/%"m" 8M. 3XK3 #-T B$*#$N-T. 8&%"'+-$ 0G+/%""
G"%` "e#Q3 0'893-3 #Q3 '"&%Q3 G,"*(83-.l
qsGGFH--. ',8&3x #8TW83 #FB8 *f`~ qG8%(3Ho,
-e #()-3 0' H"3(#-$ J/"HQ3 g8%-9- G-,+#-$l
0. j(H-. 0' /"+K., P3H-. ',$#F3, 63 G-#8 B"+43
%G"*83 03 G8,(/8& 8/(,-$ G38=*"3#-. JX#-$.
-e3#8TH83 BC 8M. C3 uA%&3H-3 0,H893 -e B&A/343, 0#%(GK3 BC B&`
ab*+". 0GR %$/+"3 #13 8/(,K3 '"R u")$,+"3l 'J3#"TH" JG-%+ j+-$
'"R JH$+ #f. *$)-%O. 0GAB4'" 0"$#Q3 ,o*#K%+h. y"R #V C3
G%I#" ^GK%A#K. ,o*#K%+-$ /83F83-., #Q #8,8$#"9-3 |BC| G8%R y&,&'+"3
"e#Q. *$38*#K*(K3 ,o*#X%&-3, 8eB-'&f*"3 0GR G-,=, [4. 0,X)HK*"3
-g *w3 0-R -e G%Q G-,,-T #-T *8 MB893. bY#K C3 z #I3 0I3
B&K/K(#43 #=NKl *w BA, p )+,#"#8, 8MGA -& #V E"$#-Tl Bf,-. /V% 8r
8/(,o #&3R J3(/'o # '"#V #13 G,(3K3 N%;83-..

And it was as they were carousing that Hippothous let out a moan and
began to weep. Habrocomes asked him why he was weeping. 'It`s a long
2. 1 SORTING THE FRAGMENTS 91
story, he replied, 'and a very tragic one. Habrocomes asked him to tell
it and promised to tell his own as well. As they were alone, Hippothous
told his story Irom the beginning.
'I belong, he said, 'to one oI the leading Iamilies oI Perinthus, a
city close to Thrace. And as you are aware, Perinthus is an important city
and its citizens are well to do. There while I was a young man I Iell in
love with a beautiIul youth, also Irom Perinthus, called Hyperanthes. I
Iirst Iell in love with him when I saw his wrestling exploits in the gym-
nasium and I could not contain myselI; during a local Iestival with an all-
night vigil I approached Hyperanthes and begged him to take pity on me.
He listened to me, took pity on me, and promised me everything. And
our Iirst steps in lovemaking were kisses and caresses, while I shed
Iloods oI tears. And at last we were able to take our opportunity to be
alone with each other; we were both the same age, and no one was suspi-
cious. For a long time we were together passionately in love, until some
evil spirit envied us. One oI the leading men in Byzantium (the neighbor-
ing city) arrived in Perinthus: this was Aristomachus, a man proud oI his
wealth and prosperity. The moment he set Ioot in the town, as iI sent
against me by some god, he set eyes on Hyperanthes with me and was
immediately captivated, amazed at the boy`s beauty, which was capable
oI attracting anyone. When he had Iallen in love, he could no longer re-
strain himselI but Iirst made overtures to the young man; when that
brought no result (Ior Hyperanthes would let no one near him because oI
his relationship with me), he won over the boy`s Iather, a villainous man
subservient to money. And he made over Hyperanthes to Aristomachus
on the pretext oI private tuition, Ior he claimed to be a teacher oI rheto-
ric. When he Iirst took the boy over, he kept him under lock and key, and
then took him oII to Byzantium. I Iollowed, ignoring all my own aIIairs,
and kept him company as oIten as I could; but that was seldom, there
were Iew kisses, and he was diIIicult to talk to: too many were watching
me. At length I could hold out no longer. Nerving myselI, I went back to
Perinthus, sold everything I had, got my money together, and went to
Byzantium; I took a sword (Hyperanthes had agreed to this as well),
made my way into Aristomachus` house during the night, and Iound him
lying in bed with the boy. I was enraged and struck him a Iatal blow. All
was quiet, and everyone asleep: I leIt secretly with Hyperanthes without
Iurther ado; traveling all through the night to Perinthus, I at once em-
barked on a ship Ior Asia, unknown to anyone. And Ior a while the voy-
age went well. But a heavy storm struck us oII Lesbos and capsized the
ship. I swam alongside Hyperanthes, gave him support, and made it eas-
2 STORY 92
ier Ior him to swim. But night came on, and the boy could not hold on
any longer, gave up his eIIorts to swim, and died. I was only able to res-
cue his body, bring it to land, and bury it. I wept and wailed proIusely
and removed the relics. I could only provide a single stone to serve as a
memorial on the grave, and inscribed it in memory oI the unIortunate
youth with a makeshiIt epigram.
Hippothous Iashioned this tomb Ior Iar-Iamed Hyperanthes,
A tomb unworthy oI the death oI a sacred citizen,
The Iamous Ilower an evil spirit once snatched
Irom the land into the deep,
On the ocean he snatched him as a great storm wind blew.
AIter this I decided not to return to Perinthus but made my way through
Asia to Phrygia Magna and Pamphylia. And there, since I had no means
oI supporting myselI and was distressed at the tragedy, I took to brigand-
age. At Iirst I was only one oI the rank and Iile but in the end I got to-
gether a band oI my own in Cilicia; it was Iamed Iar and wide, until it
was captured not long beIore I saw you. This, then, is the misIortune I
am telling you about. But you, Habrocomes, my dearest man, tell me
your own story, Ior I am sure that there was some great necessity that
Iorced you to become a wanderer.
196


A stranger`s narrative to a stranger, this story is only told aIter a speciIic
request and a promise to reciprocate with another story. As Glenn Most has
argued persuasively, a stranger`s personal recollection oI his woes in the
Greek narrative tradition is typically not oIIered voluntarily, but is, as it
were, 'wrung Irom his lips by a moment oI overwhelming compulsion.
197

Such a moment is implicit Ior the narrative setting oI the Satvrica as well
and explains the oIten painIul, or at least embarrassed Ieelings experienced
by Encolpius at the recital oI the story to his audience oI good Romans. The
typical symptoms oI the very storied liIe oI Hippothous are his Odyssean
tears and wailing which directly prompt the request Ior his telling his liIe`s
tale.
198
The circumstances oI Odysseus` telling his tale oI woe are so oIten
invoked by later narratives that they cannot count as speciIic reIerence, but
should rather be taken as a generic marker.
199
The same narrative paradigm

196
The translation is based on that oI Anderson, in Reardon 1989, 1478.
197
Most 1989, 127.
198
Alkinous, having observed the guest`s misery, says: 8MGC B` 6 #& ',"+8&. '"R 2B=%8"& Z3B-H&
H$x . 'tell (me) why you weep and lament in your heart . (Od. 8.577), and in re-
sponse to the request Odysseus tells his story.
199
Similar situations with explicit or implicit reIerences to the Homeric paradigm occur, e.g.,
2. 1 SORTING THE FRAGMENTS 93
is used in Sat. 61.16, when Trimalchio asks Niceros, who has remained
silent during the party, to tell a story, and he agrees aIter some apologies,
while the narrator describes his beginning with phrases reminiscent oI Ae-
neas` narrative at the Carthaginian court, which is oI course directly modeled
on the Odyssean paradigm.
200

The tone and mood oI the story oI Hippothous show that it was quite
possible to tell a romantic adventure story oI two boys including envious
divinities; cruel competitors who pose as educators; Iathers who are de-
scribed as 'subservient to money (cI. Sat. 84.3, infra pecuniam); jealousy,
murder, travel, shipwreck, piracy and brigandage, without the 'homosexual-
ity oI the romantic couple in any way undermining the seriousness (iI that is
the proper term) that we are Iamiliar with in the 'heterosexual so-called
Ideal Greek Romances. These stories about young men and their oIten vio-
lent and lawless love aIIairs are at least as old as Thucydides.
201
A Irequent
item in the plot is the killing oI an older and more powerIul competitor; they
also oIten involve embarking on a career oI brigandage and piracy, which
includes living in bands oI desperate young men who are outlaws Irom nor-
mal society. This may have some relation to the sometimes initiatory nature
oI ancient pederasty, and its relation with military camaraderie.
202

Interestingly, Hippothous is a role-player, just like Encolpius; one day he
is a pirate, the next he and his company 'pose as tourists (X.Eph. 4.1).
Role-playing in the Satvrica is a Iunction oI outlawry and the liIe-style oI
vagabonds, Ior the marginal condition oI vagabonds and driIters does not
allow them to speak Irankly or claim their right among strangers directly.
203

The many lies and deceptions oI Odysseus in the last twelve books oI the

in Pl. R. 614b; Verg. Aen. 2.3; Chariton 4.3; X.Eph. 5.1; Juv. 15.16; Ps.-Lucianus Onos 1;
Apul. Met. 1.12; Lucianus JH 1.3, Merc. Cond. 1; Ach. Tat. 1.2, 2.34, 7.34; Longus 2.3;
Hld. 1.8 and 2.21. For reIerences to 'Iirst-person narratives told to strangers in tragedy and
comedy, see Most 1989, 120121. For a discussion oI the Odyssean 'Ich-Erzhlung with
respect to narrative technique in the ancient novel, see Suerbaum 1968.
200
61.5, haec ubi dicta dedit, talem fabulam exorsus est ('These were the words he uttered;
then he embarked upon this tale). The words haec ubi dicta dedit are a Virgilian Iormula
(Verg. A. 2.790, 6.628, 7.323 and 471, 8.541, 10.633, 12.81 and 441), also used by Eu-
molpus in his poem on the civil war, Sat. 121.1 v. 100.
201
See Thucydides (6.54), who relates a story Irom Athenian history; Parthenios (Parth. 7)
who preserves a late Iourth, early third-century B.C.E. account by Phanias oI Eresos; see
also Parth. 24; Ach. Tat. 1.714, 2.34; and two oI Iive short narratives in Plutarch`s Love
stories (Moralia 772d774d).
202
See Sergent 1986, 4054. See also the classic Iormulation oI 'the black hunter / ephebe
complex in Vidal-Naquet 1986.
203
Consider Ascyltos` words 'Who in this place knows us, or who will take our words Ior
anything (Sat. 14.1, 'quis` . 'hoc loco nos novit aut quis habebit dicentibus fidem?`).
2 STORY 94
Odvssev are the ultimate literary model Ior this tradition,
204
as is signaled in
the Satvrica by Encolpius` adopting as a pseudonym an epithet oI the hero,
Polvaenos (u-,="&3-.),
205
'much praised, during the deceiving mimus
(117.4) invented by Eumolpus to trick the legacy hunters oI Croton. The
Homeric scholia include the Iollowing comment on Athena`s strange praise
oI Odysseus` cunning lies to her: 'Travelers abroad are Iorced to lie, since,
being among Ioreigners, they are exposed to harm (Schol. Od. 13.294).
206

Odysseus is the archetypal liar in later Greco-Roman literature (Arist. Po.
1460a; Hor. Ars 151; Juv. 15.16; Lucian JH 1.2I.; Eust. Comm. ad Od.
14.199) and, in Iact, he is universally admired, even in the Homeric poems
themselves, Ior his rhetorical skill (Il. 3.204224) and his guile and deceitIul
tales (Od. 13.287310). His cunning verbal manipulations are generally suc-
cessIul, regardless oI whether there is a grain oI truth in them.
Just like Encolpius, Hippothous does not only chase aIter boys, he also
becomes the object oI women`s desires (X.Eph. 5.9). We must be careIul not
to generalize about ancient sexual personae on the basis oI our own modern
assumptions, Ior there are clearly major diIIerences between our categories
and the ancient ones, as has been shown in recent studies.
207
The apparent
normality oI the relationship oI the boys should not come as a surprise ei-
ther, Ior we have seen that the respectable audience oI the Satvrica considers
both boys and beautiIul young women desirable as sexual partners and does
not seem to rank one higher than the other. Neither does it make the least
diIIerence whether we are dealing with pederasty in a 'Greek or 'Roman
literary work. This dichotomy in the scholarly literature is traditionally so
steeped in nineteenth-century ideologies that it is best leIt completely out oI
the picture.
208
Encolpius` 'homosexuality, thereIore, is not what makes the

204
See Trahman 1952, 3442; Walcott 1977. For Athena, Odysseus assumes the role oI a
Cretan exile who has a Iamily at home (Od. 13.256286); Ior Eumaeus, he is a grateIul
beggar who originates again Irom Crete and is the son oI a rich man and a concubine (Od.
14.199359); Ior Antinous, his background is more condensed but similar, but the account
oI how he got to Ithaca has changed completely (Od. 17.415444); Ior Penelope, he is
Aethon brother oI Idomeneus, Iriend oI Odysseus (Od. 19.165360); and Iinally to his Ia-
ther Laertius, he is Eperitus Irom Alybas (Od. 24.303314).
205
Od. 12.184. The epithet is used by the Sirens, when they address Odysseus, in the Iamous
episode when they try to lure him to wreck his ship (Od. 12.3954; 158200).
206
CI. Eumaeus` words: 'nay, at random, when they have need oI entertainment, do vaga-
bonds lie, and are not minded to speak the truthJ,,` P,,4., '-&Bf. '8N%KA3-&, P3B%8.
J,f#"& / m8=B-3#`, -eB` 0HA,-$*&3 J,KHA" $HX*"*H"& (Od. 14.1245).
207
For an excellent treatment oI the complicated question oI modern homosexuality and an-
cient pederasty, and the various structural paradigms oI ancient sexual relationships, see
Konstan 1994.
208
Encolpius is a Greek, and as I shall argue below, the Satvrica is most likely a Roman adap-
2. 1 SORTING THE FRAGMENTS 95
Satvrica a parodyiI that is what it issince the pederastic paradigm was
open to many uses and variations. Nor is the tragic nature oI the story oI
Hippothous, the death oI the loved one, in any way determined by the gender
oI the couple.
209
Facile critical dichotomies will merely pose obstacles to our
understanding.
The supposed 'homosexuality oI the protagonist was indeed not the
main point oI Richard Heinze`s thesis, that the Satvrica was a parody oI the
'Greek Ideal Romance.
210
Heinze`s condemnation oI Petronius Ior being at
his most 'shameless when he expected his readers to accept Encolpius and
Giton as a romantic couple (Unter den vielen Frechheiten Petrons scheint
mir die frechste die, das er uns :umuthet, als Liebespaar Encolpios und Gi-
ton uns gefallen :u lassen) may have been no more than an obligatory nod to
proper morals, although in recent scholarship the statement is at times as-
sumed to have been the basis oI his parody-thesis.
211
The main achievement
oI Heinze`s thesis, in my view, is not to have shown that the Satvrica was a
parody oI the Greek erotic Iictions, but in Iact to have shown that the work is
just that, an erotic Iictional narrative with a structure and plot organization
comparable to those oI the Iully extant works.
Heinze demonstrated by means oI a close comparative reading oI the
Satvrica and the Greek Iictions that there was a clear schematic analogy
between the Iortune and behavior oI the couple in the Iragmentary Latin text
and the Iully extant Greek erotic Iictions. Encolpius and Giton get their Iair
share oI wandering Irom place to place, and they even experience the generic
storm at sea and shipwreck. Typically, their own beauty or desirability is
their worst enemy, since this attracts many rivals who threaten the integrity
oI their bond. Such external threats, then, naturally lead to outbursts oI jeal-
ousy, as well as instances oI real betrayal, comparable to Clitophon`s and
Daphnis` inIidelities, or in some degree to Callirhoe`s marriage to Diony-
sius. Heinze convincingly supported his sentimental reading oI the work by
reIerence to the Irequent qualiIication in the text itselI oI the boys` relation-
ship as 'a very old relationship, vetustissima consuetudo (80.6), and conjec-

tation oI a Greek model, which was just as prosimetric and comic as the Latin work. I shall
discuss in general the modern 'Roman / 'Greek dichotomy in section 3.2.
209
For no less tragic love stories oI male-Iemale couples, see e.g. the story oI Charite, Tlepo-
lemus and Thrasyllus (Apul. Met. 8.114); see also Parthenios (4, 5, 9, 10, 14, 15, 19, 20,
21, 27, 28, 31, 32, 34, 35, 36); Plutarch, Moralia 771I772c; 774e775b.
210
Heinze 1899.
211
See Bernd EIIe 1987 Ior a criticism oI the assumption that 'homosexuality in the Satvrica
is a parody oI the 'heterosexuality oI the Greek erotic Iictions. Clearly, no such parodic
transIormation was needed in the Greco-Roman world to give rise to a Iictional love story
oI two boys.
2 STORY 96
tured that, unlike Ascyltos and Eumolpus, Giton must have been introduced
at the beginning oI the work and Iollowed Encolpius until the end.
212
He also
notes that Encolpius portrays himselI as being very sentimental, and Ire-
quently bursting into tears over the adversity oI Iortune (24.1, 81.1, 91.4,
99.2, 113.9, 115.12, 134.5), much like the couples oI the Greek works, and
both he and Giton oIten claim to live only Ior their mutual love, and rou-
tinely turn suicidal at the prospect oI losing each other (80.7, 94.8, 114.913,
127.4). Despite the supposedly outrageous 'homosexuality in the story,
Heinze nevertheless recognized that serious erotic stories in Greek prose
literature, Ieaturing two males as the Liebespaar, certainly predate the Sa-
tvrica, and was Iully aware that examples oI this type are Iound in the Greek
erotic novels themselves, such as the Hippothous story above.
213
He claimed,
oI course, that there were important generic diIIerences between the Big Five
extant erotic Iictions and the Satvrica, but let us leave that problem until
later, when we treat the comic narrative stance oI Encolpius and his inIerior
moral status with respect to his audience (see section 3.1.5), and examine
here the evidence Ior the beginning oI Encolpius` story.
2.1.3 Encolpius the Massaliot
Encolpius is certainly oI Greek origin, although his audience is Roman and
the extant text deals with episodes set in Italy, though mostly in Greek com-
munities. The vast majority oI characters in the extant story are Greek, and
we can thereIore with some justice say that the Satvrica is a Greek story,
notwithstanding the language and the audience.
214
The evidence Ior the an-
cient and long independent Greek city oI Massalia as Encolpius` birthplace
comes Irom two Iourth and IiIth-century Iragments, which read side by side
with a Iew passages oI the Satvrica yield this inIormation easily. Servius`
commentary (Aen. 3.57) provides the Iollowing description (Fr. I) culled
Irom the Iull-text Satvrica:


212
He also cites 10.7, iam dudum enim amoliri cupiebam custodem molestum, ut veterem cum
Gitone meo rationem reducerem ('I had been eager Ior some time to get rid oI this annoy-
ing custodian, so that I could resume the old relationship with my Giton); and indications
that Ascyltos is ignorant oI, or pretends not to know about, Encolpius` relationship with Gi-
ton (9.410, 11.3I.).
213
Note, however, that Heinze 1899, 497 n. 3, qualiIies his comparison by saying that this
story and those oI Clinias and Menelaus in Achilles Tatius Iigure 'Ireilich nur episodisch.
214
At least three quarters oI the proper names in the extant Satvrica are purely Greek; cI. Index
personarum in Ernout 1962, 20710.
2. 1 SORTING THE FRAGMENTS 97
auri sacra fames| sacra id est execrabilis. tractus est autem sermo ex
more Gallorum. nam Massilienses quotiens pestilentia laborabant, unus
se ex pauperibus oIIerebat alendus anno integro publicis et purioribus ci-
bis. hic postea ornatus verbenis et vestibus sacris circumducebatur per
totam civitatem cum execrationibus, ut in ipsum reciderent mala totius
civitatis, et sic proiciebatur. hoc autem in Petronio lectum est.

accursed hunger for gold| sacra means accursed. This manner oI speak-
ing derives Irom a custom oI the Gauls, Ior whenever the Massaliots suI-
Iered Irom a pestilence, one oI the poor citizens oIIered himselI to be Ied
Ior a whole year on public and pure Iood. This individual was then
equipped with branches and dressed in sacred attire and led around the
whole city with curses, so that on him would descend the evils oI the
whole city, and thus he was banished. This can be read in Petronius.

Servius is, unsuccessIully, attempting to explain the word sacra in Virgil by
assuming that, since he was a Mantuan and thereIore originally Irom Gallia
Cisalpina, he used the word in a speciIically Gallic sense. Hence the associa-
tion with the Petronian passage which Servius takes to be reliable evidence
Ior religious customs in the Greek city oI Massalia (also in Gallia) in accor-
dance with the grammarian`s practice oI extracting historical and biographi-
cal inIormation Irom literary works. It is evident that Servius` Petronius was
as much a Massaliot in custom and language as Virgil was a Mantuan.
It is oI scant importance to us whether the inIormation thus acquired is
reliable.
215
What matters is that Servius read in Petronius that one oI the poor
citizens oI Massalia, unus ex pauperibus |sc. civibus|, had volunteered to act
the role oI the 'scapegoat ()"%"'F.) in return Ior being Ied Ior a whole
year at public expense, and was then expelled Irom the city when that time
was up. As we learn Irom textbooks on Greek religion, the human scapegoat
is sacriIiced only in a social sense.
216
His treatment is reminiscent oI that oI a
beast marked Ior sacriIice. The beating and cursing oI the )"%"'F. to ward

215
Another commentary, that oI Lactantius Placidus on Statius` Thebais (10.7934), has some-
times been adduced as Iurther evidence Ior the historical truth oI this alleged Gallic custom,
but as Paratore 1933, 1, 152, has shown, it is entirely derived Irom Servius` clause, using
very similar language, and thereIore oIIers no independent evidence. Lactantius Placidus
mentions neither Massalia nor Petronius.
216
Walsh 1970, 73 n.3, seems to mistake the meaning oI proiciebatur and translates 'pushed
oII a cliII. The verb could possibly carry the sense 'to oIIer as a sacriIice, but this is
doubtIul (probably always a corruption oI porricio; see OLD under proicio 3b and 7b.),
whereas the sense 'to drive out (a person) or 'to banish (e.g., proicere in exilium) is well
attested (Ovid, Silius Italicus, Tacitus, Apuleius and Seneca).
2 STORY 98
oII sin, plague or Iamine was no doubt oI importance in actual ancient ritual
(e.g., the beating oI boys in the ritual oI Artemis Ortheia at Sparta), but the
behavior oI the human scapegoat was likely conventionalized and may have
resembled that oI a writhing dancer or an actor in comedy, such as the stu-
pidus oI mime.
217

A relevant ancient account oI such a ritual survives in the poems oI Hip-
ponax (Frs. 511 |West|). In threatening his enemies with destruction Hip-
ponax provides a description oI how the )"%"'F. should be dealt with: A
deIormed and repulsive male is selected and Ieasted on Iigs, barley broth,
and cheese, then whipped with Iig branches and sea onions, and struck seven
times on his membrum virile. Walter Burkert explains how there is a moral
condemnation implicit in the rejection oI this supposedly depraved individ-
ual:
218


It is clearly essential that the creature to be driven out be Iirst brought
into intimate contact with the community, the city; this is the sense oI the
giIts oI Iood which are constantly mentioned. Figs are doubly contrasted
to normal culture, to the Iruits oI the Iield and to the Ilesh oI the victim;
they point to sweetness, luxury, licentiousness, a breath oI a golden age
Irom which reality must be rudely distinguished .; the outcast is then
called the one wiped oII all around, peripsema. There is not active kill-
ing, but simply a matter oI oIIscourings which must be thrown across the
boundaries or over the cliIIs, never to return.

It is easy to see how this episode would Iit into the Satvrica`s plot. Encol-
pius, Ascyltos or Eumolpus are just the types to recklessly exploit such a
situation despite the consequences.
219
Constantly penniless and needy, they
gladly take every opportunity that comes along to get Iood, money and sex.
In the extant Satvrica, religious cults and rituals are generally represented as
pretexts Ior sexual and Iinancial exploitation, and we may accordingly imag-
ine the tone and mood oI the episode as anything but solemn. But most im-
portantly, the possibility that the branches mentioned in the account oI Ser-
vius have something to do with the beating oI the scapegoat on his penis,
and, in any case, the general prominence oI Encolpius` phallus in the extant

217
See Wylie 1994, 489.
218
For Greek scapegoat rituals generally, see Burkert 1985, 8284. The same source is also to
be credited Ior the inIormation about the pharmakos in Hipponax.
219
99.1, 'ego sic semper et ubique vixi, ut ultimam quamque lucem tamquam non redituram
consumerem ('I have always and everywhere lived my liIe as iI I was enjoying the last
light and would never see another day`).
2. 1 SORTING THE FRAGMENTS 99
story, make him exceptionally well suited to play the )"%"'F. in such a
ritual. In the extant text oI the Satvrica the shaving oI Encolpius` and Gi-
ton`s heads, which is then interpreted as sinister Ior the entire ship`s com-
pany (103.5), might be cited as a parallel. The Ieeding and Iattening oI En-
colpius is also an important part oI the Croton episode,
220
where there is
likewise a sense oI imminent danger which spoils the pleasure oI temporary
well-being (125.24).
221
Moreover, his humiliating procession through the
streets oI Massalia has a partial but striking resemblance to the Risus-Iestival
in Apuleius (Met. 3.112), where Lucius is made the butt oI the entire citi-
zenry oI Hypata.
222

As we have seen above, a ritual or a religious Iestival is used in three oI
the Iive Iully extant erotic Iictions to get the plot going (Chariton, Xeno-
phon, Heliodorus). In the Greek cultural context, so preoccupied with the
preservation oI civic cohesion, to be thus cursed or mocked by a whole city,
especially one`s own, is nothing short oI a nightmare and certainly the ulti-
mate in humiliation. Servius says he read about this humiliated Massaliot in
Petronius, and Irom the extant part oI the Satvrica we know that, oI the char-
acters in the story, Encolpius himselI is the most susceptible to humiliation.
Fragment IV, a Iew lines Irom a poetic eulogy oI Sidonius Apollinaris
(Carm. 28.1457), also ties the Satvrica to Massalia:

145 quid vos eloquii canam Latini,
Arpinas, Patavine, Mantuane,
|.|
155 et te Massiliensium per hortos
sacri stipitis, Arbiter, colonum
Hellespontiaco parem Priapo?

why should I sing oI you as sires oI Latin eloquence,
Arpinian, Patavinian, and Mantuan
|.| and you, Arbiter, worshipper oI the holy trunk,
that is Iound throughout the gardens oI Massalia,
yourselI on a par with Priapus oI the Hellespont?`

220
125.2, quotidie magis magisque superfluentibus bonis saginatum corpus impleveram ('each
day I Iilled my stuIIed body as the situation with material goods became more and more
overabundant).
221
125.4, 'nempe rursus fugiendum erit et tandem expugnata paupertas nova mendicitate
revocanda ('no doubt it will be necessary to Ilee again and our poverty, that had at long
last been taken care oI, will again call Ior a new liIe oI begging`).
222
It may be added here in a Iootnote that Fellini incorporated the Risus-Iestival into his cine-
matographic version oI the Satvrica, creating some quite memorable scenes.
2 STORY 100
The late IiIth-century Christian bishop here apostrophizes three Roman liter-
ary worthies (Cicero, Livy and Virgil) by noting only their birthplaces (Ar-
pinum, Patavium, Mantua). He goes on to address others and amongst them
'Arbiter, who is presented as being in Massalia, as iI Petronius Arbiter, the
author, were that character oI the Satvrica whom Servius reIers to in the Iirst
Iragment.
223

Birt rightly rejected Cichorius` interpretation oI the words sacer stipes in
Sidonius` poem as a hollow tree trunk wherein the statue oI Priapus was
placed,
224
but he also rejected Bcheler`s interpretation, unnecessarily in my
view, that the word reIerred to the wooden image oI Priapus. For Birt stipes
was the removable phallic stake (Knttel), which was stuck into the simula-
crum oI Priapus and withdrawn by the vilicus to penetrate the behind oI
thieves who had stolen Irom the garden. A similar usage oI the word stipes,
although not mentioned by Birt, is attested in Seneca (Dial. 6.20.3, alii per
obscena stipitem egerunt), where it is the executioner`s stake Ior impaling
criminals. It is relevant that Priapus is not elsewhere called stipes. Although
Birt`s interpretation seems possible, oI the two stakes, Priapus and his phal-
lus, I am inclined to preIer the god, especially because oI the word sacer,
which properly applies to the god (although metonymy cannot be ruled out
completely). It should, thereIore, be saIe to accept Bcheler`s explanation
that 'sacer stipes est ligneus Priapus. As is pointed out by Birt, the word
colonus was in late Latin associated more broadly with cultus and so in the
poem oI Sidonius it should mean a worshipper or practitioner oI a religion or
virtue, in this case the sacri stipitis which is the objective genitive.
225
Ac-
cordingly, it isn`t Petronius who 'cultivates, as a gardener, the gardens in
Massalia where statues oI Priapus are Iound, but Petronius who cultivates
Priapus whose wooden eIIigies are Iound throughout gardens in Massalia.
226

There is an important diIIerence, because the phrase Massiliensium per hor-
tos does not indicate any movement or action perIormed by Petronius that
could possibly reIer to a lost episode in the Satvrica.
The reductive method oI reading Iictive personal recollections became
the dominant approach oI the ancient grammarians in their commentaries.

223
Bcheler 1862, ad Fr. IV, who says the idea had been adumbrated by Lilius Gyraldus, was
the Iirst to unravel the biographical Iallacy in Sidonius` reading by noting that the poet
'thought that Petronius was the same as Encolpius (ratus uidelicet eundem esse Petronium
atque Encolpium).
224
Birt 1925, 956; Cichorius 1922, 439.
225
This usage oI colonus is attested in Iourth and early IiIth-century Christian writers, i.e., in
the writings oI the immediate predecessors oI Sidonius: Hier. In Is. 54.15; RuI. Clem. 6.2;
Paul. Nol. Carm. 26.333, Ps.-Cypr. Carm. 2.31.
226
A very diIIerent interpretation is advanced by Anderson 1934, 22.
2. 1 SORTING THE FRAGMENTS 101
Maurus Servius above and Aelius Donatus were just such erudite Iigures.
They belong to the Iourth century and are thereIore earlier than Sidonius.
227

(In this tradition the Eclogues oI Virgil, Ior example, were read as the
masked autobiography oI the poet containing important historical testi-
mony.) Similarly, Augustine shows some doubt as to whether he should
believe the truth oI Apuleius` statement 'about himselI (inscribit sibi ipsi
accidisse) in the Metamorphoses that he had been changed into an ass, but he
does not hesitate to apply Lucius` statement to Apuleius himselI.
228
This
merely conIirms what the Christian Saint himselI conIesses to in his work;
as a young man he had gone through the grinding mill oI pagan education in
the classics. Medieval scribes even supplied the praenomen 'Lucius to
Apuleius` name in MSS oI his works.
Now, beIore we go any Iurther in interpreting this Iragment, let us iden-
tiIy its intertextual relationship with the Satvrica. The three relevant lines:

155 et te Massiliensium per hortos
sacri stipitis, Arbiter, colonum
Hellespontiaco parem Priapo?

are clearly modeled on Satvrica 139.2:

14 me quoque per terras, per cani Nereos aequor
Hellespontiaci sequitur gravis ira Priapi.

me, too, through lands, over hoary Nereus` surIace,
haunts the heavy wrath oI Hellespontiac Priapus.`

To my knowledge, this observation has never been made beIore, although
the similarities (underlined) are too strong to be coincidental and Encolpius`
last line is, in turn, a reworking oI Virgil (G. 4.111, Hellespontiaci servet

227
Fulgentius (Fr. VII) oIIers more oI the same when he states that |.| Petronius Arbiter ad
libidinis concitamentum mvrrhinum se poculum bibisse refert ('. Petronius Arbiter says
that he drank a cup oI myrrh to excite his lust).
228
August. C.D. 18.18, . nec tamen in eis mentem fieri bestialem, sed rationalem humanam-
que servari, sicut Apuleius in libris quos Asini Aurei titulo inscripsit sibi ipsi accidisse, ut
accepto veneno humano animo permanente asinus fieret, aut indicavit aut finxit ('. not
that their minds would become beastly, but instead they would retain a rational and human
mind, as Ior example Apuleius either discloses about himselI or deliberately makes up that
aIter being poisoned he was turned into a donkey and yet retained his human conscious-
ness).
2 STORY 102
tutela Priapi).
229
We may thereIore be certainiI certainty is ever attainable
in such argumentsthat Sidonius was practicing what must have seemed to
him a clever biographical reading, reducing the Iiction oI the story by twist-
ing the words oI Encolpius and applying them to Petronius.
It is noteworthy that Rose 1971, 55, and others who have tried to recon-
struct the author`s biography, do not take issue with the implications oI the
ancient biographical tradition that associates Petronius with Massalia, al-
though one would think this to be oI major importance Ior the subject. For
Rose, Massalia (in Fragments I and IV) belongs both to the missing text oI
the Satvrica and is a part oI Petronius` biography, since he conjectured that
Petronius possibly got his education there. The biography oI Petronius is
clearly written on the basis oI the text oI the Satvricaa case oI what I
called Iictionalizing Petronius in section 1.1but the text oI the Satvrica is
also seen to supply the motivation Ior its own production: 'It might be that
|Petronius| was unIavorably impressed by the austere and puritanical reputa-
tion oI the town, and took a literary revenge by making it the scene oI ribald
adventures (Rose loc. cit.).
My interpretation raises a question about the condition oI the Satvrica`s
text in the late IiIth century. Why did Sidonius choose Ior his parody a poem
taken Irom a later book oI the Satvrica (book 16?) that happens to be extant
today? Considering the probable vastness oI the original work, the likelihood
that this should happen accidentally does not seem great. It certainly would
have been more to the point Ior him to parody the beginning, because that is
where Massalia was the scene and where the causes oI Encolpius` wander-
ings were laid out. Or was Sidonius` Satvrica already a reduced version re-
sembling our own?
230
The Iact that Sidonius somehow knew that the narrator
was Irom Massalia might perhaps indicate that he possessed more oI the text
than we do. However, the name and hometown oI an ancient author do not
necessarily come Irom the author`s work. Sidonius himselI was Irom Lu-
gudunum (Lyons) in Gallia, a city closely associated with Massalia through
traIIic on the Rhne, and this might explain his interest in it. We can only
know Ior certain at this point that he had poem 139.2 at hand,
231
and it might

229
A line which had been alluded to beIore by Ovid, Fast. 1.440, Hellespontiaco victima grata
deo. The victima in Ovid`s amusing story was the asellus, another phallic creature. The
lines in the Satvrica, moreover, have Odyssean resonances (Od. 1.14); cI. Cat. 101 and
Verg. Aen. 6.69293, and the discussion oI them by Conte 1986, 3239.
230
See Richardson 1975, 292II., Ior an attempt to understand when and why most oI the text
was lost.
231
Note, however, my reading oI a letter by Sidonius in section 3.2.6.
2. 1 SORTING THE FRAGMENTS 103
very well be the seemingly grand statement Irom the hero, which this par-
ticular poem contains, which drew his attention.
Two things about the meta-text oI Sidonius in relation to the text oI this
particular poem in the Satvrica demonstrate the conIusion oI author and ego-
narrator: what the bishop says about Petronius clearly recalls in Iorm and
content what Encolpius says about himselI, but more obviously the detail
about Petronius` supposed phallic looks betrays without doubt the identity oI
the narrator oI the Satvrica. Just like Lichas, we too may recognize Encol-
pius by his mentula (105.9).
232
The result is a jocular picture oI a Petronius
who stays in his home town Massalia worshipping the sacred stake oI Pria-
pus Iound all over or throughout, per, the gardens oI that city, being himselI
a phallic Iigure on a par with the god.
The humor is oI the type 'send-it-back-to-where-it-came-Irom. In Si-
donius` reading oI the poem oI Encolpius it is now Arbiter who is the
speaker oI the lines, and Arbiter is a Massaliot, and so the bishop deIlates
Encolpius` Iabulous hyperbole, per terras, per cani Nereos aequor, by rede-
Iining the speaker`s relationship with the grotesque pagan deity, and setting
it in the proper biographical ambiance, Massiliensium per hortos. The impli-
cation is that the only dealings Petronius had with Priapus were in the gar-
dens oI his hometown. As Ior epic wanderings and persecution at the hand oI
a deity, in this he was merely spinning a yarn. That Sidonius is treating his
catalogue oI eloquent Roman writers in a playIul manner is clear Irom what
he has just said about Tacitus (Carm. 28.153, et qui pro ingenio fluente nulli,
Corneli Tacite, es tacendus ori). The catalogue`s Iunction in the poem, a
eulogy to Consentius, also makes this appropriate, Ior the literary worthies
are listed in a praeteritio simply in order to be put down in comparison with
the eulogized addressee.
2.1.4 Priapus
However, we must avoid giving in to interpretive delirium at this point, be-
cause Sidonius says nothing about any 'crime or 'wrath oI Priapus. These
ideas are Iound only in Encolpius` poem (139.2, gravis ira Priapi |'the

232
CI. the explanation oI Bcheler 1862, ad Fr. IV: par Priapo Arbiter uocatur quia Encolpius
nilo deterius mutoniatus quam Mutunus tot tantasque res mentula duce gesserat, quibus
etiam hodie quae extant chartulae refertae sunt ('Arbiter is said to be on a par with Priapus,
because Encolpius who is no worse equipped than the |phallic deity| Mutinus commits so
many and such great deeds under the leadership oI his penis as even the Iew sheets oI the
story that still exist today bear so copious a witness).
2 STORY 104
heavy wrath oI Priapus|) where they are best read in context with the hero`s
own conjecture that he was poisoned (128.2, veneficio tactus sum), his pri-
vate parts, he thinks, put to sleep by sorcery/poisoning (138.7, partes venefi-
cio, credo, sopitae) and unmanned by an 'angry hand, manus irata (140.12;
cI. 139.2, gravis ira Priapi), as a punishment Ior an oIIense he committed
against Priapus out oI poverty and not with all oI his body (133.3, inops et
rebus egenis attritus / facinus non toto corpore feci).
233
In earlier scholarship
it was possible to expand the 'wrath oI Priapus into an overarching epic
structure by reading Sidonius` poem in conjunction with Sat. 139.2, as iI the
two poems were one continuous context.
234
Such contextual Iusion can have
disastrous consequences Ior interpretation. Sidonius merely says that Encol-
pius (alias Arbiter) is Irom Massalia, and that he is the 'equal oI the garden
deity Priapus. In other words, the poem oI Sidonius does not provide inde-
pendent evidence that the gravis ira Priapi was a uniIying motiI in the Sa-
tvrica Irom the beginning oI the plot in Massalia, contrary to what has oIten
been asserted.
235

This conjecture is not called Ior and in Iact it creates obstacles when we
try to reconstruct the opening oI the plot, because in consequence a certain
over-determination oI Iactual causes Ior Encolpius` leaving his home city
occurs: he is exiled as a scapegoat, and Iorced to leave the city because oI

233
Sat. 133.3 is discussed in detail below in section 2.2.7.
234
This is originally an idea oI Elimar Klebs, though he attributes it to Bcheler. Bcheler`s
1862 comment on the text runs as Iollows: re vera denotasse mihi illis uerbis Sidonius uide-
tur ea quae in satiris Petronius narrauerat, ratus uidelicet eundem esse Petronium atque
Encolpium qui primas harum fabularum partes agebat. sacer stipes est ligneus Priapus,
sacri stipitis per hortos Massiliensium colonum dicit eum qui sacra Priapi apud Massilien-
ses coluit, aut fortasse respiciens ad rem singularem relatam a Petronio eum qui aliquando
in Massiliensibus hortis dedicauit Priapum (ad Fr. IV) ('in Iact, it seems to me that with
these words Sidonius has pointed to material told by Petronius in his Satires, no doubt un-
der the impression that Petronius was Encolpius, the main character oI these tales. The hal-
lowed tree trunk is the wooden Priapus, and he calls him a worshipper oI the hallowed tree
trunk throughout the gardens oI Massalia, he who practiced the sacred rites oI Priapus
among the Massaliots, or perhaps having in mind a speciIic incident told by Petronius, he
who at some time dedicated a Priapus in gardens oI Massalia). What is put Iorward by
Bcheler as a possible (aut fortasse) episode or incident in the story (res singularis) be-
comes in the words oI Klebs 1889, 623, much more deIinite: Bcheler :uerst aufmerksam
machte, da Priapus im Roman eine bedeutende Rolle spielte. The supposed Priapic hap-
penings in Massalia were Iurther elaborated by Cichorius 1922, 438442, in an untenable
interpretation oI the Iragment.
235
This view seems to be making a come-back in the scholarship; see Schmeling 1994/5, 213,
who adds the detail that Encolpius was struck impotent by Priapus already in Massalia. But
iI he has been impotent all along, how can we explain his surprise at Iinding himselI unable
to get an erection in Croton?
2. 1 SORTING THE FRAGMENTS 105
some facinus he committed there against Priapus. The interpretation also
makes too much oI the anger oI Priapus, which is likely to be no more than
the subjective understanding oI Encolpius, who in the episode where it oc-
curs is desperately trying to explain and remedy his impotence. The old the-
sis oI the wrath oI Priapus was indeed never Iully accepted, since it tended
not to solve but to complicate the problem, and since the arguments it relied
on were tenuous in the Iirst place. UnIortunately, it eventually provoked
hyper-skeptical responses which, so to speak, threw the baby out with the
bath water.
236
For even iI there is no evidence Ior a Priapic episode in Mas-
salia, it does not Iollow that a comic conception oI the 'wrath oI Priapus
did not play an important role in the Croton episode which may have con-
nected it with an earlier episode.
The earliest and only incident that Iits the description oI a crime against
Priapus belongs to Encolpius` dealings with Quartilla`s cult oI Priapus (dis-
cussed Iurther in section 2.1.9).
237
In that episode, moreover, we have an
explicit facinus against the god (20. 1, facinus, and 133.3 v. 9, facinus),
committed in a temple, out oI poverty, and very likely with the help oI En-
colpius` over-sized mentula, which made him able to impersonate Priapus
himselI, and so would Iit his own description oI the 'crime (133.3 vv. 7
10). In Iact, the trick played on the cult oI Quartilla beIore the Crvpta con-
Iorms nicely to Encolpius` retrospective speculations, which reIer to a previ-
ously narrated episode, although the presumed causal link between that past
crime and his present impotent condition is based on suspicious evidence to
say the least.
238
The epic description oI his wandering, 'through lands, over
hoary Nereus` surIaceper terras, per cani Nereos aequor (139.2 v. 14),
which is spoken in Croton, is accurate enough iI taken to allude to the pro-
tagonist`s travels since the incidents in the urbs Graeca. He has voyaged by
sea Irom the harbor oI the city to the gulI oI Tarentum, and thence moved on
Ioot to Croton (the plural in terras need only be parodic hyperbole to en-

236
See especially Baldwin 1973, 29496; Slater 1990, 40. On the century-old debate, see
Klebs 1889, 62335; Cichorius 1922, 43842; Perry 1925, 31I.; Birt 1925, 956; Anderson
1934, 20; Bagnani 1956, 237; Pack 1960, 312; Courtney 1962, 956; Sullivan 1968, 40
et passim; Mulroy 1970, 2546; Walsh 1970, 73, 77; Rankin 1971, 5267; and McDermott
1983, 8285.
237
The killing oI the sacred goose, 136.4137.12, occurs aIter Encolpius Iinds himselI impo-
tent and also aIter he 'ascertains that the cause oI it lies in the crime he committed against
Priapus. In any case he quickly expiates the crime with a payment oI two gold pieces.
238
The 'evidence is Giton`s statement (133.2) that Ascyltos, too, did not perIorm sexually on
that night aIter the dinner at Trimalchio`s, which was his Iirst opportunity to have sex aIter
the eIIects oI the aphrodisiac they drank at Quartilla`s had worn oII. But Giton, who has
good reasons to Iear Encolpius` jealousy, is a most unreliable witness on this account.
2 STORY 106
hance the vaunted similarities with Odysseus and other oldies). The point oI
Encolpius` poem is not that he has been Ileeing Priapus over sea and land.
Not even Odysseus, his model, was Ileeing Poseidon (the god was merely
delaying the completion oI the hero`s nostos). Rather, Encolpius seems sur-
prised that the punishment Ior a crime he committed in the Greek city should
hit him so Iar away in Croton. This he comically takes as prooI oI the might
oI the deity he thinks he is up against, and on the basis oI that Iantastic idea
he Iancies himselI a hero, much needing to assert himselI psychologically in
a moment oI physical weakness.
But what iI Priapus nevertheless Ieatured in the opening episode in Mas-
salia? What could his Iunction have been? As Heinze showed over a century
ago, in response to Klebs` thesis about the 'wrath oI Priapus, the role oI
Priapus in the Satvrica does not Iollow the conventions oI epic, where gods
come down to earth and meddle directly in human aIIairs, but resembles the
more distant and mystical role oI the gods in the Iully extant ancient erotic
Iictions.
239
The Satvrica`s divine apparatus Iurther resembles that oI the
other stories in the Irequent reIerences made by the protagonists to the
mostly hostile Iorce oI Fortuna (Sat. 13.1, 13.4, 82.6, 100.3, 101.1, 114.8,
125.2) or =NK.
240
These Irequent reIerences to the vicissitudes oI Iortune
may in Iact be a generic marker oI this kind oI narrative, as Karl Brger had
argued even beIore Heinze,
241
relying on Cicero`s inclusion oI them as a
deIining characteristic.
242

In ancient erotic Iictions the gods do not interIere in the action directly in
the epic manner, but they are there in the background oIten Ior reasons oI
divine envy, and may provide theological explanations Ior the misIortune or
salvation oI the protagonists. In Hippothous` story above, the boys love each
other greatly until an evil spirit gets envious and grudges them their happi-
ness ('"R N%F3h *$3f83 G-,,x, *#A%/-3#8. J,,X,-$. B&")8%F3#4., [4.
B"+43 #&. z93 038A*K*8). In Chariton, beautiIul Callirhoe is a devout
worshipper oI Aphrodite, and so it is Chaereas` untimely jealousy which

239
Heinze 1899, 5012.
240
Heinze 1899, 502, also mentions the possible use oI a Ioreshadowing oracle in the original
Satvrica (Fr. XXXVII).
241
Brger 1892, 349.
242
Inv. 1.19, Hoc in genere narrationis multa debet inesse festivitas confecta ex rerum varie-
tate, animorum dissimilitudine, gravitate, levitate, spe, metu, suspicione, desiderio, dissimu-
latione, errore, misericordia, Iortunae commutatione, insperato incommodo, subita laetitia,
iucundo exitu rerum ('In this Iorm oI narrative there should be great liveliness, resulting
Irom variety oI events, contrast oI characters, severity, levity, hope, Iear, suspicion, desire,
deception, error, compassion, change oI Iortune, unexpected trouble, sudden joy, and a
happy ending).
2. 1 SORTING THE FRAGMENTS 107
provokes the goddess` anger (8.1, 2%/&*H89*" N",8GI. B&V #13 P'"&%-3
UK,-#$G+"3). In Xenophon oI Ephesus, Habrocomes arrogantly claims that
he is more handsome and powerIul than Eros, which oI course makes the
god Iurious (K3& G%Q. #"T#" D ?d%4.). The most direct involvement oI
these gods is when they appear in a character`s dream and provide inIorma-
tion that will inIluence the course oI events, as in Longus, when the Nymphs
appear to Daphnis and tell him where he will Iind a purse with three thou-
sand drachmas (3.27), or in the Metamorphoses, when Isis appears to Lucius
the Ass in a dream and instructs him as to where he may Iind the antidote
that will release him Irom the spell and return him to human Iorm,
243
or in
the Satvrica, when Lichas and Tryphaena dream respectively oI Priapus and
the eIIigy oI Neptune, who inIorm them that the boys have been led back to
the ship and will be Iound there (104.12). Other dream epiphanies could be
mentioned that do not provide such detailed inIormation about later events in
the story, but will nevertheless 'come true in some sense.
244
God-sent
dreams, as well as divine oracles and utterances by priests in trance,
245
con-
Iorm to normal ancient religious experience and practice. The manner oI
divine interIerence in ancient Iiction could be described as a Iunction oI the
cosmological status oI the characters, viz. they exist in a post-heroic age
when gods no longer mingle with mortals.
Since only such indirect involvement by the gods is allowed in the
genre,
246
Priapus can have entered the plot as early as Massalia no more di-

243
'I had scarcely closed my eyes (11.3, necdum satis coniuveram), he says, just beIore the
vision, and aIter it, I 'was quickly released Irom sleep (11.7, somno protinus absolutus).
244
Such as Quartilla`s dream inquiry in the temple (17.7, medicinam somno petii, iussaque
sum vos perquirere atque impetum morbi monstrata subtilitate lenire |'I sought the remedy
in a dream, and was told to seek you out and relieve the onset oI the illness by a precise rit-
ual that was revealed to me|); the dream oI Apollonius where an 'angel instructs him to
go to Ephesus where he will Iind his 'dead wiIe (48). Comparable is Osiris` appearance to
Lucius in a god-sent dream (11.29, divinum somnium) to instruct him about Iurther initia-
tion; and in Achilles Tatius, when Artemis appears to Leucippe in a dream and assures her
that she will preserve her virginity (4.1).
245
E.g. in Xenophon oI Ephesus, when an oracle oI the temple oI Apollo in Colophon at the
beginning Ioretells some elements oI the story (1.6), or when the children outside the tem-
ple oI Apis in Memphis assure Anthia that she will recover her husband Habrocomes (5.4),
or in the Metamorphoses, when the priest oI Isis prophesies to Lucius in a trance (11.14,
vultu . inhumano; 11.16, vaticinatus sacerdos).
246
Although Quartilla`s statement in 17.5, nostra regio tam praesentibus plena est numinibus
ut facilius possis deum quam hominem invenire ('our region is so Iull oI divine presence
that you are more likely to run into a god than a man), is sometimes used to suggest the
presence oI the supernatural in the Satvrica, the words are entirely subjective to the priest-
ess, and probably serve the purpose oI excusing how easily she mistook Encolpius Ior Pria-
2 STORY 108
rectly than by grudging Encolpius and Giton their erotic pleasures, or, and
perhaps more likely, because Encolpius was his equal with respect to the size
oI his mentula and may have inadvertently entered into competition with the
god Ior the attention oI worshippers. Nothing provokes divine anger like the
impersonation oI a god by a mortal, which is really what is implied by Si-
donius, when he apostrophizes Petronius and calls him, or rather Encolpius,
Hellespontiaco parem Priapo. II I am right in conjecturing that Encolpius
impersonated Priapus when he disturbed the nocturnal ceremonies oI Quar-
tilla and her Priapic cult, this would either be a repetition oI an earlier motiI
or the only cause oI the 'wrath oI Priapus. No direct epic-style conIronta-
tion or facinus is thereIore necessary or even possible and we can let the
inIormation Irom this Iragment (Fr. I) suIIice as an explanation oI how and
why Encolpius leIt his home city oI Massalia.
2.1.5 Encolpius, Scapegoat and Exile
So Iar our reading oI the two Iragments oI Servius and Sidonius has yielded
inIormation about Encolpius` citizenship, poverty, voluntary assumption oI
the degrading role oI scapegoat, and Iinal expulsion Irom Massalia. II this
inIormation is right, we would expect some oI it to be reIlected in what En-
colpius says about himselI in the Iragments oI his narrative that have come
down to us. Two passages in the extant text oI the Satvrica Iall into place as
soon as we accept this inIormation. Firstly, Encolpius reIers to himselI as
exul (81.3) in a retrospective soliloquy at a moment oI disillusion when he
has no reason to misrepresent himselI to the original audience/reader, who
already knows the Iacts Irom hearing/reading the story; and secondly, Lichas
reIers to him directly with the Greek word Ior scapegoat (pharmace): 'What
do you have to say Ior yourselI, you thieI? What stray salamander has burnt
oII your eyebrows? To what god have you oIIered your hair? Answer me,
you scapegoat! (107, 15, quid dicis tu latro? quae sola salamandra super-
cilia tua exussit? cui deo crinem vovisti? pharmace, responde'). These retro-
spective reIerences to the protagonist in the extant Satvrica match so per-
Iectly the Iragment oI Serviusin both Encolpius is an exile and a scape-
goatthat their appositeness is most unlikely to be merely coincidental. As
a result we have no choice but to accept as genuine the extant inIormation
about Encolpius in Massalia at the beginning oI the Iull-text Satvrica.
Let us Iirst look closer at the reIerence to Encolpius` scapegoatery and
then move to the question oI his exile. That pharmace should be considered

pus himselI.
2. 1 SORTING THE FRAGMENTS 109
Greek, transcribed with Latin letters, is proven by the Iact that it occurs only
here in extant Latin literature, so Iar as I have been able to ascertain. As Har-
low has shown, pharmace is correctly read as the Greek vocative )"%"'A,
'scapegoat.
247
The word belongs to the vocabulary oI Greek satiric and
comic authors such as Hipponax and Aristophanes and is used as a term oI
abuse, and so it might possibly occur here without a reIerence to anything
speciIic.
248
However, the other two items in the same address do have reIer-
ences to speciIic Iacts about Encolpius: he has stolen things Irom the ship
and he has lost his eyebrows. The Iorce oI Lichas` question above is not that
he himselI believes that 'a stray salamander leaped Irom the sea aboard the
ship and burnt oII his eyebrows, but that he is mockingly anticipating some
such Iar-Ietched explanation Irom Encolpius.
249
By rounding oII his attack
by nastily reminding Encolpius oI the humiliation he underwent in Massalia
as a 'scapegoat, Lichas delivers a serious blow to the ego oI our hero. Sig-
niIicantly, Encolpius the narrator immediately acknowledges the truth oI
Lichas` accusations: 'and I couldn`t invent anything to say against this accu-
sation oI a most obvious guilt (108.1, nec quid in re manifestissima dicerem
inveniebam).
Let us now examine the description oI Encolpius as an exile (81.3) and
the signiIicance oI this Ior the story. Besides Encolpius, there are two other
exules in the story. Tryphaena calls Giton an exul (100.4), and she herselI is
so reIerred to (100.7, exulem) by Eumolpus, when addressing Encolpius and
Giton who would certainly know the Iacts about her exile.
250
It should be

247
Harlow 1974, 377.
248
E.g. Hippon. Fr. 7 |West|, et passim; Ar. Ra. 733. The early commentator Janus Souza read
pharmace as the vocative oI )"%"'F. (Burman 1743, 2:38). LSJ (s.v.) derives the abusive
sense oI )"%"'F., scapegoat`, Irom the Iact that criminals could be used as scapegoats.
Strangely, however, translators oI the Satvrica have usually taken pharmace here Ior an-
other Greek word )(%"'-. (on the accent see the grammarian Herodianus 1.150 |Lentz|)
meaning sorcerer` and translated it as empoisonneur` (Ernout), GiItmischer, Zauberer`
(SteIenelli), poisonous Iellow` (Heseltine), poisonous creature` (Sullivan), snake in the
grass` (Branham and Kinney). The word is Iound e.g. in the vocabulary oI the Greek LXX.
The problem with this reading is that we have no reason to suspect Encolpius oI magical
practices.
249
A marine animal similar to the salamander, possibly some sort oI 'mollusc, is said by
Pliny (Nat. 10.188) to emit a substance with depilatory eIIects.
250
Encolpius at one point claims that Ascyltos was 'by his own admission worthy oI exile
(81.4, sua quoque confessione dignus exilio), which could possibly indicate that there was a
Iourth exile in the story. The editio Pithoeana has exito, but it is not supported by other wit-
nesses, and shortly beIore Encolpius has spoken oI Ascyltos and himselI as having experi-
enced similar Iortune (80.8). But even iI Ascyltos is an exile he is unlikely to originate Irom
Massalia, because he was clearly not on the ship oI Lichas with the others. He is not a
protagonist and both enters the story and disappears Irom it in Campania.
2 STORY 110
noted that the words exilium and exul were not used lightly in the Latin lan-
guage and rarely in a transIerred sense and then only oI inanimate things and
animals. Moreover, the terms are without abusive connotations (as opposed
to fugitivus, 'runaway, 'Iugitive) since they usually involve people oI
some rank and standing. An exilium is either a legal banishment (the legal
terms are expulsio, eiectio, aquae et ignis interdictio, deportatio and relega-
tio), or a voluntary emigration (demigratio, fuga, peregrinatio). There is
always in these terms an implicit contrast to patria and domus. For these
three Greek characters in our story to be called exules in Campania and
thereabouts proves that they are not Roman citizens, but come Irom an inde-
pendent city outside Roman territory. That city is most likely as Greek as
they are themselves.
The best way to explain the institution oI exile in the Roman world is to
consider it in the light oI legal arrangements between independent states. An
exiled Roman citizen could through the ius exulandi, 'the right to live in
exile, adopt a new patria and thus IorIeit his Roman citizenship.
251
This
arrangement was reciprocal and exules Irom independent cities which had a
foedus with Rome could take up citizenship there and thus relinquish their
previous status at home (Cic. de Orat. 1.177). In early times the exiled Ro-
man did not need to go Iar into exile and could Iind a new home without
leaving Latium, in cities such as Tibur, Praeneste, Lavinium and Ardea, or
he could go to the Latin colonies. In later times Tarquinii, Nuceria and Ra-
venna would serve the same purposes. But when the ager Romanus had been
expanded so as to cover the whole Italian peninsula and especially aIter the
civil wars, when all Italian cities had been granted Roman citizenship, such
places had to be sought outside Italy, in Gallia, Greece or Asia (Cic. Mur.
89). In the early principate the closest Ioreign city to the North and West,
and one that was preeminently qualiIied to accept Roman exiles, happened to
be Massalia. This independent Greek city-state, lying in the middle oI the
Roman province oI Gallia Narbonensis, had had a politically privileged
status in the area ever since the war against the Gallic tribes in 123121
B.C.E.
252
In Roman sources, moreover, it is oIten mentioned as the preIerred
destination oI Roman exules.

tagonist and both enters the story and disappears Irom it in Campania.
251
See KleinIeller 1958, 16831685.
252
Strabo has a chapter on Massalia (Str. 4.1.4I.). In the early principate Massalia was still an
oIIicially independent Greek city-state which laid great store by its ancient customs and
citizenship and had a long standing relationship oI amicitia with the Romans. In Strabo`s
time (ca. 63 B.C.E. 19 C.E.) the city had a high reputation Ior its rhetorical and philoso-
phical schools which attracted upper class Romans. Tacitus corroborates this reputation oI
2. 1 SORTING THE FRAGMENTS 111
Even beIore the civil war, in 70 B.C.E., the corrupt Iormer governor oI
Sicily, C. Verres, chose Massalia as his city oI exile and took there much
wealth. In 63 B.C.E. it seemed the obvious place to go to Ior Catilina, iI he
had chosen exile (Sal. Cat. 34.2). Milo, too, went there in 52 B.C.E. (Asc.
Mil. 32.13; 45.23), became a citizen and despite his discontent could joke
that he was happy to be in exile because oI the excellent mullet oI Massalia
(D.C. 40.54). AIter the execution oI Jullus Antonius by the order oI Augu-
stus, in relation to the adultery oI Julia (2 B.C.E.), his adolescent son Lucius
Antonius was sent to Massalia 'where his exile would be hidden by the pre-
tense oI study (Tac. Ann. 4.44, ubi specie studiorum nomen exilii tegere-
tur). Tacitus reports an interesting embassy to Tiberius in 25 C.E. under-
taken by the Massaliots to ask Ior the legitimation oI the testament oI a cer-
tain Vulcancius Moschus, who had leIt his property to the city ut patriae
(Ann. 4.43.5). This well known rhetor (Sen. Suas. 1.2; Con. 2.3.4 et passim)
was born in Pergamum (Porphyrion De Hor. ep. 1.4.9), but had to Iace
charges oI murder by poison and thereIore leIt Pergamum, despite his de-
Iense by Asinius Pollio (Sen. Con. 2.5.13) and Torquatus, Horace`s Iriend
(Ep. 1.5.9). Later he had settled in Massalia as a rhetor. The Massaliots
brought the case beIore Tiberius to test the validity oI the ius exulandi in
Massalia, which was thus reaIIirmed. Seneca wrote to Nero about a Iather
who had shown his clemency to a son who had made an attempt on his liIe:
'satisIying himselI with exileand a luxurious exilehe detained the parri-
cide at Massalia and gave him the same liberal allowance that he had beIore
(Cl. 1.15.2). Finally, in 58 C.E., Nero on Ialse charges bade Cornelius Sulla
leave Rome and stay within the walls oI Massalia (Tac. Ann. 13.47.3). These
walls had been torn down by Caesar in 49 B.C.E., but were reconstructed by
the wealthy Massaliot doctor, Crinas, with Nero`s permission (Plin. Nat.
29.9).
Given the reciprocity oI the institution oI exilium, the Irequency with
which the Romans themselves chose Massalia as their place oI exile makes
this city the most probable, iI not the only possible, place oI origin oI our
Iirst-century Greek exules on board a Tarentine ship heading south along the
west coast oI the Italian peninsula. Since we know that Encolpius is a Mas-
saliot, and we may assume that he leIt the city by sea on the ship oI Lichas, a
merchant who would have had commercial reasons Ior going to Massalia,
the conclusion is hard to resist that Giton and Tryphaena originate Irom
Massalia, are likewise exiles and were also on that ship. The great complex-
ity oI the relationships oI Encolpius, Tryphaena, Giton, Lichas and his wiIe

the Massaliots Ior rhetorical and philosophical skills (Ann. 4.44, Ag. 4). For a concise ac-
count oI ancient Massalia, see Wackernagel 1966, 21302153.
2 STORY 112
(discussed in detail below), which is evident Irom the reciprocal accusations
and apologies when the boys board the ship again in the Greek city, requires
them to have spent considerable time together on that ship beIore arriving in
Campania.
2.1.6 The SigniIicance oI Massalia
In Roman literature the name oI Massalia (or Massilia as the Romans spelled
it) is loaded with political and cultural signiIicance. The city`s destiny was
perceived as intimately connected with that oI Rome Irom its very Iounda-
tion. Legend had it that in the times oI king Tarquinius the youthIul settlers
Irom Phocaea, which is sometimes portrayed as another sacked Troy (Luc.
3.340), had sailed up the Tiber and made Iriends with the Romans beIore
continuing on their journey to Iound Massalia in the midst oI savage na-
tions.
253
For the Romans they remained a symbol oI the old Greek civiliza-
tion miraculously preserved in the heart oI barbarian darkness.
254
Severity,
gravity and discipline were the communal virtues oI Massaliots lauded by
Roman authors (Cic. Flacc. 26.63; Phil. 8.6.19; V.Max. 2.6.7). These were
virtues that the Romans did not commonly associate with Greeks, but rather
with their own vetus Roma. Massalia was believed to have provided Iinancial
aid aIter the sack oI Rome by Gauls, and Ior this, according to Justin (Just.
43.5.10), it was granted 'immunity (immunitas), 'an auditors` place in the
senate (locus spectaculorum in senatu), and 'a treaty oI equal right (foedus
aequo iure). Like Rome it Iought against the Carthaginians. It had the repu-
tation oI a IaithIul Iriend and ally to Rome in war and peace (Just. 43.5.3).
Accordingly, the siege and subsequent capitulation oI Massalia to Caesar
during the civil war was perceived as symbolic oI the irreparable harm and
madness oI that conIlict. For Rome to turn against such an ally was typical
oI the selI-destructive Iraternal slaughter that was the civil war. In the ac-
count given by Lucan in the Pharsalia (3.298I.), the Massaliots Iace Caesar
with 'an un-Greek steadIastness (Luc. 3.302, non Graia levitate) and they
appeal to him by reminding him oI the historical relationship oI the two
states and demonstrating clearly their old Iashioned hatred oI tyranny and
civil striIe. Civil wars are evil, and iI Rome has the good Iortune to negotiate

253
There is a short history oI Massalia in Justin 43.35, which is an epitome oI Trogus` His-
toriae Philippicae Irom the Iirst century B.C.E.
254
This aspect oI the city`s image is emphasized in numerous sources: Cic. Flacc. 63, Phil.
8.9; Liv. 37.54; Sil. 15.16872; V.Max. 2.6.7; Tac. Ag. 4; Mela 2.77.
2. 1 SORTING THE FRAGMENTS 113
peace, Caesar and Pompey can both come to Massalia to dwell there in exile
(Luc. 3.333335).
Thus Massalia, like Troy in the poem oI Eumolpus about the Iall oI Troy
(Sat. 89), might be presented as a projection oI Rome herselI with respect to
her Iate in the civil war, the subject oI another oI Eumolpus` poems (Sat.
119124). The logic oI the admiration oI Massalia by the Romans is ex-
plained by A. Trevor Hodge in the Iollowing manner: 'Romans, almost
without exception, were Iulsome in their admiration, praising the Massaliots
as a kind oI puritan supermen, while speaking oI their politics and Ioreign
policy in terminology that tends painIully to remind a modern ear oI a right-
winger speaking oI a Iriendly banana republic.
Greek writers, however, have quite a diIIerent story to tell, and one that
resonates better with the tenor we are Iamiliar with in the story told in our
Iragments oI the Satvrica. They saw Massalia, again in the words oI A.
Trevor Hodge, 'as a kind oI Naughty Paree, O-la-la!`.
255
In Greek texts the
Massaliots have a reputation Ior being eIIeminate and soIt, which is proven
by the Iact that they wear Iloor-length tunics (Athen. 12, 523, c; Ps. Plutarch
Proverb. Alex. 60). The phrase 'you might sail to Massalia (Suid. 8 499, 0.
:"**",+"3 G,8=*8&".) is explained in the late tenth-century Byzantine lexi-
con the Suda in the Iollowing way: 'Used oI those living an eIIeminate and
soIt liIe, since the people oI Massalia used to live rather eIIeminately, wear-
ing Iancy long robes and perIumes (0GR #I3 HK,$#A%4. '"R ","'I.
U;3#43. -g /V% :"**",&I#"& HK,$#A%-3 ZU43 *#-,"9. G-&'+,"&. '"R
G-BX%8*& '"R =%-&. N%;83-&). In the same Iashion, the phrase 'you are
coming Irom Massalia (Suid. 8 3161, 0' :"**",+". '8&.) gets the gloss:
'Used oI eIIeminate and luxury-loving people, inasmuch as the men oI Mas-
salia are said to wear eIIeminate clothing and perIume, and tie their hair up,
and are a disgrace because oI this soItness (0GR #I3 HK,$B%&I3 '"R
#8H%$A-343, G"%F*-3 0'8+3-$. )"*R HK,=#8%-3 *#-,+U8*H"& 8$%&*-
A3-$. '"R #(. #%+N"& J3"B-$A3-$., '"R B&V #"=#K3 #X3 ","+'"3
J*NK-389.). Another noted peculiarity oI the Massaliots is male proper
names with Ieminine endings: Protis, Apellis, Thespis, Zenothemis, Taxaris,
Charmis. The only Roman writer who adopts this Greek attitude towards the
Massaliots is the comic dramatist Plautus, who lets a character reIer to eI-
Ieminacy as 'practicing the morals oI the Massaliots (Plaut. Cas. 963). This
atypical attitude Ior a Roman text could be explained by Plautus` own ad-
mission that he adapted the Casina Irom a play by the third-century Greek
poet oI New Comedy, Diphilus oI Sinope.

255
Both citations are Irom Hodge 1998, 4.
2 STORY 114
As Iar as we can tell, Massalia appears to have been an old Iashioned
city state with an aristocratic constitution, and very conservative with respect
to its religious customs and the Greek language. An archaic Ionic dialect
held its ground there, and Greek continued to be the spoken language until
late antiquity.
256
Encolpius` marked preIerence Ior old Greek literature and
art, and his apparently genuine astonishment at things seen and heard on his
trip through Italy under the Julio-Claudian dynasty, are thus intelligible as
aspects oI his Massaliotic background. His surprise is thereIore not due to
stupidity, but to his Ioreign and culturally 'Iiner origin, and may to a certain
extent be used to measure the deviant mores oI those whom he encounters.
Because oI his background, Encolpius stands closer to the admirable Greeks
oI the past than do those characters oI the story who are more Iamiliar with
and accepting oI the contemporary scene. His simple-mindedness is obvi-
ously at least in part intended as an intellectual virtue, and his literary mind-
set and nostalgia Ior the past glory oI Greece is more appealing to his Iine
Roman audience than the world oI sordid acquisitiveness. At the same time
the joke is always on Encolpius, because oI his Massaliotic eIIeminacy, his
soItness and obsession with Giton and his scholastic way oI reacting to the
world. The phallic Encolpius is a contemporary satyr, a comic Iigure, but not
in any comprehensible sense a parody oI the male protagonists oI the Greek
novel.
There is another reason why Massalia is especially appropriate as the
origin oI the narrator oI the Satvrica. Encolpius` home city, which prided
itselI on having a port oI major commercial importance in the western Medi-
terranean, was Iamous Ior its Atlantic seaIarers and their incredible trave-
logues. Pytheas oI Massalia, Ior one, claimed to have sailed into the outer-
sea and north along the coast. His voyage supposedly took him to many pre-
viously unknown lands and led to the discovery oI the mysterious island oI
Thule. But he was branded the very worst oI liars by Strabo (1.4.3) and
mocked by Polybios (34.5.7) as someone too pooranother poor Mas-
saliotto undertake an expedition to Iar-away places. Antonius Diogenes
certainly parodies Pytheas amongst others in his lost The Wonders bevond
Thule (Phot. Bibl. Cod. 166). Euthymenes, another Massaliot adventurer,
claimed to have rounded the southern tip oI AIrica and located the Nile`s
source and thus solved this centuries-long debate. But he is called a braggart
by the sophist Aelius Aristides and his Periplous nothing but an 'account Ior

256
See Clerc 1971, 1:458564, on the intellectual culture. The Greek inscriptions oI Massalia
dating Irom the Roman era are notable Ior their archaic and Ionian Iorms, though this may
perhaps be due to an oIIicially cultivated archaism to boost local patriotism rather than the
survival oI the old dialect in common speech.
2. 1 SORTING THE FRAGMENTS 115
Alkinous (Aristid. Aeg. p.354 |Jebb|, JGF,-/-. ab,'+3-$), i.e. oI the same
type as the lying Iables told by Odysseus to the gullible king oI the
Phaeacians. The Younger Seneca cites Euthymenes oI Massalia only to re-
Iute his claims, and adds that in the olden days 'there was room Ior lies;
because the realms oI the outer sea were unexplored, they were allowed to
make up Iables (Sen. Nat. 4.2.2225, tunc erat mendacio locus, cum ignota
essent externa, licebat illis fabulas mittere). Lucan, with an obvious allusion
to Nero`s interest in the problem oI the sources oI the Nile (Sen. Nat. 6.8.3),
also reIers to the Massaliot`s story as hearsay, rumor, in a conversation be-
tween Caesar and Acoreus, an Egyptian priest (Luc. 10.2557). Because oI
such incredible travelers` tales connected with the city oI Massalia, Aelius
Aristides uses the term Massaliotic Iables` (Aeg. p.353 |Jebb|, TH-&
:"**",&4#&'-+) to cover this type oI travelogue and relies on his readers to
know to what he is reIerring. Whether the Massaliots Pytheas and Euthyme-
nes were mere liars or misunderstood explorers Iar ahead oI their time is
diIIicult to ascertain, but it is certain that they were known to later authors as
Odyssean spinners oI yarns, which makes their city especially appropriate as
the home oI Encolpius, the narrator oI the travelogue we know as the Sa-
tvrica.
The outlines oI what happened in the Iirst episode in Massalia are not
diIIicult to reconstruct based on such evidence as the Servius Iragment and
the Iormulaic Irames oI Greek travelogues. Like strangers in Greek literature
typically do, Encolpius will have begun his tale by identiIying himselI
through his city oI origin. He will Iurther have associated with his Mas-
saliotic identity the qualities that deIine him most as a character and a narra-
tor: a noble mindset, old-style education and travel, alluding also to his
soItness` and love Ior Giton. His education and taste Iit well with the image
oI Massalia as a university town in imperial times, and his travels Iit well
with the Iact that Massalia counted among its Iamous citizens certain travel-
ers who explored the outer-ocean and came back to tell incredible tales. One
does not have to ponder long the possibility oI a discursive strategy Ior the
opening oI this story to see how the hackneyed motiI oI the Phaeacian tales
oI Odysseus could here be given yet another creative spin in Greek literature.
The whole set-up is highly adaptable Ior an ancient Greek satire about litera-
ture, human attitudes and morals. The Odyssean traveler who goes Irom city
to city and gets to know many places and the minds oI many men is an ideal
vehicle Ior such a satire. Rather than taking a trip to the Iabulous edges oI
the world, as his Iellow Massaliots claimed to have done, the overeducated
but unheroic Encolpius goes to the heart oI civilization to Iace moral and
esthetic monstrosities oI no less Iabulous proportions. This movement in-
2 STORY 116
wards to the ordinary (and prosaic) and away Irom the mythical (and poetic)
is no doubt related to the therapeutic strategy oI Greek Cynic satire which
ridiculed scholars Ior studying in detail the errors oI Odysseus while being
ignorant oI their own. For Petronius the eIIeminate Massaliot provided, addi-
tionally, an ideal vehicle Ior a satire to subvert Roman chauvinism.
2.1.7 Sailing Irom Massalia
II our geographical and cultural remapping oI the Satvrica has made more
persuasive our thesis that Encolpius, Giton and Tryphaena boarded the ship
oI Lichas in Massalia, there is still much about their relationships and the
events oI the voyage that remains in the dark. We would, perhaps, know
considerably more iI we had Giton`s hurried exposition oI the causes oI their
enmities and the present threat, which puts Iear into Eumolpus (101.7, rap-
tim causas odiorum et instans periculum trepidanti Eumolpo exponit [sc.
Giton]). But since the audience/reader oI the original had all this inIorma-
tion, the narrator does not bother repeating it. However, as will become clear
in the Iollowing, it is Iairly easy to recover what caused the Ialling-out be-
tween the characters. The enmities between Tryphaena and Lichas, on the
one hand, and Encolpius and Giton, on the other, are virtually re-exposed by
the narrative itselI through the accusations and apologies which precede their
partly Iorced reconciliation.
Tryphaena primarily misses Giton
257
and Lichas is most eager to get his
hands on Encolpius.
258
In Giton`s words the boys are on the run Irom these

257
100.4, 'si quis deus manibus meis` inquit 'Gitona imponeret, quam bene exulem exci-
perem` ('II some god were to place Giton in my hands`, she said, how well I would re-
ceive the exile`). She dreams that the statue oI Neptune, which she had noticed three times
in the sanctuary at Baiae, says to her 'you will Iind Giton in Lichas` ship (104,2, 'in nave
Lichae Gitona invenies`); cI. 108.5, intentans in oculos Trvphaenae manus usurum me
viribus meis clara liberaque voce clamavi, ni abstineret a Gitone iniuriam mulier damnata
et in toto navigio sola verberanda ('I shook my Iist in Tryphaena`s Iace, and shouted in a
loud and bold voice that I would use violence, iI she did not leave oII insulting Giton, Ior
she was a wicked woman and the only person on the ship who deserved Ilogging).
258
Encolpius Iears Lichas especially: 100.34, sed repente quasi destruente fortuna constan-
tiam meam eiusmodi vox super constratum puppis congemuit. 'ergo me derisit?` et haec
quidem virilis et paene auribus meis familiaris animum palpitantem percussit ('But sud-
denly, as though Iate was trying to destroy my resolution, a voice on the ship`s deck
groaned: So he made me a laughingstock?` And this manly voice was somehow Iamiliar to
my ears, and my heart beat Iast as I heard it). The Priapus oI Lichas` dream says to him:
104.1, 'Encolpion quod quaeris, scito a me in navem tuam esse perductum` ('Know that I
have lead Encolpius, whom you seek, to your ship`). When Tryphaena hears Giton`s voice
2. 1 SORTING THE FRAGMENTS 117
people (101.6, 'hi sunt` inquit Giton 'quos fugimus`). The angry adults
accordingly reIer to the boys as 'runaways and 'culpable (104.11 and
105.10, fugitivi; 106.3, noxii). At this stage the relationship between the boys
and Lichas in particular is presented as that oI runaway slaves to an irate
master (Sat. 101.10). The impractical declamatory solutions which the trio
scholastically invent are another demonstration oI the uselessness oI decla-
mations in real liIe (cI. Encolpius` claim at Sat. 1.12.4) and they never
manage to represent the boys as anything but slaves who have committed
some wrongdoing against their master, although Eumolpus is the one who
would pose as their dominus (102.9, 103.4, 105.2). In his deIense oI Encol-
pius and Giton the poet presents them as 'Iree men, 'noble and 'honest,
but even so he too reIers to them as 'runaway slaves and 'enemies who
have surrendered in the same sentence (107.35). His only eIIective rhetori-
cal argument is to appeal to whatever residual sentiments there may be aIter
the amorous liaisons oI the adults and the boys (107.34).
259

Eumolpus is, oI course, lying outrageously in claiming that the boys
willingly returned to the ship and will say whatever he thinks is going to be
oI help. Lichas protests to his calling the boys ingenui and honesti (107.35),
not however by arguing that they are slaves in a legal sense, but because they
have become 'guilty(107.9, noxii) and 'liable (107.10, rei) in his eyes. II
Encolpius was his Iriend beIore, all the more reason to call him, besides a
thieI, a parricide as well (107.12).
260
Neither boy is actually the slave oI Try-
phaena or Lichas. What has so debased them with respect to their Iormer
Iriends is that they have Iallen captives (113.7, captivitate) to people whom
they have hurt (107.10, laesi) and who now wish to wreak vengeance upon
their heads (108.9, dimicantium furor, illis pro ultione, nobis pro vita pug-
nantibus |'the Iury oI the Iighters, they Iighting Ior revenge, we Iighting Ior
our lives|). This double nature oI the relationship between the people on
board the ship shows that there is a dramatic switch, a 'beIore and 'aIter

and runs to him, Lichas runs to Encolpius: 105.9, Lichas, qui me optime noverat, tamquam
et ipse vocem audisset, accurrit et nec manus nec faciem meam consideravit, sed continuo
ad inguina mea luminibus deflexis movit officiosam manum et 'salve` inquit 'Encolpi`
('Lichas, who knew me intimately, came running as iI he too had heard my voice, and did
not inspect my hands or my Iace, but immediately looked down and applied his busy hand
to my groin, saying How are you, Encolpius?`).
259
CI. 106.2, volebat Trvphaena misereri, quia non totam voluptatem perdiderat ('Tryphaena
wanted to Iorgive, because her pleasure had not wholly died away).
260
107.11, 'at enim amici fuerunt nostri. eo maiora meruerunt supplicia ('But they were
once our Iriends |you say|: then they deserve the harsher punishment`).
2 STORY 118
in the boys` relationship with the adults, and that the partly erotic and partly
criminal departure Irom the ship marks that turning point.
261

Harder to Iigure out are the original relationships in the period beIore the
Ialling-out when things were apparently going more smoothly. And even
beIore this happy period we must posit an initial encounter, no doubt when
the woman and the boys Iirst boarded the ship oI Lichas in Massalia. During
this initial period, then, we may presume that Tryphaena, Encolpius and
Giton were primarily exules in the eyes oI Lichas, who knew that Encolpius
was exiled as a result oI playing the scapegoat. The wiIe oI Lichas (her name
was most likely 'Hedyle)
262
was also on board the ship in the beginning and
seems to have had an important role to play, especially in the boys` departure
Irom the ship. Eumolpus` answer to Encolpius` question about the owner
and the passengers gives us the basic Iacts about the captain and his most
prominent passenger, Tryphaena. Lichas and his ship are returning to their
homeport in Tarentum (100.7, dixero Licham Tarentinum esse dominum
huiusce navigii), and he has been collecting merchandise to sell presumably
on the local market (101.4, onus deferendum ad mercatum conducit). Try-
phaena is being brought on the ship to Tarentum as an exile (100.7, huiusce
navigii, qui Trvphaenam exulem Tarentum ferat?). The same would seem to
apply to the boys. Originally, they were most likely going to Tarentum.
Tryphaena is also, according to Eumolpus, the most beautiIul oI all
women and travels hither and thither because oI pleasure (101.5, omnium
feminarum formosissima, quae voluptatis causa huc atque illuc vectatur).
The reason he gives Ior her travels (voluptatis causa) might seem to conIlict
with her status as exile. However, iI her exile had something to do with her
lust, as is likely, since this seems to be her dominant character trait, this is
not a problem.
263
Eumolpus might also be reIerring to her and Lichas` search
Ior the boys, which seems to have been their sole activity Irom the time the
boys leIt the ship. Tryphaena somewhat resembles an unusually wealthy
Greek hetaira, but considering her status as exile she is more likely to be
on the analogy oI Lichas` wiIe and the recently married Circe in Crotonthe

261
113.3, non dubie redierat in animum [sc. Lichae] Hedvle expilatumque libidinosa migra-
tione navigium ('no doubt it was Hedyle who came to (Lichas`) mind and how his ship had
been pillaged on her libidinous elopement).
262
Hedvle is Bcheler`s conjecture Ior the hedile or edile oI the tradition. That Lichas had a
wiIe on board and that she played an important role, however, is not in question: 106.2,
Lichas memor adhuc uxoris corruptae iniuriarumque, quas in Herculis porticu acceperat
('Lichas, still remembering the seduction oI his wiIe and the insults he took in the Portico
oI Hercules).
263
According to the conventions oI New Comedy, going into exile is the natural reaction to
Irustrated love, see Zagagi 1988, 193209.
2. 1 SORTING THE FRAGMENTS 119
libidinous wiIe oI a wealthy Massaliot who has abandoned husband and
home. Besides, her amorous attachments would hardly be so culpable were
she a proIessional harlot. But perhaps most signiIicantly she blushes at the
end oI Eumolpus` satire about the Widow oI Ephesus, a story introduced as a
demonstration oI Iemale levity, how easily they Iall in love, how Iast they
even Iorget their children; that no wiIe is so virtuous that she isn`t willing to
sacriIice everything Ior the love oI a stranger (110.67, peregrina libidine).
Why would she have blushed so violently (113.1, erubescente non mediocri-
ter Trvphaena) iI she had nothing in common with the widow oI the story?
When peace has been brokered she and Giton are very close again
(109.8; 110.3; 113.1 and 5). That they have been close beIore is shown by
the Iact that Tryphaena`s 'most IaithIul slaves (114.7), especially the ancil-
lae, recognize Giton`s immediate cry oI pain even beIore their mistress.
264

She knows his voice well enough to be subliminally 'upset or 'disturbed,
turbata, at hearing it (105.5), even iI she does not recognize it immediately.
When all her slave girls have run to his aid and called on their mistress Ior
help (105.6), she is quick to respond.
265
Likewise, when Giton threatens to
cut oII his genitals, she stops him by showing herselI earnestly willing to
Iorgive (108.10, inhibuitque Trvphaena tam grande facinus non dissimulata
missione |'Tryphaena prevented this great disaster Irom happening by ear-
nestly oIIering us pardon|). Such unconditional Iorgiveness is a sure symp-
tom oI love (in the Satvrica a sentiment no diIIerent Irom lust) as Encolpius
so well demonstrates by his willingness to take Giton back whatever he has
done.
266
The reason Ior the Iamiliarity oI the handmaidens with the boy is
not that Giton is Tryphaena`s slave, as some have suggested, but that these
creatures were indispensable intermediaries in comic love aIIairs. Accord-
ingly, they are especially knowledgeable about the most intimate oI their
mistresses` secrets. To convince ourselves oI this we need only observe the
likeness oI Tryphaena`s ancillae to Quartilla`s Psyche and Circe`s Chrysis.

264
The text is strange here: 105.6, non solum ergo turbata est, sed ancillae etiam omnes fami-
liari sono inductae ad vapulantem decurrunt, but need not be corrupt. Bcheler hesitatingly
suggests sola Ior solum, but prints a lacuna aIter sed and suggests the missing words: acces-
sit quoque propius et acrius uociferantem intuetur. Ernout changes ergo to era, and adduces
Bcheler as the authority. Mller adopts Novk`s emendation and adds ea~ aIter ergo.
Even iI era is accepted, this does not necessarily imply that Giton is Tryphaena`s slave,
since her status as mistress would be justiIied by the reIerence to her handmaidens, and
does not necessarily have anything to do with Giton.
265
105.8, deflectit aures Trvphaena iam sua sponte credentes raptimque ad puerum devolat
('Tryphaena lent a ready ear to the cry and hurried to the boy).
266
Lust also motivates Encolpius when he wishes to break up his Iriendship with Ascyltos
(10.7), and love when he receives Giton back and Iorgives him later on (91.6).
2 STORY 120
The chambermaids oI Tryphaena typically repair the beauty oI the boys by
restoring to them their lost hair and eyebrows with their mistress` cosmetics
(110.15). In one isolated Iragment one oI them seems to be ingratiating
herselI with Encolpius (113.11), just as Chrysis later attempts to replace her
mistress as the young man`s lover (139.4). These subordinates can at times
be quite imperious and they are not always under their owner`s control,
much like the slaves oI Trimalchio (e.g. the dispensator).
The idea that Giton is Tryphaena`s slave is contradicted by her initial
reIerence to him as exul (100.4), likewise by Eumolpus` statement that the
boys were once close Iriends with the adults (107.1, aliquando amicissimis),
a statement which is acknowledged by Lichas (107.11, at enim amici fuerunt
nostri). This seems an improbable way to reIer to slaves in the ancient world.
Furthermore, although now she does not talk to him, at an earlier point in the
story Encolpius was intimate with Tryphaena and she was happy to have him
as her lover (113.8, neque Trvphaena me alloquebatur tamquam familiarem
et aliquando gratum sibi amatorem). But this was beIore Giton took his
place (113.7, nec tamen adhuc sciebam utrum magis puero irascerer quod
amicam |sc. Trvphaenam| mihi auferret, an amicae quod puerum cor-
rumperet |'I couldn`t make up my mind whether to be more angry with the
boy Ior stealing my girlIriend Irom me, or with the girlIriend Ior seducing
the boy|). It seems quite pointless to assume that she lost interest in Encol-
pius and Iell in love with her own slave whom she would have known well
beIore.
II the little inIormation we have oI the boys` relationship with Tryphaena
is matched with the pattern which Vincenzo CiaIIi was Iirst to point out (Iur-
ther discussed below) oI Encolpius making Iriends and lovers oI people who
then become his enemies as soon as they take an interest in Giton, we can
account Ior both Tryphaena`s switch Irom Encolpius to Giton and the devel-
opment Irom Iriendship to animosity. With a certain amount oI plausibility
we may thereIore assume that Encolpius Iirst had an aIIair with Tryphaena in
Massalia. True to type, she would like Circe have lusted aIter Encolpius
during that year when he was receiving giIts oI Iood in his role as scapegoat
and when the reputation oI his penis would have been likely to attract the
interest oI the libidinous women oI the town. In both the Greek and the Latin
Ass stories we have a similarly needy lady who is attracted to the donkey
penis only (Asin. 51, 56; Met. 10.1922). According to the satiric ethos,
wealthy and beautiIul matrons like Tryphaena, Hedyle and Circe are ex-
pected to lust aIter sexy outcasts, slaves, gladiators and condemned criminals
(Sat. 126.10, matronae, quae flagellorum vestigia osculantur |'married
women, who kiss the scars oI a Ilogging|). Later then, when Encolpius had
2. 1 SORTING THE FRAGMENTS 121
been expelled, she may have run away Irom home with himin the same
manner as Hedyle, Ascyltos and even Eumolpus, other initial Iriends and
lovers oI Encolpius who join the 'brothers on their travels, only later to
become unwanted and suspected by the protagonist. An alternative (which
does not necessarily exclude the Iirst option) would be that she was Iound
out by her husband, and thus too became a voluntary exile to escape the con-
sequences oI her inIidelity, i.e., the wrath oI her husband. A third possibility
would be that the boys and Tryphaena met on the ship. But this seems to me
less likely, since the boys typically need a third partner to help them move
Irom one episode to another. The initial aIIair is certain to have been compli-
cated, but the logic oI erotic liaisons which primarily motivates the action oI
the Satvrica is relatively simple and constitutes a remarkably reliable reIer-
ent Ior Iiguring out the lost parts oI the plot. Only later, then, when they
were on board the ship oI Lichas would Tryphaena have developed her Ilam-
ing passion Ior Giton, which so excited the rabid jealousy oI Encolpius as to
cause him to plan a desperate escape (108.5, 108.8, 108.14 v. 5).
267

The voyage presumably lasted long enough Ior Encolpius to have had
erotic relationships also with Lichas and later his wiIe, although a long nar-
ration is not as such an indication that a long time must have passed. How-
ever, Encolpius had at diIIerent times and in diIIerent situations been erotic-
ally involved with each oI the important individuals on board the ship: Gi-
ton, Tryphaena, Lichas and Hedyle. When Tryphaena had lost interest in
Encolpius and had made Giton the new object oI her lust, Encolpius was Iree
to begin the relationship with Lichas, which must have been initiated by the
captain. Lichas` wiIe, then, would typically have been angered at her hus-
band`s marital inIidelity, and might have used it as an excuse to do like-
wise.
268
In Apuleius` Metamorphoses we have the story oI the pistor who
punishes his wiIe`s youthIul lover by taking him to bed and Ilogging him the
day aIter (9.278), but Lichas seems genuinely to lust aIter Encolpius and so
it is more likely that he preceded his wiIe as Encolpius` lover. I suspect that
Eumolpus` ostensible Iiction about the boys (supposedly his slaves) having

267
113.5, Trvphaena in gremio Gitonis posita modo implebat osculis pectus, interdum concin-
nabat spoliatum crinibus vultum. ego maestus et impatiens foederis novi non cibum, non
potionem capiebam, sed obliquis trucibusque oculis utrumque spectabam. omnia me oscula
vulnerabant, omnes blanditiae, quascumque mulier libidinosa fingebat ('Tryphaena was
now lying in Giton`s lap, covering his breasts with kisses one moment, and sometimes ca-
ressing his shaven head. I was depressed and unhappy about our new treaty. I didn`t touch
Iood or drink, but kept looking askance at them both with anger in my eyes. All their kisses
wounded me, all the pleasant wiles that the libidinous woman invented.).
268
Compare this to the marital row oI Fortunata and Trimalchio. Note especially the reIerence
to ex aequo ius firmum (74.9).
2 STORY 122
spent the night with a Iictitious amica (105.3) may somehow in the solipsis-
tic world oI this story mirror how Hedyle was 'corrupted. The boys and
Lichas` wiIe seem to have made a pact against her husband, although Encol-
pius was primarily thinking oI getting Giton away Irom Tryphaena. Accord-
ing to the central erotic pattern Hedyle would sooner or later have taken an
interest in Giton and so would eventually have to be gotten rid oI as well by
our lovesick hero, although some variation may have been built into the mo-
tiI. Something, in any case, caused them to part company Ior when we meet
the 'brothers in the Iirst extant scenes she has ultimately been replaced by
Ascyltos as the third man and rival. (We will leave to section 2.1.8 and 2.1.9
the discussion oI the many adventures that Encolpius experienced in the long
interval while he wandered about in Campania, aIter he escaped Irom the
ship and beIore we meet him in the Greek city).
2.1.8 The urbs Graeca
The location where the conIrontation and escape oI these three characters
(Encolpius, Giton, and Hedyle) took place is named in the text as Herculis
Porticus (106.2). Julius Beloch, in his magisterial survey oI ancient Cam-
pania, locates this portico in Puteoli by conjecturing that a part oI the long
Sulenhallen by the harbor, which Cicero calls Porticus Neptuni, may have
been called Porticus Herculis.
269
This is probably the strongest evidence
there is to locate the place. Scholars most oIten assume that Herculis Porti-
cus was somewhere in the complex oI monumental buildings which adorned
the resort at Baiae, because Tryphaena reIers to having thrice seen there in a
sanctuary the simulacrum oI Neptune.
270
Here at least we have a deIinite

269
See Beloch 1890, 134: 'Die verschiedenen Theile dieser Portiken waren nach Gttern
benannt. So die Porticus Neptuni, die Cicero erwhnt |Acad. pr. 2.25.80|, nach der ab-
bildung bei Bellori etwa zwischen dem grossen Molo und der Kirche PuriIicazione a mare
gelegen. Ein anderer Theil dieser Sulenhallen war wohl die Porticus Herculis, von der
Petron erzhlt. Sie boten den Puteolanern einen beliebten Spaziergang. Other explanations
Ior the change Irom Neptune to Hercules could be poetic license or Iaulty memory, but in-
venting such arguments is too easy Ior them to count as evidence. The Iact is that no one
can positively identiIy the place.
270
The transmitted text runs like this: 104.2, exhorruit Trvphaena et 'putes` inquit 'una nos
dormiisse, nam et mihi simulacrum Neptuni, quod Baistor asvlo notaveram, videbatur
dicere. in nave Lichae Gitona invenies.` Scaliger isolated Bais (oIten written Baiis with
Bcheler). Heinsius emended tor asvlo to in~ peristylo; Gronovius to in~ peristylio;
Bcheler to in~ tetrastylo, and that reading was adopted by Ernout, Heseltine and Mller.
But no peristvlum, peristvlium, or tetrastvlum has been Iound at Baiae (Beloch op.cit., 186).
However Ribezzo 1930, 57, has provided by Iar the best solution: 'Bais ter asylo nota-
2. 1 SORTING THE FRAGMENTS 123
location, but nothing Iorces us to assume that the boys leIt the ship at Baiae,
just because Tryphaena says she had been there. However, the urbs Graeca
is bound to be close to Baiae. II we consider the Iirst encounter with Quar-
tilla, which precedes the arrival oI the boys at the urbs Graeca, everything
points to their having leIt the ship somewhere else than in the city oI Aga-
memnon and Trimalchio, which they entered on Ioot. And Baiae does seem
the appropriate setting Ior the corruption oI a matron (Hedyle), considering
its reputation as a sexually corruptive beach resort (e.g. Varr. Men. 44 |Ast-
bury|; Prop. 1.11). Thanks to natural hot springs, Baiae was a Iashionable
spa and resort, and the emperors built palaces there, while the wealthy built
their villas. For Petronius to pass by Baiae without making use oI it in his
licentious work would have been an uncharacteristic act oI restraint, and
would certainly have leIt his audience disappointed. However, iI the boys
leIt the ship beIore it reached Baiae, there wouldn`t have been any episode
set there, Ior nowhere in the preserved part oI the story does Encolpius as
narrator oIIer a narrative oI things to which he hasn`t been a direct witness
himselI.
271
We can just imagine that iI the boysand thereIore the narrative
as wellhad leIt the ship beIore reaching the Iamous resort, Tryphaena`s
reIerence to having seen the eIIigy oI Neptune at Baiae would only have
inIlamed the interest oI the reader without satisIying the desire Ior knowing
more about so congenial a setting Ior the Satvrica. The risk is that the
reader/hearer would have seen this as a missed opportunity. It is one thing
never to arrive in Tarentum, but Baiae is a diIIerent story altogether. I think
it Iair to assume that since Baiae is mentioned at all in the Satvrica it is

veram. By emending only one character, which is likewise emended by all other editors,
we get a Iine sense to the passage. Asvlum is a general word Ior a sanctuary or a temple (and
perhaps 'resort, see quotation Irom Fronto below). The local ablative asvlo without in is
also possible (Verg. Aen. 2.761, Iunonis asvlo, cI. Austin 1980 ad.loc.; Fro. Amic. 2.3, has
asvlo recreari but the ablative here could be instrumental), and ter adds an appropriately
superstitious tone to Tryphaena`s statement (Petronius is very Iond oI ter in this Iormulaic
sense both in prose and verse; 88.4, 98.4, 123.1 v. 240, 131.5, 132.8 v. 1 and 2, 133.3 v. 16;
the model is without doubt Verg. Aen. 6.229, 10.873, 11.188I.). OI some relevance here is
perhaps Ribezzo`s report oI a 'recent underwater Iind at Baiae oI a statue oI Neptune.
271
In this respect he diIIers Irom Achilles Tatius` Clitophon (Reardon 1994, 82), although we
cannot know Ior certain what Encolpius did in the lost parts oI the work. This argument
seems to me to make Walsh`s 1970, 74, 'tempting emendation oI Herculis Porticus to
Herculis Portus, or Monaco, lose its plausibility. Moreover, our inability to locate exactly
this particular portico is surely not evidence Ior corruption in the text. Our modern know-
ledge oI Campanian cities in the Iirst century is certainly not so complete. Finally, Herculis
Portus on its own might be misunderstood, since there was more than one place by that
name, and ancient Monaco would perhaps be better reIerred to in Iull as Herculis Monoeci
portus (v. Pauly-Wissowa s.v.).
2 STORY 124
bound to have been treated in at least one episode. AIter all, it is summer and
the right season Ior Baiae.
272
Would Tryphaena and Hedyle, given their type,
have agreed to pass by the hedonistic Baiae? We know at least that Try-
phaena did not.
Now, iI Beloch was right that the Herculis Porticus reIerred to is the
same as the long portico by the harbor in Puteoli, the southward movement
oI the large vessel oI Lichas would mean that it Iirst entered the port oI
Puteoli with the boys still on board, Ior there would not be any means Ior a
large ship like that to stop at Baiae itselI. Puteoli had been the biggest com-
mercial harbor in Campania Ior two centuries and it was especially Iurnished
with a giant mole, over three hundred meters long, whose remains are still
visible.
273
From there the characters could have gone on Ioot to Baiae, which
is very close. Thus the boys, by returning to the ship beIore the others, could
have stolen the rattle and robe oI the ship`s eIIigy (113.3, 114.5) and then
made their escape in the company oI Hedyle, but only to be conIronted by
Lichas and Tryphaena in the portico oI the harbor itselI, in an incident Irom
which the captain and his pretty passenger are still smarting and Irom which
they evidently came away scathed and humiliated (106.2).
274
AIter running
away Irom Puteoli, the boys would eventuallythey had many adventures in
between which we shall discuss in the Iollowing sectionhave gone through
the Crvpta Neapolitana and so entered the Greek city oI Neapolis by the
normal route.
But is the urbs Graeca Neapolis? This equation cannot oI course be
proven, but it has the merits oI making some sense oI the otherwise conIus-
ing reIerences in the extant text. The problem is that the Greek city is also
characterized as a colonia, which may be taken to denote a Roman colony.
Rose argued that the urbs Graeca could not be Neapolis because it didn`t
become a colony until Antonine times.
275
But Puteoli, Rose`s candidate,
oIIicially became a colony only in 63 C.E. (Tac. Ann. 14.27), and Hermeros

272
See Rose 1962, 4067.
273
Beloch 1890, 131: 'Unterhalb des BurgIelsens und durch ihn vor dem Scirocco geschtzt
liegt der HaIen der Stadt, der erste an dieser ganzen Kste nach dem von Misenum. Als
aber Puteoli anIing, sich zum ersten Handelsplatz Italiens auIzuschwingen, gengte dieser
natrliche Schutz nicht mehr und es wurde jener Molo in`s Meer geworIen, den schon das
Alterthum als eins der grssten Wunderwerke pries und der noch heute von allen Ruinen
Pozzuoli`s unser grsstes Interesse in Anspruch nimmt.
274
106.4, nec se [sc. dixit Trvphaena] minus grandi vexatam iniuria quam Lichas, cuius pu-
doris dignitas in contione proscripta sit ('Tryphaena said that she had been just as gravely
wronged as Lichas, considering that her reputation Ior chastity had been publicly and ad-
versely shown up).
275
Rose 1962, 4045.
2. 1 SORTING THE FRAGMENTS 125
says that he came to the colonia as a boy (57.9, puer capillatus in hanc colo-
niam veni). Are we to imagine that the dramatic date oI the Satvrica is that
late? Are the claims oI Ganymedes (44.12, 16), Hermeros (57.9), and Tri-
malchio (76.10) that their city is a colonia reliable evidence Ior its legal
status according to Roman law? One could counter this argument by saying
that they could just as well be using the Latin word colonia in the Greek
sense, as the equivalent oI JG-&'+", and in that sense Neapolis was certainly
Irom its very Ioundation the colonia oI Cumae.
276
Loose and non-technical
language would be highly characteristic oI these men.
Encolpius` statement (81.3) that he is in a city with a Greek identity is
intrinsically more reliable and inIormative than the Ireedmen`s use oI colo-
nia.
277
Now, oI the three principal candidates, Cumae, Puteoli and Neapolis,
only the last one could be, and was regularly, reIerred to as a Greek city.
278

Neapolis happened to be the Campanian center Ior Greek culture, and it was
known as a place which oIIered rhetorical and philosophical education to
youth and civilized peace and quiet to retired politicians and other wealthy

276
Serv. Aen. 1.12, veteres colonias ita definiunt. colonia est coetus eorum hominum, qui
universi deducti sunt in locum certum aedificiis munitum, quem certo iure obtinerent. alii.
colonia est quae graece!JG-&'+" vocatur ('The ancients deIined colonies thus: a colony is a
gathering oI men who together are led to a speciIic place to construct buildings there, which
they possess by certain right. Others deIine it thus: colonia in Latin is what is called JG-&'+"
in Greek).
277
Rose 1962, 404, quotes Juvenal (3.6061), non possum ferre, Quirites, / Graecam urbem
('Roman citizens, I cannot endure this Greek city), which reIers to Rome itselI. But this
statement oI Juvenal is backed up with much context which contrasts successIul Greeks in
the capital itselI with the poor local citizens. I do not deny that there are indications that the
place is Roman, beyond the word colonia. As I shall explain in section 3.2.5, these are most
likely Roman elements added to the description oI a Greek city, during the process oI adap-
ting the Satvrica Irom a Greek model.
278
Tac. Ann. 15.33; Str. 5.246, 6.253; Cic. Tusc. 1.86, Arch. 5.10. The oIIicial language, even
aIter Naples became a municipium, was Greek (Cic. Fam. 13.30.1, Att. 10.13.1). And the
city also had a Greek calendar. On the Greek language in ancient Naples, see Leiwo 1994.
Cumae, on the other hand, although a very ancient Greek settlement (Str. 5.243), was
sacked in 421 B.C.E. by Campanian Samnites and became aIter that predominantly Oscan,
while the Greek inhabitants Iled to Naples, their own colony. According to Livy (40.42.13)
Rome granted Cumae in 180 B.C.E. the right to use Latin as the oIIicial language at the re-
quest oI the citizens themselves. As Ior Puteoli, its origins are obscure, but in 194 B.C.E.
three hundred Roman Iamilies were settled there (Liv. 32.29, 34.45). Later, Sulla and Au-
gustus may also have sent coloni, and Irom the second century onwards its strategic and
commercial importance as the main harbor oI Rome was such that its identity could not be
other than Roman. Its oriental and Jewish inhabitants did not necessarily contribute to mak-
ing its character Greek.
2 STORY 126
Romans.
279
The road Irom Neapolis to Puteoli was lined with the villas oI
the great statesmen (Cicero, Pompey, Caesar, Domitian, Lucius Piso, Cato
Uticensis, Lucullus). An important part oI Neapolitan identity was its
schools and education (Stat. Silv. 5.3.112), which gave to the city the by-
name 'learned Naples, docta Neapolis (Col. 10.134; Mart. 5.78.14). Many
known literary Iigures were citizens.
Agamemnon, accordingly, and the other scholastici, the pinacotheca in
the temple portico, and last but not least Eumolpus the poet himselI, are
thereIore very much at home in Neapolis. Furthermore, the lack oI an atrium
and other signiIicant details about the house oI Trimalchio show it to be
Hellenistic and unlike the typical Roman houses on Campanian excavation
sites.
280
The plan oI Encolpius and Ascyltos to earn a living Irom their
knowledge oI letters (10.46), a plan which they undoubtedly Iormed prior
to arriving, thus seems to spring Irom the reputation oI docta Neapolis, in the
same manner that the conception oI Eumolpus` proIitable mimus rises Irom
inIormation about the reputation oI the ghost-town oI Croton. Finally the
similar cultural identities oI Massalia and Naples with respect to Rome make
this city a likely place Ior the Iugitive Encolpius to want to seek out.
As Ior earlier stops in other seaports on the way Irom Massalia, there is
no reason why we should not accept Encolpius` statement that he was in
Rome during the Saturnalia (69.9). It seems appropriate that a big commer-
cial vessel sailing Irom Massalia to Tarentum would make a stop in Ostia.
This was aIter all Rome`s main harbor Ior vessels coming Irom the west, and
thanks to costly improvements it was slowly taking over the role oI Rome`s
biggest harbor, which Puteoli had played Ior two centuries.
281
Besides, it is
absurd to imagine that Encolpius` visit to Rome belongs to some other occa-
sion than the present journey. Being a Massaliot youth, he would never have
leIt his city on any other occasion. The Saturnalia in Rome, just like Baiae, is
an ideal setting Ior the Satvrica. By taking the narrative straight Irom Mas-
salia to the topsy-turvy world oI a Roman Saturnalia Petronius would have
driven home the contrast and similarities oI the two places and provided an

279
Str. 5.246, Hor. Epod. 5.43, otiosa Neapolis; Verg. G. 4.363; Ov. Met. 15.712, in otia na-
tam Parthenopem; Stat. Silv. 3.5.85.
280
See Maiuri 1945, 244: '|L|a casa di Trimalchione || indubbiamente modellata pi sul tipo
della casa ellenistica che della casa romana. Ha un atriensis ma, in luogo dell` atrium, ha,
subito dopo la Iauce, una porticus nel cui mezzo una piscina, e un hospitium come le case
di Delo.
281
A voyage without a stop Irom Ostia to Massalia was quite possible (D.C. Hist. Rom.
60.21.3, '"R '"#"G,8=*". 0. #V ?*#&" 0'89H83 0. :"**",+"3 G"%8'-+*HK).
2. 1 SORTING THE FRAGMENTS 127
exceptional basis Ior the Satvrica`s theme oI Roman, and thereIore the
world`s, degeneration.
2.1.9 The Quartilla Episode
We now come to the beginning oI the extant text, but beIore we proceed
with our reconstruction some words must be said about the most adventurous
attempt to reorganize the Iragments. In 1930 the Neapolitan scholar Italo
Sgobbo hypothesized that the Quartilla episode was wrongly placed in our
tradition; that its proper place was beIore the initial encounter with Aga-
memnon at the school oI rhetoric.
282
This rearrangement would make the
Cena take over more or less directly Irom the initial scene in the Campanian
urbs Graeca (Sgobbo was convinced that it was Puteoli) and thus solve the
problem oI the apparent lack oI continuity Irom the invitation to dinner pro-
cured by Agamemnon (10.6) and the dinner at Trimalchio`s, which he as-
sumed were the same. The thesis involved locating the forum scene and the
pervigilium Priapi in Naples. Sgobbo`s thesis, however, creates more prob-
lems than it solves. The sacrum or sacellum Priapi is clearly not supposed to
be in Naples, but outside the Greek city. It is reIerred to in a manner that
shows it to be at some distance Irom the lodging house, where the second
encounter with the priestess initially takes place (16.4, ipsa venit; 17.6, huc
venisse). This distance is the distance between the Greek city and the shrine
outside the Crvpta. Puteoli and Neapolis (the strongest candidates Ior the
urbs Graeca) were connected by the Jia Puteolana and midway between
them (a short walk Irom either city) is the Pausilvpum promontory through
which the tunnel Crvpta Neapolitana runs.
283

We need to demonstrate that the phrase in 16.3 (Quartillae . sacrum
ante crvptam) reIers to the Crvpta Neapolitana, Ior there is the possibility
that crvpta (see OLD s.v.) can reIer to an underground room Ior religious
rites, a vault or a crypt. However, it is not the crvpta itselI that is reIerred to
as the shrine oI Priapus. The shrine sacrum or the diminutive sacellum Pria-
pi (16.3; 17.8, quod in sacello Priapi vidistis) is expressly said to be beIore,
ante, the crvpta, and so crvpta might well be written Crvpta. The phrase ante

282
Sgobbo 1930, 35461.
283
This tunnel was constructed by Cocceius, an architect oI Augustus, and is oIten reIerred to
in literary sources. It still exists, although many times restored and remodeled, and now
measures over 700 meters. On the Neapolitan side there was a necropolis by the road and
there somewhere stood Virgil`s tomb. For ancient reIerences and a map, cI. Beloch 1890,
83II.
2 STORY 128
Crvptam is a speciIication oI where the shrine oI Priapus is located and not a
reIerence to the shrine itselI. The singular in sacrum, 'shrine (here only in
the Sat.) and the alternative Iorm sacellum show that sacrum does not denote
'sacred rites or 'worship (something which could be perIormed outside a
shrine, iI the crvpta were the shrine itselI), Ior which Petronius always uses
the plural sacra.
284

The urbs Graeca is on one or the other side oI the important landmark
Crvpta Neapolitana and the boys came upon the cult somewhere outside the
city where a rusticus Iound Encolpius` shirt abandoned, in solitudine. II we
add to this the observation that the word crvpta is a rather obscure architec-
tural term, and that it is certain that the Crvpta Neapolitana did Ieature in an
episode oI the Satvrica,
285
which must necessarily be the extant episode in
the urbs Graeca, it seems that the phrase ante Crvptam is most naturally
taken as a reIerence to the Iamous tunnel between Neapolis and Puteoli.
Indeed, the reIerence in 16.3 to a crvpta would be highly problematic, iI the
Crvpta Neapolitana was not intended. Quartilla went to the city on the same
side oI the tunnel as her sacellum (cI. 17.5, nostra regio). Moreover, iI the
Greek city is Neapolis (truly the only urbs Graeca oI the possible candi-
dates) it would be redundant to reIer to the tunnel there in Iull as Crvpta
Neapolitana, since it had most likely already been mentioned as such (Fr.
XVI) and any mention oI it simply as the Crvpta would be immediately un-
derstood. The sacellum Priapi accordingly stood by the road beIore the en-
trance to this tunnel (whether any such place existed in reality does not mat-
ter) on the Neapolitan side.
Paratore did his best to reIute Sgobbo`s thesis three years aIter it was
Iirst presented;
286
nevertheless it still seemed plausible to Sullivan in 1968,
who Ielt that 'the Quartilla episode (1226.6) |was| very much out oI
place,
287
and is still regarded by Schmeling as the leading hypothesis.
288
In
my opinion, the diIIiculties caused by the traditional order oI the Iragments
have been greatly exaggerated. As I intend to show, we have not lost a day
or more somewhere in the Iragments. The Cena is supposed to take place on
the second, not the third day oI the boys` stay in the urbs Graeca. Let us

284
The early commentary oI Janus Souza, likewise, located the sacrum oI Priapus beIore the
Crvpta Neapolitana, and identiIied the urbs Graeca as Neapolis (Burman 1743, 2: 9I).
285
From the glossary oI Dionysius comes Fr. XVI, Petronius 'satis constaret eos nisi inclina-
tos non solere transire Crvptam Neapolitanam` ('Petronius writes: So it was quite clear
that they were wont to pass through the Crvpta Neapolitana only by crouching low`).
286
Paratore 1933, 1:155158.
287
Sullivan 1968, 35.
288
Schmeling 1996a, 463, claims that the Quartilla episode (1626) 'is generally believed to
be out oI place and to precede the opening scene with Agamemnon.
2. 1 SORTING THE FRAGMENTS 129
now go through this part oI the narrative in some detail to show that it is
aIter all quite possible to make sense oI the Iragments.
The encounter with Agamemnon at the school and the invitation to din-
ner (10.6) Iall in the morning
289
oI the Iirst day oI the young men`s stay in
the city, which is completely new to them (6.3, nec quod stabulum esset
sciebam; 11.1, cum errarem). Encolpius and Ascyltos meet Giton again at
the lodging house around lunch time (9.2, prandium), and the invitation to
dinner is scheduled that evening (10.6, hodie). Despite their quarrel, the boys
postpone the break-up oI their Iriendship until the day aIter (10.6, cras), so
as not to be deprived oI the dinner, since they are hungry (10.1, fame mori-
rer) and penniless (14.3). In the aIternoon (12.1, deficiente iam die; obscuri-
tas temporis) they go to the forum hoping to sell the stolen pallium (12.2).
By surprise they regain their lost treasure, sewn into the shirt oI Encolpius,
and can now at last buy something to eat. When they happily return to their
lodgings Giton prepares dinner Ior them and they stuII themselves with Iood
(16.1, nos implevimus cena). No sooner have they eaten (16.1, ut primum),
than the woman Irom the market scene just beIore (16.3, paulo ante)
290

shows up at their lodgings and identiIies herselI as the maid oI Quartilla.
Next, the priestess herselI, leaving the sacrum or sacellum oI Priapus, where
the young men surprised her on a previous night (17.7, nocte; 17.9, noctur-
nas religiones), arrives at their lodgings (16.4, ipsa venit in stabulum; 17.6,
huc venisse), in the same area (16.4, suam regionem; 17.5, nostra regio). At
Iirst she is polite and merely pleads with them to be silent about what hap-
pened and what they saw and to help her overcome her tertian Iever accord-
ing to the remedies revealed to her by Priapus in an incubational dream. But
when Encolpius shows himselI most ready to please her, the women`s mood
suddenly changes and they become threatening (18.719.1). Quartilla an-
nounces that she has taken control over the lodging house and is keeping out
all visitors (19.2, vetui) Ior the rest oI that day (19.2, hodie). The boys pre-
pare to Iight assuming their gender, iI nothing else, will secure them victory.
Something upsets their calculations (19.6).
291
Encolpius expects death, and
begs Ior a speedy execution (20.1). Psyche spreads a mat on the Iloor and
tries to stimulate his inguina to no eIIect (20.2). And here there seems to be a
change in Quartilla`s plan, perhaps because she Ieels that more drastic meas-
ures are needed to secure their cooperation. The boys` Ieet and hands are tied
(20.4), and thus they are apparently carried back to the scene oI the crime, to

289
The regular hours Ior school activities in the Greco-Roman world.
290
I retain the connecting phrase (16.3) considered by Mller to be a 'gloss; see my discus-
sion in section 1.1.1.
291
Most likely Quartilla`s auxiliary Iorces: 18.5, parata erat in crastinum turba.
2 STORY 130
the sacellum oI Priapus, where Quartilla lives, in the same manner that her
colleague Oenothea in Croton has her home (137.3, domicilium meum) in
theadmittedly less grandcella sacerdotis (134.3) oI the templum (136.7)
oI Priapus in Croton.
292
We note in a later reIerence to Pannychis the tempo-
ral primum, 'the Iirst time (25.2, ea ipsa quae primum cum Quartilla in
cellam venerat nostram),
293
which would not be needed had the company not
moved Irom the stabulum to the sacellum Priapi. Not much needs to be
missing here Ior describing the move to the other location, because Encol-
pius` narrative transitions are usually precipitous.
294

Suddenly, however, we are in the middle oI a sympotic setting and it
appears that we have missed some (erotic?) stories that were told (20.5; more
on this below), and thereIore a considerable amount oI text may be missing
in this most Iragmentary part oI the episode (19.621.3). The boys have now
leIt the stabulum (16.4) or deversorium (19.2), which is not alluded to again
in the episode. The sacellum, thanks to its location outside the city, would be
ideal Ior keeping hostages Ior there would be no one near to hear them
scream (21.1, volebamus miseri exclamare, sed nec in auxilio erat quis-
quam). Here Encolpius comes close to overdosing on the aphrodisiac sa-
tvrion,
295
whose properties make Quartilla sexy in his eyes (20.7). The boys
are then tortured, worked upon by a cinaedus and made to swear not to tell
the IrightIul secret oI the cult (21.3, tam horribile secretum). Next they are
rubbed down by masseurs and led into the adjacent cella (21.5, proximam
cellam), which is a luxurious triclinium (22.3, 25.3) with silverware (22.3)
and a Iamily oI servants (22.2). They are treated to Iine Iood and Falernian

292
Other reIerences to temples in Priapic sources include, sacellum (Priapea 14.2; Appendix
Jerg. Pr. 3.8), aedicla (CIL 5.3634), templa (CIL 5.2803).
293
Mller marks the clause with square brackets as an interpolation, but the text is no less
sound here than in 16.3 and other similar cross-reIerences, which are necessary to preclude
conIusion, when the narrative gets complicated.
294
One short sentence usually does the trick: 11.1, in cellulam redii; 12.1, veniebamus in
forum; 15.8, in deversorium praecipites abimus; 82.1, in publicum prosilio furentisque
more omnes circumeo porticus; 90.2, subsecutus fugientem ad litus perveni; 91.3, raptim-
que in hospitium meum pervolo; 116.1, momento temporis in montem sudantes con-
scendimus.
295
According to Pliny (Nat. 26.96I., 128) the Greek word satvrion is a general term Ior plants
with aphrodisiac properties. The roots or seeds oI these plants have phallic shapes or re-
semble testicles. One type with a testicle-shaped root causes erections iI taken in the milk oI
a Iarm-yard sheep, but makes erections subside iI taken in water. Another type arouses sex-
ual desire iI the root is merely held in the hand, but is more potent iI taken in dry wine. Yet
another can stimulate iI carried on one`s person. Pliny cites Theophrastus, a weighty author-
ity in botanical matters, Ior the anecdote that the touch oI an unspeciIied brand oI satvrion
provoked seventy successive copulations.
2. 1 SORTING THE FRAGMENTS 131
wine. When they are about to Iall asleep, Quartilla reminds them that they
are attending a pervigilium Ior Priapus and thus demands that they stay
awake (21.7). More torture and sexual exploitation ensue and Iinally the
whole household Ialls asleep out oI exhaustion (22.13).
Syrian burglars try to steal a silver lagoena (large jar with handles) and
unwittingly wake up the revelers. It is still night, and the butler adds more oil
to the dying lamps (22.6, tricliniarches [.] lucernis occidentibus oleum
infuderat). The party continues (23.1, refectum est convivium). Musical en-
tertainment is provided by a cvmbalistria (23.1). Quartilla invites the revel-
ers to begin drinking again and orders a 'bedclimber (24.1, embasicoetas)
Ior Encolpius, who knows this name Ior a speciIic type oI drinking cup.
296
It
is a prank in many ways resembling those played by Trimalchio on his
guests. Instead oI a drinking cup, a cinaedus enters who was 'obviously
worthy oI that house (23.2, et plane illa domo dignus). This person delivers
a poem in the Sotadean meter and then climbs into bed with Encolpius and
tries in vain to have sex with him. When Encolpius tearIully complains about
his treatment, Quartilla mocks him Ior his supposed lack oI urbanitas, i.e.,
Ior not knowing that a 'bedclimber is a cinaedus.
297
Encolpius now wishes
that the thing be given to Ascyltos as well. At this Giton cracks up, and
Quartilla seems to take an interest in him Ior the Iirst time.
298
To Encolpius`
dislike she Iondles his vasculum and plans to have it as an erotic appetizer
the day aIter, since she has already had something bigger that day: 'This will
make a good starter to rouse our desire tomorrow, since I`ve already had the
donkey today, I don`t want small rations (24.7, belle cras in promulside
libidinis nostrae militabit, hodie enim post asellum diaria non sumo; cI.
20.7).
299
The maid proposes to let the boy Giton devirginare (25.1) the
young girl Pannychis, and this depraved idea is immediately put into prac-
tice. Finally, aIter spending most oI the night at Quartilla`s, the boys some-

296
embasicoetas, (-ae) |Gk. 0j"*&'-+#". (0j"+34 '-+#K)| is an obscure Greek term Ior a
cup (Ath. 11.469a, #Q '",-=83-3 G-#X%&-3 0j"*&'-+#"3) which according to the name
seems to be intended Ior drinking in bed.
297
The OLD (s.v.) assumes, on the basis oI the Satvrica, that cinaedus is the primary meaning,
but the TLL (s.v.) correctly explains this sense as derived Irom a playIul interpretation oI
such a strange name Ior a drinking cup.
298
When Giton Iirst laughed (20.8), the virguncula was said to have put her arm around his
neck and given him 'numberless kisses. Giton characteristically made no attempt to resist
her amorous advances.
299
This apparent allusion to bestiality (OLD s.v. asellus 3) is interesting in the light oI the
desirability oI the donkey penis in the Greek and Latin Ass-Stories. Quartilla is clearly re-
Ierring to the two-legged donkey Encolpius (cI. Juv. 9.92, alium bipedem sibi quaerit asel-
lum). On the topos oI the desirability oI large penises, see Williams 1999, 8695.
2 STORY 132
how escape and make it to their beds at the lodging house where they spend
the rest oI this long and wakeIul night (26.6, abiecti in lectis sine metu reli-
quam exegimus noctem). They have provided the remedium they were asked
Ior and thus they can put aside Iear Ior the moment.
The mysterious third day arrives:

venerat iam tertius dies, id est expectatio liberae cenae, sed tot vulneri-
bus conIossis Iuga magis placebat quam quies. itaque cum maesti deli-
beraremus quonam genere praesentem evitaremus procellam, unus ser-
vus Agamemnonis interpellavit trepidantes et quid vos?` inquit nescitis,
hodie apud quem Iiat? Trimalchio, lautissimus homo horologium in tri-
clinio et bucinatorem habet subornatum, ut subinde sciat quantum de vita
perdiderit`. amicimur ergo diligenter obliti omnium malorum, et Gitona
libentissime servile oIIicium tuentem |usque hoc| iubemus in balneum
sequi. (Sat. 26)

Now arrived the third day, that is we were anticipating a dinner oI lib-
erty, but we were transIixed by so many wounds that escape seemed
more appealing than relaxation. So when in our dejection we were dis-
cussing a stratagem to avoid the approaching storm a slave oI Agamem-
non`s interrupted our trepidation, and asked: 'What`s wrong with you?
Don`t you know at whose place it is today? His name is Trimalchio, a
man oI exquisite reIinement, who keeps a water clock in his triclinium,
and has a trumpeter ready, so that he always knows Irom time to time
how much oI his liIe is lost. So Iorgetting all our evils we dressed with
care, and asked Giton who had until now most kindly played the role oI
our slave to Iollow us to the baths.

This is where H begins, almost certainly because a new book began here,
since it is common in long epic and prose narratives to use such temporal
shiIts to mark the breaks between books, and according to a Iragment Irom
the Satvrica preserved by Boethius such was Petronius` practice as well.
300


300
In Fr. V
b
|Mller|, taken Irom Boethius` commentary on Porphyry`s Isagoge, we Iind: et
ego. 'faciam` inquam 'libentissime. sed quoniam iam matutinus, ut ait Petronius, sol
tectis arrisit, surgamus, et si quid illud est, diligentiore postea consideratione tractabitur`
('And I said: I`ll do so very gladly. But since, as Petronius said, the morning sun has now
smiled on the rooftops, let us now rise Irom our discussion, and iI the matter deserves atten-
tion, it will be treated later with more care). The Iragment comes at the very end oI a book
and looks Iorward to a new beginning, which strongly suggests that the Petronian passage
being reIerred to had a similar place in the original and that at least one other book oI the
Satvrica (besides the one which began with the Cena) had the same Iormulaic opening. CI.
2. 1 SORTING THE FRAGMENTS 133
This is important Ior our present argument because it shows that the copyist
who is responsible Ior H, and who we may assume was working with the
complete text, did not begin with the words Jenerat iam tertius dies, id est
expectatio liberae cenae because he thought that libera cena reIerred to the
dinner at Trimalchio`s. He began so because the book containing the Cena
began with those words.
As CiaIIi has explained, the tertius dies can only have reIerence to the
much-Ieared tertian Iever oI Quartilla.
301
According to Celsus, tertian Iever
is so called because a second attack may be expected on the third day (expec-
tandus est dies tertius).
302
It is this tertian attack which Quartilla Iears, or
pretends to Iear (17.7, ipsa quidem illa nocte vexata tam periculoso inhorrui
frigore ut tertianae etiam impetum timeam |'That night I myselI Ielt uneasy
and shivered Irom so dangerous a chill that I even Ieared an attack oI tertian
Iever|), and it is the orgiastic remedy prescribed by Priapus in a dream
which provides her with a pretext Ior the detainment and sexual exploitation
oI the boys. The motiI oI the 'love Iever as a physical sickness, oI course, is
too well known Irom other ancient novels to require speciIic examples. The
night reIerred to in 17.7 (illa nocte) is thereIore the night beIore the Iirst day
in the urbs Graeca, since by inclusive reckoning the impetus oI the tertian
Iever would be expected to Iall on the second day aIter the Iever is Iirst Ielt.
This shows that we have not lost an entire day, or even more, somewhere in
the Iragments, as has oIten been assumed.
303
It also shows that the Cena

Apul. Met. 2.1, Ut primum nocte discussa sol novus diem fecit et somno simul emersus et
lectulo ('As soon as the new sun had dispelled the night and made day, I rose at once Irom
sleep and my bed); 3.1, Commodum punicantibus phaleris Aurora roseum quatiens lacer-
tum caelum inequitabat, et me securae quieti revulsum nox diei reddidit ('Just as Aurora
with her crimson disk brandished her rosy arm and began to drive her chariot across the
sky, I was harshly awoken Irom quiet sleep as night returned me to day); 7.1, Ut primum
tenbris abiectis dies inalbebat et candidum solis curriculum cuncta conlustrabat, quidam de
numero latronum peruenit ('As soon as darkness was dispelled with the dawn oI day and
the sun`s bright chariot shed light on all, there was a Iresh arrival); 10.1, Die sequenti [.]
('The Iollowing day .).
301
CiaIIi 1955a, 40.
302
Cels. 3.5.2, quamuis unam accessionem secuta integritas est, tamen quia tertiana timeri
potest, expectandus est dies tertius ('although return to good health Iollows a single onset
oI Iever, nevertheless, because there can be Iear oI tertian Iever, one must wait Ior the third
day |by inclusive reckoning|). The eight preserved books oI A. Cornelius Celsus, the en-
cyclopedist and contemporary oI Tiberius, are all on medicine and constitute the most im-
portant source Ior our knowledge oI ancient medicine aIter Hippocrates.
303
Which is not to say that much text may not be lost. The whole oI the Cena, a third oI the
extant work, covers only one dinner-party, and in the Quartilla episode there seems to have
been entertainment over dinner in the Iorm oI erotic story-telling, chatting and/or more per-
2 STORY 134
takes place on the second day oI the boys` stay in the urbs Graeca and not
on the third day.
The boys wake up late since they are soon oII to the baths. Although
they have been Iorced to provide Quartilla with a 'remedy to counter the
expected tertian attack, they do not know whether it has worked (unlikely
considering the sexual nature oI Quartilla`s Iever). Whatever libera cena was
intended to signiIy, it certainly does not mean 'a Iree dinner or 'a meal Iree
oI cost, because the Latin word liber did not have such modern economic
connotations.
304
More to the point would be a Iinal dinner at Quartilla`s, the
dinner oI their promised liberty, or a dinner which she has promised would
be Iree oI the captivity and harassment they had just suIIered (26.7, tot vul-
neribus confossis; cI. 22.1 and 2, tot malis).
305
We recall that Quartilla had
expressed plans Ior enjoying Giton the day aIter (24.7), an intention no
doubt particularly upsetting to Encolpius and enough to make him want to
escape at all costs. Thus the narrator is displaying his characteristic irony by
reIerring to the Ieared next encounter in Quartilla`s euphemistic terms. The
word quies, however, reIers to the waiting (expectatio) until the libera cena,
and not to the cena itselI which evokes the image oI stormy clouds gathering
on the horizon (26.8, praesentem procellam). The boys themselves are now
as apprehensive about the onset oI the tertianae impetus as the priestess her-
selI seemed, Ior they even deem preIerable the hazards oI Ileeing Irom the

Iormances like the poetry delivered by the cinaedus (20.5, iam deficiente fabularum con-
textu; note the plural in fabulae and the metaphor oI weaving in contextu, which indicate a
series oI stories or speeches). In similar situations at Trimalchio`s, and on the ship, stories,
speeches and perIormances are reIerred to as fabulae: 37.1, longe accersere fabulas coepi;
39.1, interpellavit tam dulces fabulas; 42.1, excepit Seleucus fabulae partem; 47.1, eius-
modi fabulae vibrabant; 59.3, scitis quam fabulam agant; 61.5, talem fabulam exorsus est;
110.6, ne sileret sine fabulis hilaritas; 113.1, risu excepere fabulam nautae. The gap be-
tween 20.4 (where the boys are tied up) and 20.5 (were the context oI fabulae is said to
have been broken) may be considerable. The presence oI a gap there supports my assump-
tion that the boys were tied up in order that they might be carried to another location and
that the transition Irom the lodging house to the quarters oI the priestess came here. Beyond
the transition, the accommodation in the new location and the early part oI the party are
thereIore missing.
304
Puccioni 1972, 32326, argues diIIerently and reads libergratuitus on the basis oI aedes
liberae (Liv. 30.17.14, 35.23.11). But as Puccioni himselI acknowledges this meaning oI
liber is nowhere else attested, does not stand with cena, and seems to belong to an archaic
diplomatic Iormula. For the normal idiom, cI. e.g. Apul. Met. 1.7, cena grata atque gra-
tuita.
305
Trimalchio says oI his slaves, whom he is promising Ireedom: 'they will soon drink the
water oI libertycito aquam liberam gustabunt (71.1). The promise oI liberty is a carrot
used to motivate slaves in their work. The boys` wretched Iortune oIten makes them as vul-
nerable as slaves. See also TLL, s.v. aqua.
2. 1 SORTING THE FRAGMENTS 135
city to a much needed rest (26.7, fuga magis placebat quam quies). But just
as their plans to go to the Iirst dinner-party (as they had promised Agamem-
non) were interrupted, they never attend this 'dinner oI liberty either, be-
cause while they are gloomily deliberating (26.8, cum maesti deliberamus)
what stratagem they can employ to shun the oncoming storm (26.8, quonam
genere |sc. evitandi| praesentem evitaremus procellam), they are distracted
by a character coming Irom the episode beIore the pervigilium Priapi who
oIIers them a diIIerent kind oI escape. The slave oI Agamemnon seems sur-
prised and irritated at their not knowing at whose place today`s party will be
(26.9, 'quid vos` inquit 'nescitis, hodie apud quem fiat?`), and he immedi-
ately sets their thoughts and actions on to a new course, making them Iorget
the threat Irom Quartilla (26.10, amicimur ergo diligenter obliti omnium
malorum). Agamemnon seems to have sent his slave to Ietch them, because
they didn`t show up the day beIore as they had promised. The rhetorician has
taken the boys under his protection. In Horatian terms they are his umbrae or
the companions oI a properly invited guest, Agamemnon himselI.
306
The
word hodie (26.9) clearly sets Trimalchio`s dinner apart Irom the other din-
ner oI the day beIore. From their reactions it is evident that the boys have
never heard oI Trimalchio beIore, and were not considering the possibility
that Agamemnon might procure another dinner invitation so soon, which
shows that even iI we agree with Sgobbo in transIering the Quartilla episode,
we simply cannot identiIy the Iirst invitation to dinner (10.6) with the dinner
at Trimalchio`s.
There may be some points in this interpretation that others would settle
diIIerently, but on the whole it shows that the episode is suIIiciently intelli-
gible. What is more important, it sits well where it is and its components
require no reordering. Considering the gaps in the text and the nightmare
quality oI the orgy, it would be unreasonable to expect complete intelligibil-
ity oI this part oI the narrative.

306
Hor. S. 2.8.212, cum Servilio Balatrone / Jibidius, quos Maecenas adduxerat umbras
('With Servilius Baltro was Vibidius, Maecenas had brought them as shadows); Ep.1.5.28,
locus est et pluribus umbris ('there was also room Ior many shadows).
2.2 Retrospective Soliloquies and Dialogues
2.2.1 Retrospective Surveys in Greek Erotic Fiction
In the oldest oI the extant Greek romances, those oI Chariton (Iirst century
C.E.) and Achilles Tatius (second century C.E.) a common motiI is the step-
ping aside oI the hero or heroine to utter an emotionally charged statement
containing a retrospective survey oI IateIul events thus Iar unIolded. In Cal-
lirho, such outbursts (mostly Callirhoe`s) take the Iorm oI soliloquies,
prayers and dialogues with other characters, and tend to Iocus on the turning
points oI the story (the Iestival oI Aphrodite and the wedding, Callirhoe`s
Scheintod, the robbing oI the tomb, the voyage to Ionia and her sale to the
new husband).
307
In Leucippe and Clitophon, most oI which is a personal
narrative like the Satvrica, this narrative Iigure is common as well. There the
enumeration oI Iormer evils leads up to the present moment which is re-
garded as the deIinitive culmination.
308
As a rule the recapitulations in the
Iully extant erotic Iictions are reliable and they summarize the events that
have been narrated earlier.
309
Nevertheless, an analysis oI the context oI
character statements necessarily Iorms a part oI the evaluation oI their use-
Iulness Ior plot reconstruction.
As Richard Heinze was the Iirst to show, the extant Satvrica, too, has a
Iew such retrospective passages,
310
which were surely important in the Iull-
text original, but have taken on added importance Ior us the readers oI the
Iragments because they help us in reconstructing the story in the missing

307
Chariton 1.8.34, 1.14.510, 3.8.9 (prayer to Aphrodite), 3.10.48, 4.1.1112, 4.3.10 (dia-
logue), 5.1.47, 5.5.24, 5.10.69, 6.2.811 (dialogue), 6.6.25, 7.5.25. See Hgg 1971,
262, Ior detailed analysis.
308
Ach. Tat. 3.16.35, 4.9.5, 5.7.89, 5.11.12, 5.25.28, 7.5. Hgg 1971, 283.
309
Heliodorus, however, lets his characters introduce red herrings to mislead other characters
oI the story, and even to mislead the reader. The Iigure is a Ieature oI Heliodorus` in medias
res narrative. He literally starts in the middle oI the story (the striking opening scene) and
only at the end oI the IiIth book does the reader know the events leading up to the begin-
ning. The missing inIormation is gradually Iilled in by the characters, especially by the
Egyptian priest Calasiris, who, with interruptions, narrates the best part oI the Iirst halI oI
the story (2.24 through 5.33). On his 'mendacity, see Winkler 1982. There is obviously lit-
tle room Ior this Iigure in chronologically linear narratives like the (extant) Satvrica.
310
Heinze 1899, 514.
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 137
parts oI the work. Although scholars have lately tried to minimize the sig-
niIicance oI these retrospective allusions, there is no reason to assume that
they are any less reliable in the Satvrica than in other ancient erotic Iictions.
Read careIully they indeed make good sense and complement the picture
that we have already constructed on the basis oI external Iragments and the
geography oI the story. The Iirst such passage comes early in the extant
Iragments and has the Iorm oI a dialogue, or more speciIically a shouting-
match, between Encolpius and his newly Iound, and soon to be lost, Iriend
Ascyltos. Here it is Ascyltos, rather than Encolpius himselI, who provides
inIormation about some oI the hero`s past crimes and humiliations. As we
shall see, Ascyltos merely knows about those recent adventures which they
have experienced together.
2.2.2 The Shouting Match
The reader will recall the incident as narrated. While Encolpius had been
listening to Agamemnon`s poetic rendering oI the ideal education, he had
suddenly noticed that Ascyltos had sneaked away. Ever IearIul oI rivals Ior
the pleasures aIIorded by Giton he had immediately set oII aIter his Iriend
but had not been able to Iind his way back to the lodging-house, being unIa-
miliar with the city. Eventually, he had been tricked into a brothel by 'an
urbane old lady, where he by chance had run into Ascyltos, who told him
that he too had been lost but had been led to the lupanar by a gentleman who
at Iirst seemed helpIul, but as it turned out had only wanted to hire him Ior
sex. AIter escaping Irom the brothel Encolpius Iinally Iinds his way to the
guesthouse when he glimpses Giton standing in a street. No sooner are the
'brothers reunited than Giton starts crying. Under pressure Irom Encolpius,
he reluctantly tells oI how Ascyltos had arrived in haste a little earlier and
had attempted to rape him. Hearing his worst suspicions conIirmed, Encol-
pius is enraged and points his Iingers into Ascyltos` Iace demanding an ex-
planation. Let us now print their important quarrel in Iull (9.610.3):

'quid dicis inquam 'muliebris patientiae scortum cuius ne spiritus purus
est? inhorrescere se Iinxit Ascyltos, mox sublatis Iortius manibus longe
maiore nisu clamavit: 'non taces inquit 'gladiator obscene, quem de
ruina harena dimisit? non taces, nocturne percussor, qui ne tum quidem,
cum Iortiter Iaceres, cum pura muliere pugnasti, cuius eadem ratione in
viridario Irater Iui qua nunc in deversorio puer est? 'subduxisti te in-
quam 'a praeceptoris colloquio. 'quid ego, homo stultissime, Iacere de-
2 STORY 138
bui cum Iame morerer? an videlicet audirem sententias, id est vitrea
Iracta et somniorum interpretamenta? multo me turpior es tu hercule, qui
ut Ioris cenares poetam laudasti.
itaque ex turpissima lite in risum diIIusi pacatius ad reliqua seces-
simus.

'So!I said'what have you to say Ior yourselI, you prostitute, sub-
missive as a woman, whose breath is not even pure?Ascyltos Iirst
pretended to be shocked, but then came on more aggressively raising his
Iists and yelling with considerably more vehemence: 'Won`t you shut
up, you Iilthy gladiator, whom the amphitheater dismissed when it col-
lapsed? Won`t you shut up, you night-time assassin, who then, when you
were at your strongest, weren`t even a match Ior a pure woman, whose
brother I was in the same sense in the garden that the boy is now in the
lodging-house?'You sneaked away Irom the colloquium with our in-
structorI added. 'What was I supposed to do, you stupidest oI all
men, when I was dying oI hunger? Should I perhaps have listened to his
rhetoric, nothing but broken bottles and dream interpretations? By Her-
cules, you`re much baser than I; you Ilatter the poet to get an invitation
to dinner.
So out oI this completely disgraceIul quarrel we dissolved into
laughter and backed oII Ior a more peaceIul remainder.

We can begin our analysis oI the passage by noting that Encolpius appears to
have concealed Irom Ascyltos the nature oI his love relationship with Giton,
perhaps as a part oI some ploy to keep the other Irom developing designs
against the boy`s chastity, but more likely because Ascyltos himselI had
been Encolpius` lover in the viridarium, as emerges Irom the passage. En-
colpius now demands an explanation Irom his Iriend as to why he, who be-
Iore has submitted to penetration and whose breath isn`t even pure (Irom
having engaged in fellatio), is now posing as a dominant male and trying to
rape Giton. AIter having made his young selI utter this accusation, the narra-
tor then supplies the inIormation that Ascyltos was not truly oIIended by the
accusation, although he Iound it convenient at the moment to Iake indigna-
tion (inhorrescere se finxit).
Accordingly, Ascyltos` even louder answer does not seek to oIIer a de-
Iense against the assault on his virility, but instead aims to drag Encolpius
down with him, and demonstrate that he is in no position to criticize, or even
to speak (note the repeated 'non taces`), since he too is seriously lacking in
virility. The logic oI Ascyltos` counterattack on Encolpius` virility is not the
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 139
accusation oI impotence that some have thought (that condition comes as a
great surprise to the characters as late as the Croton episode) but seems in-
stead to appeal to a more general deIinition oI the dominant male as some-
one who displays military prowess and has sex with a 'pure woman. An
obvious analogy is drawn between Iighting, or stabbing, and sexually pene-
trating. The phrase 'pure woman picks up the quality oI the 'pure breath,
spiritus purus, which is what Ascyltos supposedly lacked. A 'pure woman
seems thereIore to be mainly a woman who does not engage in fellatioand
by implication in other 'dirty sexual activities. In principle, then, the domi-
nant male here earns his reputation Ior sexual virility primarily by engaging
in vaginal intercourse.
311

In the sexual department, thereIore, Encolpius` dominance ('cum fortiter
faceres`), over Ascyltos in the viridarium and Giton in the deversorium, Iail
to qualiIy him as a dominant male, since buggery does not really register in
this respect. His exploits in the military department (as an obscene gladiator
rejected by the amphitheater and as a night-time assassin) are likewise Iound
to be very much lacking in manliness. By thus applying a positive standard
oI virility, instead oI the negative deIinition employed by Encolpius ('you
are sexually submissive and thereIore not virile), Ascyltos puts a diIIerent
rhetorical color on the Iacts oI the case and argues that Encolpius cannot
criticize another Ior lacking a virtue he does not possess himselI. Encolpius
says no more about the issue and thus implicitly acknowledges that his case
has been destroyed.
Encolpius, however, does not give up completely, and now accuses his
Iriend oI having deliberately sneaked away Irom their instructor. The charge
is that Ascyltos did so with the intention oI catching Giton alone in the
guesthouse to sexually abuse him. Ostensibly, this goes to show that he
would preIer the pleasures oI buggery to his own literary ediIication. Again
Ascyltos interrupts Encolpius beIore he can make an explicit case and claims
a legitimate reason Ior leaving: he was dying oI hunger. He then mounts a
counter-attack and reminds Encolpius oI his motives in staying to listen to
Agamemnon: he was dishonestly praising bad poetry in order to earn an

311
The use oI purus or impurus to denote this speciIic type oI deIilement is attested, beside the
above passage, in several poems oI Martial. Adams 1982, 199, provides Iurther examples.
Soverini 1976, 99107, rightly stresses the importance oI the interpretation oI purus, but his
argument is unnecessarily marred by his anachronistic insistence that the boys are accusing
each other oI 'homosexuality. In Iact, Encolpius accuses Ascyltos only oI not being domi-
nant, but submissive like a Iemale whore, to the point oI engaging in fellatio; cI. Williams
1999, 197II. on oral sex in Roman sources. In the Roman discourse on sexuality a distinc-
tion between 'homosexuality and 'heterosexuality did not have the Iundamental impor-
tance that it seems to have in modern sexual vocabulary; see Williams 1999, 4II.
2 STORY 140
invitation to dinner. At this blow, young Encolpius is outwitted and all he
can do is to laugh in embarrassment at having been seen through. Ascyltos,
who has won the argument with the help oI his quick wit, joins him in the
laughter.
It has been necessary to oIIer in some detail an explication oI the crude
logic oI this quarrel as a preliminary to using the inIormation it contains Ior
the reconstruction oI the story. II I am right that the phrases 'gladiator ob-
scene and 'nocturne percussor are used by Ascyltos to demonstrate Encol-
pius` lack oI military prowess, the Iact that Encolpius Iully accepts these
examples demands that there be a Iactual basis behind them. The narrative,
thereIore, at some point told oI Encolpius as some sort oI gladiator, 'an ob-
scene gladiator, who was rejected by the (h)arena or amphitheater in rela-
tion with its collapse (quem de ruina harena dimisit).
The words, de ruina, have unnecessarily troubled editors.
312
The preposi-
tion, de, can here either have a temporal ('Iollowing Irom, 'aIter), or a
causal ('on account oI) sense.
313
At least one collapsed amphitheater is
known Irom contemporary history. According to Suetonius (Tib. 40) twenty
thousand people died in the reign oI Tiberius when the amphitheater at Fide-
nae just north oI Rome collapsed during a gladiatorial show.
314
The disaster
entered the collective memory (Suet. Cal. 31). The gladiatorial ludus in
Pompeii is known to have been destroyed in an earthquake in 62 B.C.E.
315

That Ascyltos has knowledge oI the disgraceIul gladiatorial experience oI
Encolpius indicates that the incident belongs to that part oI the story in
which Ascyltos played a part. Accordingly, the episode was set in Campania.

312
The editio Tornaesiana alone oI the textual witnesses prints an asterisk between de and
ruina, and Bcheler concurs with it by indicating a lacuna. Nodot had tried to improve upon
the passage by supplying the words hospitis homicidam aIter quem; based on 81.3. Mller
1995 prints de ruina between daggers. Ernout, on the other hand, accepted the unproblem-
atic text oI codex Leidensis.
313
Burriss 1941, 276, provides a deIense Ior the Latinity oI the expression: '|t|he preposition
de is used in all periods oI Latin literature in the sense 'as a result oI, because oI.
314
Suet. Tib. 40, supra viginti hominum milia gladiatorio munere amphitheatri ruina perierant
('During a gladiatorial show more than twenty thousand men perished in the ruin oI the
amphitheater). Bagnani 1956, 25I., connected the incident to the story oI Petronius, but be-
cause he was arguing that Ascyltos` address oI Encolpius as gladiator obscene was merely
a general term oI abuse and not a reIerence to a lost episode in the work, he used the asso-
ciation with the disaster at Fidenae to hypothesize an unattested expression oI abuse,
gladiator Atilianus (the Ireedman Atilius produced the show Ior the sake oI proIit and his
parsimony was blamed Ior the accident). According to him then '|t|he general meaning oI
the remark oI Ascyltos would be: You lousy gladiator, whose only chance oI dismissal was
to be engaged by such a down-at-heel contractor that his shoddy amphitheater collapsed!`
315
Sen. Nat. 6.13, 27.12; Tac. Ann. 15.22.
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 141
As we shall see Iurther on there is indication that an earthquake caused the
collapse oI the building. Campania`s reputation as a seismically active re-
gion would provide a suIIicient excuse Ior presenting such a story, and even
iI no comparable incident had actually been known, it could nevertheless
sound credible as the sort oI thing that might well happen in that area. The
Fidenae incident and the one at Pompeii had proven that such disasters could
happen.
A Iragment oI Petronius might explain how exactly the collapse oI the
amphitheater caused or preceded the dismissal oI Encolpius. Fulgentius in
his treatise on the contents oI Virgil`s works makes a note oI an unusual
word, aumatium, meaning 'a private place in public as in theaters or in a
circus, and he goes on to quote this phrase Irom the original text oI the Sa-
tvrica: 'I hurled myselI into the latrine.
316
Such a turn oI events would have
given a typically humiliating spin to Encolpius` escape Irom a grave dan-
ger.
317
Its particular useIulness Ior the plot oI the Satvrica would be that oI
providing a sarcastic 'happy ending to the episode, i.e., the salvation oI the
hero by way oI the destruction oI others. This would be analogous to the
escape Irom Lichas and Tryphaena through the wreckage oI Lichas` ship and
his drowning (114.6), and partly like the escape Irom Trimalchio when the
vigiles break down the door oI his house (78.6I.), although on that occasion
casualties are not required Ior the boys` salvation.
We have argued above that the adjectives in the phrases, 'gladiator ob-
scene and 'nocturne percussor, have the Iunction oI degrading Encolpius`
exploits in the military department. Some scholars have argued that Ascyl-
tos` language has limited or no relevance at all to the protagonist`s hypo-
thetical 'criminal dossier.
318
The general tenor oI these objections relies on
the assumption that either the rhetorical and abusive character oI the boys`
quarrel, or the sexual content,
319
is so strong that it neutralizes any possible

316
Fr. XIII, aumatium dicitur locum secretum publicum sicut in theatris aut in circo. unde et
Petronius Arbiter ait 'in aumatium memet ipsum conieci` ('aumatium means a privy |la-
trine| in a public place` as in theaters or in the circus. WhereIore Petronius Arbiter says: I
hurled myselI into the latrine`). The word could be a corrupt Grecism Ior 2(#&-3, see
OLD s.v.
317
Sullivan 1968, 43, made the same association between Fr. XIII and the lost gladiatorial
episode.
318
The phrase is Irom the title oI Pack 1960.
319
Thus Mulroy 1970, 225, proposes a Iull-blown allegorical interpretation which is entirely
sexual: 'Ascyltos` insults are not meant literally, but rather as a Iigurative description oI
Encolpius` sex liIe. This is also suggested by the adjectives, obscene and nocturne. On this
interpretation, the ruina involved is a previous collapse oI Encolpius` virility. Gladiator and
percussor suggest erotic exertions. Cum fortiter faceres reIers to past heterosexual aIIairs,
2 STORY 142
retrospective allusions. Gilbert Bagnani so argues that Ascyltos is 'an artist
in abuse, |who| knows the Iorce oI alliteration and assonance, and who
practices mere abuse and elaborate name-calling; accordingly, 'one should
not inquire too closely as to the exact meaning |.| oI the expressions
used.
320
Walsh echoes Bagnani, when he speaks oI 'the scholastic nature oI
this exchange and 'the artiIicial nature oI the controversy. According to
him it is all 'a charade devoid oI any realistic reIerences.
321
Roger Pack
readily agrees with Bagnani 'that gladiator obscene is mere abuse, al-
though he does not think the case is thereby closed, and rightly points out
that the passages in 81.3 (they will be treated below) must be accounted Ior
beIore the prima facie content oI Ascyltos` phrases can be written oII.
322

Recently, Gareth Schmeling has once more urged the reader 'not to accept at
Iace value that which Encolpius hands him.
323
We disagree because on the
Iace oI it these statements are retrospective allusions, and since we have lost
so much text Irom the original Satvrica, it seems better in principle to as-
sume that they do indeed reIer to lost episodesuntil our attempts at recon-
struction clearly show that we have been misled. My Iindings indicate that a
reconstruction according to the apparently retrospective allusions in the Sa-
tvrica is less problematic than has oIten been thought.
It remains Ior us to explain what exactly Ascyltos could be reIerring to
by calling his Iriend 'an obscene gladiator.
324
The Iirst idea that comes to
mind is the association with Encolpius` most signiIicant attribute, his phal-
lus. Lucius, the narrator oI the Metamorphoses, while in the Iorm oI an ass

like the one with Tryphaena (113.7). The pura mulier represents the object oI competent
seduction and stands in contrast with women like Tryphaena.
320
Bagnani 1957, 245. The same scholar suggests a term, 'prosopographical, Ior the type oI
invective allegedly used by Ascyltos in the passage, and describes it as 'the unIlattering and
usually imaginary description oI the antagonist`s career. He provides an example Irom
Pompeian graIIiti, in which the object oI ridicule is Iirst reminded oI all the low proIessions
he has practiced and still practices, and then this is topped by the Iollowing statement: 'iI
you have licked cunt, you have tried everythingsi cunnum linxeris, consummaris omnia
(della Corte 1954, 329). The problem with this evidence is that we do not know whether the
receiver oI this insult had actually done what he is accused oI. For obvious reasons, insults
which allude to something real, even iI they put a malicious spin on the Iacts, are more bit-
ing.
321
Walsh 1970, 87.
322
Pack 1960, 31.
323
Schmeling 1994/5, 211.
324
The other phrase nocturne percussor I simply take to reIer to a murder committed by En-
colpius at night or at least in a secretive, non-virile manner. A sexual interpretation oI the
phrase would perhaps be possible in another context but here it would indicate that Encol-
pius was, contrary to what Ascyltos is arguing, a dominant male.
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 143
and thus no less a phallic Iigure than Encolpius, is sent into the arena to
copulate with a condemned woman (Met. 10.34). This Apuleian episode has
its direct counterpart in the Greek Ass-Story. However, Encolpius is not a
donkey and thereIore cannot have aIIorded the spectacle oI intercourse be-
tween an animal and a human being. A more likely and possible scenario
may be that he was made to Iight with a woman, an Amazon oI the arena,
and may have escaped Irom that encounter only because oI the collapse oI
the amphitheater.
325
Let us explore this Iurther.
In the extant Satvrica, Echion, a guest at Trimalchio`s dinner, in praising
the upcoming munus given by Titus, probably that year`s aedile, stresses that
there will be a woman Iighting Irom a chariot, mulier essedaria (45.4).
326

This passage, which is the only one on the subject in the extant text, also
presents a scathing criticism oI the previous year`s games in the amphithea-
ter, presented by one Norbanus, and especially oI the weaklings then Iighting
in the arena. In the Quartilla episode moreover Encolpius, Ascyltos and Gi-
ton line up Ior battle with Tryphaena, Psyche and Pannychis respectively
(19.46). Here the language and subject matter is military in a non-
metaphorical sense.
327
Encolpius is certainly not a soldierly type. On the
contrary, he strikes other characters in his narrative as having the qualities
and looks oI a prostitute,
328
and he was certainly no match Ior the soldier he
met on that night oI jealous rage when he intended to kill Ascyltos and Giton
(81.682.4). It is thereIore hard to imagine that any patron oI gladiatorial
games would dare to present someone as unsoldierly as Encolpius to do
combat with proIessional Iighters. II, however, he was made to Iight a
woman gladiator, something which may have suggested the imagery oI
Ascyltos` language ('ne quidem cum fortiter faceres cum pura muliere pug-
nasti`), although here the reIerence is to sexual intercourse rather than Iight-
ing in the literal sense, Encolpius, the delicate phallic youth, Iighting a much
stronger Iemale warrior in the amphitheater would have provided good enter-
tainment to please such audiences as Echion and his Iriends.

325
See Colin 1952, 31586, Ior a documentation oI 'gladiatorial perversions.
326
Statius reports that women and dwarIs were used as gladiators by Domitian in Rome (Silv.
1.6.51II.).
327
The phrase, contra nos, si nihil aliud, virilis sexus ('on the other hand, iI nothing else, we
had at least our male gender), especially, shows that the idea oI conIronting Encolpius with
a gladiatorial Amazon, to Iurther demonstrate his disqualiIication Irom the category oI the
dominant male, is present in the work.
328
7.2, 'hic [sc. in lupanari]` inquit 'debes habitare` ('here |in the brothel|,` she said, is
where you should live`); 126.1, 'quia nosti venerem tuam, superbiam captas vendisque
amplexus ('Because you are aware oI your sex appeal, you play hard to get, and sell your
Iavors`).
2 STORY 144
Admittedly the evidence concerning the exact practices oI ancient
gladiatorial games is limited, but as Steven Cerutti and L. Richardson have
attempted to show the Satvrica`s reIerence to 'obscenity in the amphithea-
ter may be more speciIic than has oIten been assumed. It appears that the
games in the amphitheater were not all about bloody killing but 'included
mock Iights and slapstick duels as comic relieI.
329
It seems that the gladiato-
rial schools had more than one division; beside the part where the really
virile gladiators were kept, there was also an 'obscene part, pars obscena
(Sen. Nat. 7.31.3), Ior those who were deliberately chosen Ior their lack oI
virilitas. The so-called 'OxIord Fragment oI Juvenal speaks with outrage oI
a certain type oI men, similar to cinaedi (O3, similesque cinaedis), who are
allowed to taint respectable households with obscene words and behavior
(O12, vivit luditque professus / obscenum), but would more appropriately
be kept separate in the gladiatorial trainer`s ludus and in a separate cella in
his jail (O113; at Juv. 6.365366). The logic oI this arrangement is hardly
obscure.

It is as the opposite oI this stereotype |oI the virile gladiator| that we
must see the retiarius tunicatus, a mock gladiatorial Iigure, oI equivocal
sex, regularly dressed in costume oI some sort, possibly usually as a
woman, and matched against a secutor or murmillo in a mock gladiato-
rial exhibition |.| we know that there were women gladiators, and the
sight oI a woman got up in body armor matched against a light-armed
man in drag would surely have been a spectacle the Roman crowd would
have relished.
330


Ordinarily, male gladiators Iought almost naked, apart Irom wearing the
subligaculum, which may explain the otherwise obscure reIerence which
Encolpius makes, that beyond being oI the male sex the boys` tunics were at
least girt higher than the women`s (19.5, sed et praecincti certe altius era-
mus) when they matched themselves with Quartilla and her maids.
The sexual innuendo traditionally associated with gladiators and mock-
gladiators Iits the passage in the Satvrica well and shows that the insulting
language used by Ascyltos would work at least as well, iI not better, iI there
had been a previous episode involving Encolpius` participation in an actual
perIormance in the amphitheater. In order to explain the complicated insults
exchanged between Encolpius and Ascyltos it is thereIore not necessary Ior
us to assume that the language is Iactually meaningless, i.e., a mere embel-

329
Cerutti and Richardson 1989, 589.
330
Cerutti and Richardson 1989, 593.
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 145
lishment oI style. In Iact, insults in general are clearly more to the point iI
there is some real or apparently real reIerence behind them. II Encolpius had
actually been an obscene gladiator in the arena, as Ascyltos says he had,
reminding him oI that Iact would be a very useIul way to silence him aIter
his accusation. Even Schmeling, though the uncompromising premise oI his
article is that all retrospective allusions in the extant Satvrica must be mere
embellishments oI language and Ascyltos` words here must be empty invec-
tive, reluctantly concedes the possibility that 'in a small town |.| Encolpius
had acted out the part oI a gladiatorwhether in a private house, garden, or
small harena.
331
In order to deal with the actual meaning oI the phrases
'nocturne percussor and 'viridarium, it will be necessary to introduce
another important passage, this time a genuine soliloquy, in which scholars
have generally recognized similarities to the phrases so Iar discussed.
2.2.3 Encolpius` Soliloquy
Let us Iirst brieIly rehearse the context. AIter two diIIicult nights in a row
(the pervigilium with Quartilla and the dinner with Trimalchio), Encolpius
Iinally had Giton all Ior himselI, but he had been too drunk to accomplish
more than kisses and soon Iell asleep (79.89). While he was sleeping,
Ascyltos had taken Giton away Irom him and carried him oII into his own
cubicle. When Encolpius woke up and discovered the truth, he had Iirst con-
sidered killing both oI them, but then decided against this violent plan and
merely roused Giton with a beating and demanded Ascyltos` immediate de-
parture. Already two days beIore he had wanted to break up the Iriendship
with Ascyltos in order to re-establish his old sexual relationship with Gi-
ton.
332
Ascyltos now agreed to leave and they had divided their spoils with-
out mistrust, but when it came to the boy Ascyltos demanded that they split
him as well. Encolpius Iirst thought he was joking, but when Ascyltos drew
his sword and threatened to cut oII his part oI Giton he prepared to meet him
in battle. Giton, however, by pleading with them and blaming himselI Ior all
that had happened, succeeded in averting this imminent 'Theban tragedy.
Next, Ascyltos had suggested they solve the crisis by allowing the boy him-

331
Schmeling 1994/5, 215. I Iail to see, however, on what evidence Schmeling can claim that
both the town and the harena were 'small.
332
10.7, hanc tam praecipitem divisionem libido faciebat, iam dudum enim amoliri cupiebam
custodem molestum, ut veterem cum Gitone meo rationem reducerem ('It was lust that
caused this very hurried separation; I had been eager Ior some time to get rid oI this annoy-
ing custodian, so that I could resume the old relationship with my Giton).
2 STORY 146
selI to choose whom to Iollow. Encolpius agreed, trusting that his old rela-
tionship with Giton would weigh heavily in the boy`s decision. Without even
giving the matter much thought, Giton had chosen Ascyltos and leIt with
him. Thunderstruck and abandoned by his two Iriends and lovers, Encolpius
collected his baggage and rented 'a secret place by the shore, where he
locked himselI in and 'Irequently lamented in this manner (81.36):

'ergo me non ruina terra potuit haurire? non iratum etiam innocentibus
mare? eIIugi iudicium, harenae imposui, hospitem occidi, ut inter au-
daciae nomina mendicus, exul, in deversorio Graecae urbis iacerem de-
sertus? et quis hanc mihi solitudinem imposuit? adulescens omni libidine
impurus et sua quoque conIessione dignus exilio, stupro liber, stupro in-
genuus, cuius anni ad tesseram venierunt, quem tamquam puellam con-
ducit etiam qui virum putavit. quid ille alter? qui tamquam die togae vi-
rilis stolam sumpsit, qui ne vir esset a matre persuasus est, qui opus mu-
liebre in ergastulo Iecit, qui postquam conturbavit et libidinis suae solum
vertit, reliquit veteris amicitiae nomen et, pro pudor, tamquam mulier se-
cutuleia unius noctis tactu omnia vendidit. iacent nunc amatores obligati
noctibus totis, et Iorsitan mutuis libidinibus attriti derident solitudinem
meam. sed non impune. nam aut vir ego liberque non sum, aut noxio san-
guine parentabo iniuriae meae.

'So why couldn`t the earth swallow me in the collapse, or the sea Ior that
matter, who gets angry even with innocent people? Have I escaped trial,
have I cheated the amphitheater, have I killed my host, so as to lie amid
suspicions oI delinquency,
333
poor as a beggar, an exile abandoned in the
lodging-house oI a Greek city? And who imposed on me this solitude?
An adolescent, deIiled by every sort oI libidinous act and according to
his own conIession worthy oI exile; Ireedman through buggery, Ireeborn
through buggery, whose youth is Ior sale at the toss oI a coin, who was
hired as a girl even by the man who knew him Ior a guy. And what about
that other one? who put on a woman`s dress as iI on the day oI assuming
a man`s toga, who was persuaded by his mother that he wasn`t a man,
who took on a Iemale Iunction in the jailhouse; who aIter having reshuI-
Iled and shiIted his sexual ground, betrayed the name oI old Iriendship,
andthe shame oI itlike a Iollowing girl traded all his assets Ior one
night oI groping. Now the lovers lie entire nights locked in each other`s

333
The word nomen here means 'suspicion or 'ground oI accusation, complaint; cI. Cic. de
Orat. 1.120, impudentiae nomen; Fam. 2.1.1, nomine neglegentiae; Suet. Tib. 3.2, levitatis
nomine.
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 147
embraces, and maybe, when they are worn out Irom their mutual satis-
Iaction, they mock my solitude. But not with impunity! For I`m not a
man and a Ireeborn citizen, iI I don`t avenge these insults by spilling the
guilty blood.

In this passage the same underlying erotic considerations as beIore are easily
recognizable. Although not named, we must take Ascyltos to be the Iirst one
mentioned and Giton the second. Ascyltos` deIilement through impure sex-
ual acts, especially fellatio, was the topic oI the earlier passage as well, and
he is again here described as a male prostitute. It is now Giton who is the
'Iemale in the ergastulum, 'the private jail, whereas earlier it was Ascyl-
tos who submitted to Encolpius in the viridarium, 'the garden, although his
sexual passivity was then likened to that oI Giton in the deversorium, 'the
lodging house. The Iirst two places mentioned, the ergastulum and the viri-
darium, are most likely related. As we shall see Iurther on, both probably
reIer to the villa oI cruel Lycurgus,
334
which is the kind oI place, a villa,
which we would typically Iind in Campania, and where we could easily Iind
both a garden and a private jail all in one. Giton and Encolpius go a long
way back together ('veteris amicitiae nomen`), whereas Ascyltos is a rival
who has recently entered the picture, and now has temporarily managed to
abduct the boy Irom the sorry protagonist. This erotic melodrama between
the three boys exhibits Iew mysteries that are not readily understood.
The initial death wish, however, has caused much speculation which
directly impinges upon how much material Ior reconstructing lost episodes
can be extracted Irom the passage. Here again scholars have argued that the
Iorm somehow overpowers and cancels out some oI the content. Ever since
Otto included the opening sentence oI Encolpius` remorseIul soliloquy
('ergo me non ruina terra potuit haurire?`) in his collection oI Roman
Sprichwrter,
335
the phrase has passed Ior 'a more or less stereotyped ex-
pression.
336
From this it has been thought to Iollow that the words, ruina
and terra, ought not be connected with the words de ruina harena (9.8) in

334
83.6, 'at ego in societatem recepi hospitem [sc. Ascvlton] Lvcurgo crudeliorem` ('I, on
the other hand, took into my companionship a Iriend |Ascyltos| more cruel than Lycur-
gus`); 117.3, placeret vestis, rapinae comes, et quicquid Lvcurgi villa grassantibus prae-
buisset ('so long as the garment, my associate in the robbery, was good Ior it, and whatever
the villa oI Lycurgus had yielded when we broke into it).
335
Otto 1890, 345, s.v. terra (3).
336
The quoted words are Bagnani`s 1956, 25. Klebs 1889, 626, however, was writing too early
to be inIluenced by Otto`s work. Paratore 1933, 1:148, consequently accuses him oI being
naive in interpreting the phrase 'ergo me non ruina terra potuit haurire?` as a reIerence to
a lost episode.
2 STORY 148
Ascyltos` earlier statement: 'you obscene gladiator, whom the amphitheater
dismissed when it collapsed (9.8, 'gladiator obscene, quem de ruina
harena dimisit`), despite the reappearance oI (h)arena only ten words later,
in 'harenae imposui` ('I have cheated the amphiteather), where the reIer-
ence, even according to Bagnani, is 'clearly autobiographical.
337
However,
to include the supposed stock-phrase in his collection, Otto was Iorced to
separate it Irom its parallel phrase which is syntactically dependent upon it:
'non iratum etiam innocentibus mare |sc. me potuit haurire?| The reason
why Otto did not include the whole expression is, oI course, that no stock
phrase is attested Ior the latter halInor is any Ior the Iirst halI, as we shall
see.
A close parallel Ior the latter halI is Iound in Achilles Tatius, likewise in
a soliloquy oI the younger selI oI the narrator Clitophon (3.10.6):

'(#K3 *-&, p H(,"**", #13 N(%&3 -,-/X*"83l A)-"+ *-$ #
)&,"3H%4G+l N%K*#-#A%" /A/-3". G%Q. -. JGA'#8&3"., zO. BC
*;*"*" O,,-3 JGA'#8&3".. 0)HF3K*". z93 J,o*#8=#-&. JG-H"3893.

'O sea, Ioolishly did we thank you Ior your mercy. Now I have only
blame Ior your philanthropy; more useIul were you to those you killed,
Ior by saving us you have killed us twice over. You meanly reIused us
our death, and saved us Ior bandits to kill.

The reIerence here is to a speciIic voyage which ended in shipwreck. In Iact,
in the case oI the Satvrica, too, scholars have not doubted that a speciIic
voyage Irom a lost part oI the work was reIerred to. We are, nevertheless,
asked to believe that in Encolpius` opening sentences he Iirst expresses (in a
rhetorical question) the wish that he had been swallowed up by the earth in a
collapsewithout reIerence to any speciIic incidentand then reIers to a
speciIic incident on sea in a parallel phrase using the same grammatical con-
struction. This seems like an overly complicated explanation.
II we take a better look at the 'stock phrase oI Otto, whose origins are
traced to Homer,
338
we note that it consists oI a wish Ior a supernatural re-
moval oI the hero Irom the Iace oI the earth beIore he should accept doing
something cowardly or shameIul, and is something oI a Iixture in the pep-

337
Bagnani 1956, 25. He accordingly concedes 'a condemnation to the arena, but then hurries
to add that Encolpius 'was never condemned to the arena, though he would have been had
he not made his escape.
338
Il. 4.182, #F#8 -& N(3-& 8e%89" NH;3 ('then let the broad earth gape Ior me); 17.416, J,,`
"e#-T /"9" A,"&3" GO*& N(3-& ('nay, even here let the black earth gape Ior us all).
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 149
talk oI epic warriors beIore a deadly conIrontation. Being a wish, Homeric
Greek uses the optative and the reIerence is always to the near Iuture. In
Latin, Otto quotes as the prime example oI the epic phrase the words spoken
by queen Dido to her sister Anna: 'I would sooner pray that the earth would
gape to its depths (Verg. Aen. 4.24, tellus optem prius ima dehiscat). The
other Latin examples provide several variations, but none deviates Irom the
basic wish Ior a supernatural intervention beIore some disgrace happens in
the Iutureexcept the Petronian passage.
339
For what Encolpius says is not
that he wishes Ior the earth to swallow him beIore he does something dis-
graceIul, but that he wishes that the earth had swallowed him in the collapse
instead oI saving him Ior Iurther disasters. The meaning is quite diIIerent.
We are thereIore entitled to wonder whether Otto was justiIied in including
Encolpius` wish that he had not been saved at some time in the past as an
example oI this epic stock-phrase. Schmeling adduces Giton`s words at 98.9,
utinam me solum inimicus ignis hauriret vel hibernum invaderet mare'I
wish some terrible Iire would burn me up, just me, or some Ireezing sea
would cover me, as a parallel to Encolpius` words at 81.3, and adds that van
Thiel called such phrases 'stereotype rhetorische Fragen, but neither state-
ment is a rhetorical question, and the more typical Iuture optative mode oI
Giton`s words obviously contrasts with the past perIect tense oI Encolpius`
statement.
340

At issue here is not young Encolpius` heroic sense oI shamehe is Ior
the most part shamelessbut, as in the Achilles Tatius passage above, the
pointlessness oI his past salvation in the light oI his continued misIortune.
Unlike the other examples oI the stock-phrase this one poses the question:
when, exactly, was there a particularly opportune moment in the past narra-
tive oI Encolpius Ior him to be 'swallowed up by the earth in the collapse?
We, oI course, know when he could have been swallowed up by the sea; this
was on his earlier voyage on the ship oI Lichas, a character who is later in
Iact swallowed up by the sea. A past incident involving a collapsing amphi-
theater would certainly provide us with a speciIic moment to serve as a paral-
lel Ior the other element. How else are we to account Ior the mysterious word
ruina? To wish that 'the earth swallow one up in the collapse is hardly a
Iormulaic turn oI phrase.
341
There may well be in Encolpius` words a play on

339
Otto lists one other occurrence in the past tense (Ov. Ep. 6.144, 'hiscere nempe tibi terra
roganda fuit`), but in the passage Hypsipyle is addressing Jason and recalling a speciIic
moment in the past when he ought to have wished Ior the earth to swallow him.
340
Schmeling 1994/5, 218.
341
Bagnani 1956, 25, claims that 'ruina terra haurire is a more or less stereotyped expression
equivalent to our own would the earth had swallowed me up`, but this modern English id-
2 STORY 150
the epic cliche, but the resemblance is no more than superIicial, and the ad-
aptation oI the collapsing amphitheater to the paradigm oI the gaping earth is
certainly Iorced.
342

Haurire seems a singularly inappropriate substitute Ior the Virgilian de-
hiscere, and is obviously better suited to mare in the latter halI oI the sen-
tence than terra. Since Ascyltos is supposed to know about the gladiatorial
adventure oI Encolpius and the collapsing amphitheater, this episode must
necessarily have come aIter the earlier voyage(s) on the ship oI Lichas, since
Ascyltos was not on the ship and so was introduced into the story aIter the
Iirst voyage(s) and beIore the last one. ThereIore we can conclude that in
Encolpius` soliloquy the collapse oI the amphitheater in an earthquake, and
the storm at sea, appear in an inverse temporal sequence. This may be due to
a deliberate aIIectation oI a similarity with the epic cliche. In this order En-
colpius can start with something which has a superIicial resemblance with
the stock phrase. But more likely the inverse temporal order is caused by the
Iact that the latter statement is a correctio oI the Iirst, a sort oI aIterthought:
'or the sea Ior that matter, who gets angry even with innocent people? I can
see no reason to preclude the possibility that these retrospective statements
concern episodes not Iound in the extant Iragmented text oI the Satvrica.
In the light oI his present abandonment, Encolpius muses that his two
miraculous salvations Irom the brink oI disaster, Iirst on sea and then on
land, have been completely in vain. But he also lists two more narrow es-
capes and one murder that have ultimately brought him nothing but loneli-
ness and excommunication. In a rhetorical question he asks: 'Have I escaped
trial, have I cheated the amphitheater, have I killed my host, so as to lie amid
suspicions oI delinquency, poor as a beggar, an exile abandoned in the lodg-
ing-house oI a Greek city? The answer required by this rhetorical question
is, oI course, negative. Moreover, since the opposite oI being leIt alone by
Ascyltos and Gitonthis is what provokes the state he is in and is empha-
sized in the passageis being with Ascyltos and Giton, Encolpius seems to
be claiming that the impudent acts he has committed were motivated by a
desire to be in their company. We know that he has been trying to get rid oI

iom, which I have been told does indeed exist, is hardly reliable evidence Ior Latin usage in
the Iirst century C.E.
342
I have Iound no examples oI ruina used with terra to denote the gaping oI the earth, as in
the epic stock-phrase. Ruina properly denotes 'Iall or 'collapse and is most common in
reIerences to collapsing buildings and ruined cities: Sat. 115.16, illum diis vota reddentem
penatium suorum ruina sepelit ('A man paying his vows to the gods is buried in the col-
lapse oI the building housing his ancestral gods); Suet. Tib. 40, supra viginti hominum
milia gladiatorio munere amphitheatri ruina perierant ('During a gladiatorial show more
than twenty thousand men perished in the ruin oI the amphitheater).
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 151
Ascyltos to be alone with Giton, and so it is not the loss oI Ascyltos which is
bothering him, but the loss oI Giton, his old passion, who has motivated all
the deeds listed by Encolpius.
This interpretation is less Iorced than Mulroy`s, who answers the ques-
tion: 'Did I escape judgment, cheat the arena, murder my host, so as to lie
with a reputation Ior audacity, a beggar, an exile, deserted in an inn in a
Greek city?, thus: 'No, I have committed no such crimes. I am the innocent
victim oI unrequited love.
343
Mulroy identiIies correctly that Encolpius is
primarily blaming Giton Ior what he has gone through. As the 'victim oI
unrequited love he may have some excuse, but this certainly does not show
that he didn`t do what he indeed claims to have done. The general thrust oI
Encolpius` story is his desire to have Giton all Ior himselI. In so doing he has
indirectly exiled himselI Irom his city, won enemies, and accumulated a
'criminal dossier. When the boy then chooses oI his own accord to leave
him Ior Ascyltos, Encolpius is leIt to Iace the consequences oI his actions
and the enormity oI his many-Iaceted isolation.
II we Iind it diIIicult to believe Encolpius capable oI killing somebody,
we should remember that he ends this very soliloquy by deciding to kill both
Giton and Ascyltos, which is the second time he entertains the idea (cI.
79.10). Encolpius` character certainly has the necessary violent streak, as is
evidenced Ior example in the scene where he takes pleasure in watching Eu-
molpus being beaten up by cooks and lodgers while he himselI strikes Giton
a blow on the head Ior wanting to open the door to rescue him (96.14). That
the comic plot oI the Satvrica would allow someone to actually die is clear
Irom Lichas` death at sea, while he pleads with Encolpius to return the sa-
cred objects that he had stolen (114.56). On the whole, the law oI the genre
is not at all averse to killing oII minor characters, especially those who pose
obstacles to the uniting oI the lovers, and could otherwise obstruct the con-
tinuation oI the adventurous plot. The Iully extant ancient erotic Iictions are
replete with violence and death. Moreover, as we have seen Irom the story oI
Hippothous and other examples (e.g., Thuc. 6.45; Parthenios 7), actually
assassinating one`s rivaland not just Iantasizing about itseems to be a
well known motiI in ancient erotic tales oI two boys in love.
2.2.4 Remapping the Plot
We now need to connect our Iindings here with our previous geographical
re-mapping oI the plot oI the Satvrica to establish where exactly on the route

343
Mulroy 1970, 255.
2 STORY 152
south it was that Encolpius played the gladiator, escaped trial, and killed his
host. According to the Iragment oI Servius (Fr. I), Encolpius was exiled, not
Ior any crime committed by him, but because he volunteered to play the
scapegoat in return Ior being Ied Ior a whole year at public expense. He
seems thereIore to have boarded the ship simply as a humiliated exile and as
such was merely en route to a new city where he could take up habitation,
most likely in Tarentum, the Iinal destination oI Lichas` ship. But Try-
phaena`s monopoly oI Giton becomes intolerable to him, and so he impul-
sively decides to Ilee Irom the ship. As we have seen earlier this is most
likely to have been in a harbor somewhere south oI Baiae and north oI Nea-
polis, since Tryphaena mentions Baiae in connection with the boys` escape
(104.2), and because the boys are thunderstruck at discovering that in Neapo-
lis they have again boarded Lichas` ship (100.3II.), which thereIore cannot
be the same city where they leIt it.
344
Since Puteoli had a big harbor, came
earlier on the route Irom Massalia than Neapolis, but later than Baiae, and
was close enough to Neapolis Ior the boys to have moved on Ioot, this is
probably where the boys ran away. Let us assume so, in any case, to see
whether the reconstruction will continue to make sense iI we base it on these
premises. AIter all we are engaged in assembling something like a jigsaw
puzzle, and success is measured by how many pieces Iind a comIortable and
unIorced place in the picture. The only premise that we can rely on with ab-
solute certainty is that the pieces Iit in somehow and that once there was a
clear and ordered picture.
AIter Ileeing Irom the ship oI Lichas in the harbor oI Puteoli in the com-
pany oI Giton and the captain`s wiIe, and aIter stealing the robe and sistrum
oI the ship`s tutelary deity, Isis, Encolpius Iinds himselI, in addition to being
an exile, a Iugitive with his enemies chasing aIter him. In his eIIorts to get
Giton away Irom Tryphaena, and in order to provide Ior the boy and himselI
on the run, he has embarked on a career oI petty criminality by robbing the
shrine on the ship. More than just theIt, this was technically also sacrilege.
On the whole, Encolpius is involved in the theIt oI no less than three gar-
ments, which need to be kept careIully separated. The Iirst is the 'the robe oI
the goddess, vestis divina (114.5), Irom the navigium oI Lichas, the second
is the 'small tunic Iull oI gold coins, tunicula . aureis plena (13.3), which
is the same as 'the garment, my associate in the robbery, vestis rapinae
comes (117.3), with the money stolen Irom the 'villa oI Lycurgus, Lvcurgi

344
BeIore Puteoli began to play an important commercial role, Naples had Iunctioned as the
main harbor oI Campania and the city even had its own Ileet (Polyb. 1.20.14, Liv. 35.16,
36.42, Appian. bell.civ. 1.89). There is, in other words, no evidence to preclude the docking
oI the ship oI Lichas in Neapolis as well as Puteoli.
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 153
villa (117.3), who seems to have been a miser who sewed his money into his
shirt.
345
The third is the 'stolen cloak, raptum latrocinio pallium (12.2),
taken Irom the 'shrine oI Priapus, sacellum Priapi (17.8), presided over by
Quartilla. Two are expensive on their own account and one is a rag contain-
ing hidden valuables.
Once oII the ship, Encolpius and Giton do not seem to have gotten Iar in
their Ilight. Their apprehension is alluded to in Tryphaena`s conjecture, when
she asks quid ergastulum intercepisset errantes'what jailhouse had inter-
cepted the Iugitives on their Ilight (105.11). Although the character Try-
phaena is not supposed to know at this point what really happened, she nev-
ertheless arrives at the truth Irom Ialse premises (she believes that the
painted letters on the boys` Ioreheads are truly branded marks), Ior we know
Irom a previous passage that Encolpius and Giton have in the meantime
been, indeed, in an ergastulum (81.5). Such Ielicitous guesses by characters
are possible Ior the simple reason that what is told in the story and how it is
told is entirely controlled by the narrator. The central story pattern also re-
quires that soon aIter Encolpius escaped Irom Lichas and Tryphaena he
should Iall into new captivity.
Perhaps Encolpius was caught trying to sell the vestis divina and the sis-
trum oI Isis, in the same way that he and Ascyltos are later caught trying to
sell the Priapic pallium, in a shady Iorum oI the urbs Graeca. Somehow or
other the stolen objects seem to have been discovered in the possession oI the
boys. In the name oI divine retribution, it would seem that objects stolen
Irom shrines and temples should normally lead to the capture oI the
thieves.
346
Fragment XII, tot regum manubies penes fugitivum repertae'so
many regal spoils discovered in the possession oI escapees, may well reIer
to this. The word manubies, or manubiae, which caught the eye oI Fulgen-
tius, and so led to the preservation oI Fragment XII, is, indeed, used by En-
colpius Ior the spoils that he and Ascyltos in good Iaith divide between
themselves when they part company (79.12, optima fide partiti manubias
sumus). When Encolpius again boards the ship oI Lichas, the captain at least

345
117.3, quicquid exigeret, dummodo placeret vestis, rapinae comes, et quicquid Lvcurgi villa
grassantibus praebuisset ('whatever he asked Ior, so long as the garment, my associate in
the robbery, was good Ior it, and whatever the villa oI Lycurgus had yielded when we broke
into it). One may wonder how it survived the shipwreck, but this may be asking Ior too
much verisimilitude Irom this adventurous plot. Encolpius has at least not complained oI
poverty since he recovered the shirt with the golden coins in the market scene, and although
Ascyltos would presumably have received halI, there could still be enough leIt to give Eu-
molpus the appearance oI aIIluence.
346
In Apuleus, Met. 9.10, the priests oI the Syrian goddess are captured as well.
2 STORY 154
seems to think that the boy still possesses the objects he stole Irom the
ship.
347

Foremost in Encolpius` enumeration oI his exploits, which we may take
to be in a temporal sequence,
348
comes his escape Irom trial (81.3, effugi
iudicium). The hyperbolic and accusatory style oI Fragment XII leads us to
think that these may be the words oI Encolpius` prosecutor at this trial. The
obvious exaggeration in the phrase 'so many regal spoils seems to be de-
signed Ior the sake oI rhetorical eIIect. The name oI the prosecutor may even
be provided by another oI Fulgentius` Iragments oI the Satvrica. According
to Fragment VIII, a ridiculing turn oI phrase was used by the narrator about a
certain advocate named Euscion, who 'was a Cerberus oI the courts
Cerberus forensis erat causidicus.
349
Not only does the barking oI Euscion`s
accusatory rhetoric justiIy the comparison to the hellish dog, Encolpius also
Iinds himselI on the Campanian coast, in the area oI Cumae, the Sibyl, and
lake Avernus, where the entrance to the underworld was supposedly located.
Likewise, in the dinner at Trimalchio`s scholars have noted signs that the
episode was intended as an allegorical journey to the underworld.
350
The
court case Encolpius had to Iace was probably similar to the one threatened
by the advocati in the market-scene in the extant Satvrica, i.e., Encolpius
was under 'suspicion oI theIt, latrocinii suspicio (15.3).
We do not think that Fragment XIV
351
belongs to this trial. The charge oI
dolus malus is diIIerent Irom a straightIorward accusation oI theIt. It is hard

347
114.45, Lichas trepidans ad me supinas porrigit manus et 'tu` inquit 'Encolpi, succurre
periclitantibus, id est vestem illam divinam et sistrumque redde navigio` ('Lichas trembled
and stretched out his hands to me imploringly, and said, Help us in our peril, Encolpius, re-
turn the divine robe and rattle to the ship`).
348
The sequence is the same in 9.9, 'non taces` inquit 'gladiator obscene, quem de ruina
harena dimisit? non taces, nocturne percussor; and 81.3, effugi iudicium, harena imposui,
hospitem occidi. As a rule such enumerations in ancient erotic Iiction Iollow the sequence
in which events have taken place. A similar enumeration in Chariton (Iirst or second cen-
tury C.E.) Iollows the correct temporal sequence (5.5: 'I have died and been buried; I have
been stolen Irom my tomb; I have been sold into slaveryand now, Fortune, on top oI that
I Iind myselI on trial! |Translation by Reardon, 1989|). I have explained above why En-
colpius may have inverted the sequence oI the voyage and the earthquake either in order to
begin with the earthquake and so adapt his past to the needs oI the epic stock-phrase, or
more likely because he corrects himselI to add the earlier event to the later.
349
Fulg. Expos. Jirg. Cont. 98I. Fulgentius, in Iact, assigns the words to Petronius, but as we
have seen it is common Ior commentators and later writers to reIer to the Iictional narrator
oI personal recollections as iI he were the same as the author.
350
E.g., Bodel 1994.
351
Isid. Etvm. 5.26.7, dolus est mentis calliditas, ab eo quod deludat. aliud enim agit, et aliud
simulat. Petronius aliter existimat dicens 'quid est, iudices, dolus? nimirum ubi aliquid fac-
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 155
to see how Encolpius, whose simplicitas makes him an ideal victim oI Iraud,
could be charged with dolus malus, and even harder to see him twisting the
meaning oI legal terms as is done by the speaker oI Fragment XIV. Never-
theless, Isidore does assign the utterance to Petronius and so it would be
consistent with our previous readings to reassign them to Encolpius. But it is
exceedingly diIIicult to see how they could be appropriate here to the person
oI Encolpius. It is Eumolpus who is the master oI Iraudulent schemes. He
interprets the accusation oI Encolpius, that he deliberately led the boys back
to their enemies, as that oI dolus malus (101.3) and the mimus he concocts
against the legacy-hunters oI Croton is a classic example oI Iraud.
352
It is
thereIore more plausible to assign this Iragment to a missing phrase Irom the
extant episode on the ship or to a court case in Croton brought against Eu-
molpus by Gorgias, whose name makes him an ideal player in a trial scene.
But I admit that this is a less than perIect solution.
How did Encolpius escape Irom being Iound guilty oI the charges
brought against him by Euscion? We know that he did, because he tells us
that he escaped the trial, effugi iudicium. According to his elegiac exposition
oI the nature oI contemporary justice (14.2), all juries are venal. Who bribed
the jury and why? Surely the boys did not have the means to do that, not
even with the 'regal spoils that were discovered in their possession and
must have been conIiscated. Their only asset, apart Irom their knowledge oI
letters, is usually their sex appeal. Scholarship can procure them dinner invi-
tations (10.46), but cannot motivate men oI means to undertake great ex-
penditure on their behalI (83.784.3). In the world oI the Satvrica the sexual
asset, however, seems to justiIy great expenditure by wealthy men. This is
evident Irom the career oI Trimalchio, who was his master`s Iavorite sex-
slave beIore inheriting his property (75.1176.2). II we consider Lichas`
sexual attraction to Encolpius, the eagerness oI the pater familias to pay cash
Ior having sex with Ascyltos in the lupanar-scene (8.24), and the ease with
which he Iinds a savior in the inIamous eques Romanus, when he is leIt
without clothes at the baths (92.10), there is nothing to stop us Irom assum-
ing that another gentleman oI a similar stature was motivated by the same
desire. In the Satvrica men and women alike are consistent and predictable in

tum est quod legi dolet. habetis dolum, accipite nunc malum` ('dolus means a deceiver`s
mental craItiness, derived Irom the Iact that he deludat (deceives), Ior he does one thing
while pretending to do another. Petronius thinks otherwise when he says: Judges, what
does dolus mean? OI course it is some deed that hurts (dolet) the law. There you have do-
lus; now listen to what malum means .`).
352
To give Ialse inIormation about one`s possessions in order to establish Ialse credit is a
prime example oI what the legal literature intends with the term dolus malus (dig. 4.3).
2 STORY 156
their vices. A prime candidate Ior this role is cruel Lycurgus, who is listed as
the main rival Ior Giton`s love next beIore Ascyltos (83.6), and so belongs to
this part oI the story, and whose villa (117.3), which at some point was bur-
glarized by the boys, as we have mentioned beIore, would Iit exceptionally
well in an episode set in Campania.
According to the pattern established by Vincenzo CiaIIi,
353
Encolpius`
escape Irom one captivity is usually mediated by the appearance on the scene
oI a good helper who no sooner than the escape has been accomplished turns
into a new menace, when he or she takes an interest in Giton. It is thereIore
likely that Lycurgus himselI saved Encolpius and Giton Irom justice by brib-
ing the jury, or by somehow or other exerting undue inIluence upon justice.
His reasons can have been either sexual interest in the boys, or possibly a
more Iinancially motivated desire to acquire cheap labor Ior his villa. An
unexpected verdict oI innocence, aIter a trial where Encolpius` guilt was
deIinitively established by the hostile Euscion, would have been in the right
spirit. When Ascyltos laments that no one knows them in the Greek city and
that thereIore they cannot expect to win a court case, it is likely that his
speculations about their situation are directly inIluenced by the experience oI
the Iirst trial narrated, although Ascyltos does not seem to have entered the
story until later.
354
It seems likely, however, that Ascyltos underwent a simi-
lar experience beIore or aIter Encolpius, that he too was a petty criminal
saved Irom justice by Lycurgus. Encolpius` remarks about the similarity oI
their Iortune would thus be justiIied (80.8).
Whatever happened in detail, it is likely that the appearance oI Lycurgus
indirectly resolved the trial episode by simply opening an escape route Ior
Encolpius. The boys would now have Iollowed their apparent savior to his
villa. Such a villa would, as we have said, typically have been Iound in Cam-
pania. This is aIter all the part oI Italy where wealthy Romans preIerred to
build their sumptuous Hellenistic palaces modestly called villae. Besides, as
we have also seen, Lycurgus immediately precedes Ascyltos in the remarka-
bly orderly succession oI rivals Ior the possession oI Giton.
355
This indicates
that the place where Encolpius and Ascyltos met and had their initial erotic
experience, the garden, viridarium (9.9), or the chain-gang or private jail,

353
CiaIIi 1955a, 65.
354
14.1, contra Ascvltos leges timebat et 'quis` aiebat 'hoc loco nos novit aut quis habebit
dicentibus fidem?` ('Ascyltos on the other hand Ieared the law and said Who in this place
knows us, or who will take our words Ior anything`).
355
83.6, 'at ego in societatem recepi hospitem [sc. Ascvlton] Lvcurgo crudeliorem` ('I, on
the other hand, took into my companionship a Iriend |Ascyltos| more cruel than Lycur-
gus`).
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 157
ergastulum (81.5), is indeed the same as the villa oI Lycurgus. In accordance
with CiaIIi`s model, and on the analogy oI Eumolpus` role in the escape
Irom Ascyltos and the Greek city,
356
Ascyltos is likely to have been instru-
mental in the scene where Encolpius and Giton escaped Irom Lycurgus.
This is what underscores the unwitting appropriateness oI Tryphaena`s
words when she asks what ergastulum had intercepted the boys on their
Ilight (105.11). An ergastulum ( Gr. 0%/"*#X%&-3) was a kind oI prison on
large private estates to which reIractory or unreliable individuals were sent
Ior work in chain-gangs, mostly slaves, but also Ireeborn men in debt (Sen.
Con. 10.4.18; Liv. 2.23.6). Suetonius reports in the Life of Tiberius (Tib. 8)
that one oI the Iirst responsibilities oI the Iuture emperor`s civil career was to
investigate and purge the private ergastula throughout Italy, which had
gained a bad reputation Ior holding captive not only travelers oI both Iree
and servile status, but also those whom dread oI military service had driven
to such places oI concealment. This was around 23 B.C.E. As the imperial
biographer explains in the Life of Augustus (32.1I.), the practice had survived
as a result oI the lawless habits oI the civil wars, but Octavian put an end to it
by having the ergastula inspected, and by stationing guards wherever it
seemed advisable Ior the protection oI travellers.
Other such practices were the Iormation oI associations under the title oI
a new guild (titulo collegi noui" to commit crimes oI every sort. Suetonius
says that Augustus at the same time had all guilds dissolved except the oldest
and most legitimate. The 'Caesar oI Apuleius` Metamorphoses does some-
thing similar when he bans the collegium oI the robber Haemus, and has the
band hunted out (Met. 7.7). It seems, indeed, that the boys and Eumolpus
Iorm just such a guild themselves.
357
Encolpius` choice oI adventures to tell
to his Roman audience seems to be guided by a certain tendency towards
sensationalism.
358
The abuse oI ergastula seems to have been one oI those
persistent rumors, like piracy and brigandage, which were rampant in the

356
92.4, timuique ne in contubernium recepissem [sc. Eumolpon] Ascvlti parem ('I Ieared that
I had taken into my Iriendship someone |Eumolpos| who was the equal oI Ascyltos).
357
117.5, in verba Eumolpi sacramentum iuravimus. uri, vinciri, verberari ferroque necari, et
quidquid aliud Eumolpus iussisset, tanquam legitimi gladiatores domino corpora animas-
que religiosissime addicimus ('we swore an oath dictated by Eumolpus, by which we sub-
mitted to being branded, Iettered, scourged, put to the sword, and whatever else as Eu-
molpus directed, like proIessional gladiators we most solemnly committed bodies and souls
to our master).
358
I believe it was Sandy 1979, 367, who introduced the excellent descriptive term 'sensa-
tional into the discussion oI Greco-Roman Iictions.
2 STORY 158
empire, and caused especially Ioreigners to Iear the highways, perhaps need-
lessly.
359

We could try to Iit Fragment XI into this context.
360
The mosquitoes
which troubled Encolpius` comrade (or himselI according to some manu-
script readings), and may have caused him to hallucinate, would be appropri-
ate in the viridarium. Apart Irom being Iorced to work as a slave, which he
cannot have done well because oI his inexperience oI work and youthIul
delicacy (102.12), Encolpius may have been incorporated into Lycurgus`
gladiatorial troupe. The name oI Lycurgus, namesake oI the legendary Spar-
tan lawgiver, may evoke Spartan harshness and military boot camps. It may
also evoke the myth oI the violent and drunken Thracian king who attacked
Dionysus and his party oI nurses (Il. 6.13040) and was subsequently im-
prisoned in a rocky cave (Soph. Ant. 95565), later to cut oII his Ioot (Hyg.
Fab. 132.), commit suicide (Hyg. Fab. 242.2), or be killed by horses (Apol-
lod. 3.5.1). Both connotations are ideal Ior someone engaged in the nasty
business oI training and dealing in gladiators (some oI whom were inciden-
tally called 'Thracians). The Iact that Lycurgus is the namesake oI the leg-
endary Spartan legislator and the mythical Thracian king makes his name no
less viable in this story than such names as Agamemnon, Menelaos, Circe or
Gorgias. As a real captive under the control oI Lycurgus, Encolpius could
have been sent into the arena as a member oI the pars obscena oI Lycurgus`
gladiatorial ludus to Iight, possibly, a woman who seemed likely to kill him.
He could then have escaped Irom the (h)arena, when it collapsed in an
earthquake, by throwing himselI into the latrine.
According to Ascyltos` accusation and Encolpius` own conIession, he
killed a man at night who was his host. In both cases, this comes in sequence
aIter he cheated the amphitheater (9.9; 81.3), and he certainly robbed Lycur-
gus` villa in the company oI someone else.
361
He speaks oI Ascyltos, too, as

359
On the topic oI piracy under the principate and the ideology oI imperial eradication, see
Braund 1993, 195212.
360
Fulg. serm. ant. 52 (p. 124 |Helm|), alucinare dicitur vana somniari, tractum ab alucitis,
quos nos conopes dicimus. sicut Petronius Arbiter ait 'nam contubernalem alucitae mo-
lestabant` |'contubernalem` is Bcheler`s emendation Ior a variety oI readings in the
manuscripts| ('alucinare means to have Ialse dreams`. It is derived Irom alucitae, which
we call mosquitoes`. Petronius Arbiter says: For the mosquitoes were aIIlicting my bed-
companion`).
361
117.3, atquin promitto quicquid exigeret, dummodo placeret vestis, rapinae comes, et quic-
quid Lvcurgi villa grassantibus praebuisset. nam nummos in praesentem usum deum ma-
trem pro fide sua reddituram ('Anyhow, I promise to provide whatever he asked Ior, so
long as the garment, my associate in the robbery, was good Ior it, and whatever the villa oI
Lycurgus had yielded when we broke into it).
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 159
being the co-possessor oI the money stolen, then lost and later Iound
again,
362
and oI his animosity toward this Lycurgus as analogous to that
which he directed at Ascyltos, because oI competition over Giton.
363
There-
Iore we may assume that in this lost episode Encolpius told oI a lusty aIIair
between himselI and Ascyltos in a chain-gang in the garden oI Lycurgus`
villa, while Giton was doing the 'work oI a woman in the private jail, opus
muliebre in ergastulo (81.5), under the command oI Lycurgus. AIter the
events in the amphitheater, the episode evidently ended with the murder oI
Lycurgus, the burglary oI his villa, and the Ilight oI Encolpius, Giton and
Ascyltos. The sequence oI competitors Ior Giton`s love seems to conIirm this
too. Ascyltos takes over Irom Lycurgus (83.5I.), just like Encolpius Iears that
Eumolpus is taking over Irom Ascyltos (92.4).
364

2.2.5 In the pinacotheca
The importance and logic oI the rivalry Ior possessing Giton in the Satvrica
is well explained in the scene set in the pinacotheca where Encolpius projects
his own anxiety on to the pictures oI ancient masters in the gallery. Just as in

362
13.23, Ascvltos postquam depositum esse inviolatum vidit . seduxit me paululum a turba
et 'scis,` inquit 'frater, rediise ad nos thesaurum de quo querebar? illa est tunicula adhuc,
ut apparet, intactis aureis plena` ('When Ascyltos saw that the deposit was untouched .
he took me a little aside Irom the crowd and said, Do you know, brother, that the treasure I
was grumbling at losing has come back to us? That is the little tunic and still, as it appears,
Iull oI the gold coins which haven`t been touched`).
363
83.56, 'et omnes fabulae quoque habuerunt sine aemulo complexus. at ego in societatem
excepi hospitem [sc. Ascvlton] Lvcurgo crudeliorem ('all the myths Ieature intercourse
with no rival. I, on the other hand, took into my companionship a Iriend |Ascyltos| more
cruel than Lycurgus`).
364
The mysterious Doris (126.18, itaque tunc primum Dorida vetus amator contempsi) is
probably not a woman with whom Encolpius had an amatory association in the past. The
word can either reIer to a marble statue oI the wiIe oI Nereus and the mother oI the Nereids,
since the protagonist has in the previous passage been comparing Circe`s beauty Iavorably
to marble sculptures; or, as seems more probable, it is a nom de guerre temporarily assumed
by Giton, like Polyaenus. The Ieminine gender may be explained by Giton`s role as a sexual
pathic, or possibly by reIerence to the Irequency oI male proper names oI a Iemale gram-
matical Iorm (e.g., Protis, Apellis, Thespis, Zenothemis, Taxaris, Charmis) attested on in-
scriptions Irom Massalia. In any case, Giton is properly the only person with respect to
whom Encolpius is a vetus amator (10.7, veterem cum Gitone meo rationem; 86.6, vetustis-
simam consuetudinem; 81.5, veteris amicitiae nomen), and the subject oI the Iollowing ne-
gotiation is exactly Encolpius` willingness to betray Giton (127.3, dono tibi fratrem meum)
with Circe, who has heard about their relationship, while there is no mention in the context
oI any past Iemale lover oI Encolpius.
2 STORY 160
the poem where Encolpius Iancies his Iortune to be comparable to that oI
ancient heroes struggling against divine wrath (139.2), the pictures oI mytho-
logical lovers here inspire in the protagonist a lament about his own inability
to have Giton completely Ior himselI (83):

'ergo amor etiam deos tangit. Iuppiter in caelo suo non invenit quod
eligeret, et peccaturus in terris nemini tamen iniuriam Iecit. Hylan Nym-
pha praedata imperasset amori sui, si venturum ad interdictum Herculem
credidisset. Apollo pueri umbram revocavit in Ilorem, et omnes Iabulae
quoque habuerunt sine aemulo complexus. at ego in societatem recepi
hospitem Lycurgo crudeliorem.

'So love touches the gods as well. Jupiter Iinds nothing to his taste in his
heavenly kingdom, but when he goes to earth, his sinIul intention does
injury to nobody. The nymph who raped Hylas would have controlled
her desire, had she thought that Hercules would come to Iorbid her.
Apollo makes the likeness oI his boy reappear in a Ilower, and all the
myths Ieature intercourse with no rival. I, on the other hand, took into
my companionship a Iriend more cruel than Lycurgus.

Encolpius imposes a positive interpretation oI the originally tragic stories oI
Hylas, Hyacinthus and Zeus` many loved ones in order to depict himselI as
the only tragic lover who cannot embrace his loved one without the interIer-
ence oI a rival, sine aemulo. We are reminded oI Hippothous` story, where
the same problem even causes the tragic death oI the loved one. From the
context we may determine that the crueler 'Iriend, hospes, is Ascyltos, and
the supposedly less cruel 'host, hospes, is the same Lycurgus who is the
owner oI the villa which the boys burglarize (117.3).
Taken together, all oI Encolpius` rivals Ior the possession oI Giton can
be ordered into a sequence, which provides another linear structure to sup-
plement the southward progression oI the ship oI Lichas, which has been
established according to the geographical map. II we allow Ior a possible
missing rival in Massalia itselI, that one would have been Iollowed by Try-
phaena, then possibly Hedyle, then certainly Lycurgus, Ascyltos, and Eu-
molpus himselI Ior a while, and Iinally Tryphaena again. These two linear
structures taken together, the geographical one and the amatory one, show
well how the plot was structured and episodes connected around the earlier
and the later voyage on the ship oI Lichas, with the great interlude in Cam-
pania coming in between, while Encolpius and Giton stay oII the ship. What
this shows primarily is that we had in the original Satvrica, not several radi-
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 161
cally distinct episodes, as many scholars have thought, but a novelistic struc-
ture, episodic oI course, but with a great deal oI continuity Irom one episode
to the next, Iully comparable to the plots that we Iind in the extant erotic
Iictions.
Apart Irom the interrupted and then resumed voyage on the ship oI
Lichas, one could see as symbols oI this continuity the three garments stolen
by Encolpius, which Iollow him Irom episode to episode and continue to
play a role in the plot. The vestis divina Irom the navigium oI Lichas was
stolen in the lost episode when the boys escaped Irom the ship and continues
to play a role on the ship aIter the boys board it again (114.5). It may also
have occurred in the trial scene (Fr. XII). The tunicula aureis plena is lost in
the episode prior to the beginning oI our text and then reappears in the mar-
ket scene (13.3), and again in Croton as the vestis rapinae comes (117.3),
with the money stolen Irom Lvcurgi villa, which takes us even Iurther back
to the escape Irom Lycurgus. The raptum latrocinio pallium (12.2), which
the boys try to sell in the market scene, turns out to be Irom the sacellum
Priapi (19.8), an episode beIore the boys entered the urbs Graeca. In that
scene Quartilla played a role. She disappears while the boys go to the school
oI rhetoric the morning aIter, and then reappears aIter the market scene to
Iorce the boys to remedy her tertian Iever. These connections and reappear-
ances between scenes make the episodes oI the story anything but discrete.
2.2.6 A Love Letter
To complete our analysis oI Encolpius` soliloquies and the material therein
Ior the reconstruction oI the plot oI his story, we must take a look at two
more passages. The Iormer oI the two is Encolpius` (alias Polyaenos`) short
letter, written on wax tablets, in response to the libidinous young matron
Circe, aIter having Iailed her miserably as lover. The Iollowing is the Iull
text oI the letter (130.16):

'Polyaenos Circae salutem. Iateor me, domina, saepe peccasse; nam et
homo sum et adhuc iuvenis. numquam tamen ante hunc diem usque ad
mortem deliqui. habes conIidentem reum: quicquid iusseris, merui. pro-
ditionem Ieci, hominem occidi, templum violavi: in haec Iacinora quaere
supplicium. sive occidere placet, cum Ierro meo venio, sive verberi-
bus contenta es, curro nudus ad dominam. illud unum memento, non me
sed instrumenta peccasse. paratus miles arma non habui. quis hoc tur-
baverit nescio. Iorsitan animus antecessit corporis moram, Iorsitan dum
2 STORY 162
omnia concupisco, voluptatem tempore consumpsi. non invenio quod
Ieci. paralysin tamen cavere iubes: tamquam ea maior Iieri possit quae
abstulit mihi per quod etiam te habere potui. summa tamen excusationis
meae haec est: placebo tibi, si me culpam emendare permiseris.

'Polyaenos greets Circe. I conIess, my mistress, that I have oIten done
wrong; Ior I am both human and still young. But never beIore this day
have I committed a mortal sin. You have a conIessed criminal at your
disposal: whatever punishment you order, I have earned it. I have be-
trayed; I have killed a man; I have violated a temple: punish me Ior these
crimes. II execution is Iitting, I`ll bring my steel; or iI you settle Ior the
lash, I`ll run naked to my mistress. Just remember one thing: It wasn`t
me, but my equipment, that sinned. A ready soldier, I had no arms. I
don`t know who caused the damage. Perhaps the spirit rushed ahead oI
the sluggish body; perhaps when I was aIlame with desire Ior all, I spent
my pleasure meanwhile. It`s beyond me what I did. You tell me, how-
ever, to beware oI paralysis: as iI there could be any more than the one
that took away Irom me that through which I could even have had you.
Ultimately, however, my excuse amounts to this: I shall please you, iI
you allow me to make up Ior my wrongdoing.

What stands out here are the three crimes to which Encolpius conIesses. Two
oI them at least seem Iamiliar: 'I have killed a man; I have violated a temple
(by temple robbery), but the way in which this inIormation about Encol-
pius` past is presented, and especially its irrelevance to the addressee, Circe,
combine to make it somewhat hard to explain. The Iirst conIession: 'I have
betrayed (a man), is no diIIerent Irom the other two, although less Iamiliar,
Ior it can easily reIer to the betrayal oI Lichas, who never seems to have
earned the treatment he got Irom Encolpius, since he was his savior and
lover until Encolpius robbed the shrine oI his ship and corrupted his wiIe.
Lichas` betrayal by the boys is in Iact reIerred to by Eumolpus as proditio
(107.6) in the apology he oIIers to the captain Ior the boys` behavior.
365
But
why conIess all this to Circe, who doesn`t know Lycurgus, Lichas and the
other people in Encolpius` past, and is interested only in having sex with
him?
The rhetorical structure oI the argument oIIered by Encolpius in apology
Ior his conduct is based on his being human and still young (nam et homo
sum et adhuc iuvenis) and thereIore having oIten sinned (fateor me, domina,

365
The word proditio never reIers to high treason in the Satvrica, but simply to giving oneselI
or someone else away or betraying a person`s trust (cI. 98.2, 125.3).
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 163
saepe peccasse), but never until this day, when he let down his mistress
sexually by being unable to have an erection, has he committed a sin that is
(almost) mortal (numquam tamen ante hunc diem usque ad mortem deliqui).
Then he enumerates his many previous sins and shows his willingness to
accept punishments Ior these (habes confidentem reum. quicquid iusseris,
merui . in haec facinora quaere supplicium. sive occidere placet, cum
ferro meo venio, sive verberibus contenta es, curro nudus ad dominam). The
list is the one we have discussed above (proditionem feci, hominem occidi,
templum violavi). The trivialization oI these serious crimes in comparison
with the truly trivial oIIense oI impotence is best taken as an amatory exag-
geration Ior the sake oI Ilattery oI a loved one. The truthIulness oI the crimes
conIessed to by Encolpius, and the predictable lack oI moral condemnation
on Circe`s side, is exactly the point oI this letter, which satirizes the pre-
sumed lawlessness oI certain matrons who lust aIter slaves, bare-legged
boys, gladiators, mule drivers, stage-actors, and condemned criminals
(126.510). Such women, according to the moralistic ethos underlying the
Satvrica, preIer the man they are having sex with to be all at once a traitor,
murderer, and a temple-robber, since in their eyes this makes him more ex-
citing as a lover. Encolpius can Ireely oIIer himselI up Ior punishment Ior
such crimes because he knows that Circe considers his sexual Iailure to be
his only serious oIIense.
366
Having said this, Encolpius can turn to the real
apology.
His real apology to Circe is based on the dissociation oI himselI Irom his
sexual organ, which has now become his sword or weapon, in a curious Iu-
sion oI elegiac and Priapic humor: paratus miles arma non habui (Ov. Am.
3.7.71I.; Priap. 9.13I.). Encolpius as a lover is also a soldier, and his sword
is his mentula; it is the part oI his body in which he expects to be an Achilles
(129.1), but something has gone wrong with the equipment, and Ior this, he
claims, he cannot be blamed. Perhaps he was just over-excited; in any case,
another try will certainly Iix the situation, he promises. Brought into context
with other retrospective statements in the Satvrica, our reading oI the letter
accounts suIIiciently Ior the conIessions oI Encolpius, and there is no need

366
Even iI what Encolpius is saying in habes confitentem reum etc. is 'treat me as iI I had
betrayed, killed, robbed or 'let`s pretend that I . and punish me accordingly, this does
not preclude that his supposedly Ialse conIessions are based on the truth. In Iact, iI no mur-
der took place, the audience oI the original Iull-text Satvrica might well get conIused at this
point, because Encolpius has certainly both betrayed Lichas and violated the temple oI Pri-
apus beIore the Crvpta Neapolitana. What would be the point oI mentioning here, Ior the
third time (cI. 9.7, 81.3), a murder that never took place, side by side with other crimes that
did?
2 STORY 164
to say with Schmeling that Encolpius suIIers Irom the modern psychological
condition oI 'conIession-compulsion, nor that his activity is comparable to
the selI-Ilagellation oI the priest oI the Syrian goddess in Apuleius (Met.
8.28).
367

2.2.7 A Prayer to Priapus
The last passage that we shall consider belongs to the same context oI Encol-
pius` impotence. We have seen many Iorms oI declarations, a dialogue, so-
liloquies and a letter. This time it is a poem and a prayer in one. Encolpius
has been desperately seeking a cause and remedy Ior his impotence. AIter
Iailing Circe Ior the Iirst time, he immediately claimed that he had been poi-
soned (128.2, veneficio contactus sum). He began to wonder whether he had
been permanently deprived oI physical pleasure (128.5). He Iinds himselI
impotent with Giton as well (128.7, 129.1). Circe blames his relationship
with Giton and suggests that he will regain his virility by abstaining Irom the
boy Ior three days (129.8). Encolpius asks Circe Ior another rendezvous
(130.6). Chrysis thinks it is just one more case oI the local witches having
put a spell on somebody, and thereIore a cure should readily be available
(129.1011). Encolpius has tried to cut down on luxury: he skips baths, uses
less oil, eats strong Iood, and drinks less wine; he takes a walk, and heeds
Circe`s words by going to bed without Giton (130.78). Chrysis had brought
a local witch who tried her hand at the task, and succeeded in giving Encol-
pius an erection (131.47). The cure, however, did not last and when he met
her he Iailed Circe again; she has had him whipped, spat on, and thrown out;
Chrysis has been Ilogged as well, and Proselenos, the witch, was thrown out
oI the house (132.25). Hiding in bed Encolpius has let loose his anger at the
'cause oI all evil, his penis. In Sotadean or cinaedic verse he described his
attempts to cut his member oII with an ax but it escaped into the wrinkles oI
its Ioreskin (132.8). Next he had harangued the thing, demanding an explana-
tion Irom the guilty party, but his mentula had only drooped in silence like
Dido in the underworld (132.911). AIterwards he Ielt momentarily ashamed
Ior having talked with his sexual organ, but he soon Iound justiIication Ior
this behavior in numerous literary precedents oI heroes talking to various
body parts and he had also been able to appeal to the supposed belieI oI Epi-
curus that sex was the 'end (#A,-.) oI liIe (132.1215). When he Iinished
delivering this declamation, he called Giton and demanded that the boy tell
him on his honor whether Ascyltos had stayed awake that night when he

367
Schmeling 1994/5, 221I.
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 165
abducted him, and whether he had wronged him. Giton swore that Ascyltos
did not use violence (133.2). We next Iind Encolpius kneeling on the thresh-
old in the shrine oI Priapus praying to the god in hexameters.
BeIore we consider the relationship between his impotence, his question
to Giton about Ascyltos` sexual advances, and Priapus, let us go through the
poetic prayer. AIter a very elaborate address to the god, Encolpius continues
his invocation (133.3):

'huc ades, o Bacchi tutor Dryadumque voluptas,
et timidas admitte preces. non sanguine tristi
perIusus venio, non templis impius hostis
admovi dextram, sed inops et rebus egenis
attritus Iacinus non toto corpore Ieci.
quisquis peccat inops, minor est reus. hac prece quaeso,
exonera mentem culpaeque ignosce minori,
et quandoque mihi Iortunae arriserit hora,
non sine honore tuum patiar decus. ibit ad aras,
sancte, tuas hircus, pecoris pater, ibit ad aras
corniger et querulae Ietus suis, hostia lactens.
spumabit pateris hornus liquor, et ter ovantem
circa delubrum gressum Ieret ebria pubes.

'Draw near, bastion oI Bacchus and darling oI the Dryads, hearken to
my humble prayers. I come not bathed in gloomy blood, nor have I
raised my right arm as Iaithless enemy against temples, but poor and
desperate in dire need and not with my whole body did I commit the
crime. He, who sins penniless, is less a Ielon. In prayer I beg you, ease
the mind and Iorgive a minor oIIense, and sometime when Iortune
smiles, I shall not let your glory be short oI honor; to your altar, holy
one, will go a horned goat, Iather oI the Ilock; to your altar will go the
Iarrow oI a grunting sow, a milky victim. New wine will Ioam in open
bowls, and drunken lads will dance and rejoice three times round your
shrine.

AIter having witnessed Encolpius` conIession to homicide and violating a
temple in the letter to Circe, and aIter what we know about his stealing sa-
cred objects Irom at least two shrines, we must ask whether there is a contra-
diction in his proclamation here oI innocence with respect to Priapus` tem-
ple: non sanguine tristi / perfusus venio, non templis impius hostis / admovi
dextram. What exactly is Encolpius saying? He approaches (venio) the
2 STORY 166
god/temple not as one steeped in blood, i.e. as a murderer seeking sanctuary,
or as one who has committed an act oI violence in a temple and hence sacri-
lege, but as one who committed a crime because oI poverty and only with a
part oI his body. In eIIect, he is saying: 'I have committed a crime against
you, Priapus, Ior which I plead (i) that I was driven by poverty, and (ii) that I
did it non toto corpore'not with all oI my body. Encolpius` hope is that
Priapus will Iorgive a minor oIIense, especially iI promised a rich sacriIice
in return Ior the clemency. The answer to the question above is thereIore that
we do not see any contradiction between the conIession in the present prayer
and the general crimes that Encolpius has already conIessed on other occa-
sions.
But what is the connection between Priapus and Encolpius` impotence?
We have brieIly discussed this problem in the previous section as part oI our
reading oI the Fragment oI Sidonius, and again as part oI our reconstruction
oI the Quartilla episode, or episodes, just beIore and aIter the boys enter the
urbs Graeca. Now we must review the evidence in the light oI our reading oI
the conIessional passages above.
The poem and prayer to Priapus indicates that Encolpius now derives the
cause oI his inability to have an erection Irom some oIIense, facinus, against
Priapus. As we have seen, it is because oI his continuing impotence that he
later claims that he, like other heroes, is haunted by the wrath oI a god, in
this case oI the garden deity Priapus (139.2). The alienated talk about a part,
pars, oI his body as iI it were a separate entity and not integral to him is
genuine Priapic humor, which is common in the extant corpus oI Priapic
poetry. Encolpius` excuse to Priapus in the prayer is the same that he gave
Circe: illud unum memento, non me sed instrumenta peccasse (130.4), in
both cases he blames illa pars corporis (129.1) where he once was an Achil-
les. Circe had asked him 'Well, do you come whole today? (131.10, 'ec-
quid hodie totus venisti?`), and the narrator maliciously describes how he
went toto corpore (131.11) into her embrace and enjoyed her to the Iull but
only with his unbewitched kisses! His bewitched penis is still not Iunction-
ing. Encolpius` attention continues to be Iixed on his mentula, his omnium
malorum causa (132.7). And he addresses a schizophrenic blame speech to it
as iI it were a separate individual Irom himselI (132.12, cum ea parte cor-
poris verba contulerim).
There is no room Ior doubt, then, that the facinus he conIesses to having
committed with only a part oI his body was committed with his membrum
virile. But when and where did Encolpius commit a crime against Priapus
only with his penis? Not in Croton, obviously, because the last thing Encol-
pius did beIore praying to Priapus, and what Iurnishes him with prooI oI
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 167
what is wrong, was to inquire Irom Giton about how Ascyltos had behaved
sexually, i.e., whether Ascyltos too was impotent, when he abducted Giton
Irom his bed that night aIter the party at Trimalchio`s.
368
He gets an answer
Irom Giton which is highly unreliable because Giton constantly Iears Encol-
pius` jealousy, but this is less important Ior the reconstruction than the Iact
that Encolpius himselI traces his impotence back to the urbs Graeca. Now,
what happened to both Encolpius and Ascyltos in the Greek city, which in-
volved a crime against Priapus that could have led to their impotence? The
answer to the question is clear. The boys disturbed the nocturnal rites oI
Quartilla`s Priapic sacrum ante Crvptam (16.3), and stole a pallium Irom the
members oI the cult. They were subsequently punished by Quartilla, who
came to their lodgings and sexually abused them aIter making them drink the
aphrodisiac satvrion. As we recall, they were supposed to go back to her
place the night aIter, but instead went to Trimalchio`s party. This is Encol-
pius` one and only prior crime against Priapus in the extant text.
As Christopher Faraone has shown, the satvrion that Encolpius drank so
much oI at Quartilla`s party (21.1), like other such drugs, was well known in
antiquity Ior its double eIIects: It could both cause an erection and act as
poison and render men impotent. We have already come across Pliny`s
statement (Nat. 6.96I. and 128) that a certain type oI satvrion with a testicle-
shaped root causes erections iI taken in the milk oI a Iarmyard sheep, but
makes erections subside iI taken in water.
369
Faraone adds an interesting
example Irom Achilles Tatius (4.15), where an Egyptian soldier lusting aIter
Leucippe lets his servant mix an aphrodisiac into her wine. But the servant
Iorgets to dilute the potion and Leucippe becomes mysteriously mad until
Clitophon discovers the cause oI her delirium and the antidote is adminis-
tered. The type oI liquid an ancient aphrodisiac was mixed in and the quan-

368
133.12, Gitona voco, et 'narra mihi` inquiam 'frater, sed tua fide. ea nocte, qua te mihi
Ascvltos subduxit, usque in iniuriam vigilavit an contentus fuit vidua pudicaque nocte?`
tetigit puer oculos suos conceptissimisque iuravit verbis sibi ab Ascvlto nullam vim factam
('I call Giton, and say: Tell me brother, but upon your honor: that night, when Ascyltos
took you away Irom me, did he stay awake until he had wronged you, or was he content
with spending a chaste night all alone?` The boy touched his eyes and swore in the most
precise words that he had suIIered no violence Irom Ascyltos).
369
Faraone 1990, 115, adds the inIormation that 'Mandrake, Ior instance, was used both as an
aphrodisiac (Theophr. HP 9.9.1) and as a narcotic to paralyze an enemy (Plato, Rep.,
6.488c), or to treat insomnia (Arist. PA 456b31). Theophrastus reports that the roots oI the
orchis and another unnamed Plataean drug can both encourage and suppress sexual desire
(HP 9.18.35). In the latter case, impotence could allegedly be extended as long as three
months and could be used to discipline and manipulate servants. This correlation between
debilitating narcotic and philtra is underscored in Plutarch`s advice to young brides not to
use such aphrodisiacs on their husbands (Mor., 139a).
2 STORY 168
tity consumed was evidently believed to be oI great importance Ior the
drug`s eIIect. SigniIicantly, Encolpius drank all the satvrion at Quartilla`s,
and later does not merely blame the wrath oI Priapus Ior his impotence, but
also claims that he was magically poisoned.
370
As is oIten the case in ancient
magic, the cause is over-determined in that there is both a pragmatic and a
religious explanation.
But how did Encolpius commit a facinus with only that part oI his body
which later is so disgraceIully incapable oI perIorming? At the time it evi-
dently did perIorm, and the instrument was thereIore in the erect position.
Priapus is a phallic god and his erection is his iconographic emblem. In Aris-
tophanes` Lvsistrata (982) a man with an erection is mistaken Ior Konisalos,
a phallic creature associated with Priapus, in a joke which resembles many in
the Satvrica: 'Are you a man or a Priapus?
371
II we analyze the reIerences
to the alleged facinus against Priapus, there was some urbanitas or practical
joke involved;
372
the impudent robbery is said to be more than Iabulous;
373

and a point is made oI the Iacility with which it is possible to encounter a
god in that region, as iI to provide an excuse Ior how easily Quartilla and her
acolytes were taken in by the boys.
374
Although Quartilla insists that the
facinus, or scelus, committed by Encolpius and Ascyltos cannot in principle
be expiated, she is also aware that the strange ritual which the boys, who
were driven by youthIul horseplay, saw in the temple oI Priapus could be

370
138.7, 'forsitan rediret hoc corpus ad vires et resipiscerent partes veneficio, credo, sopi-
tae ('maybe this body would come back to its strength, and the parts that were drowsed
with poison, as I believe, might be themselves again`). It is true that the word veneficium
also means 'the use oI magical arts, sorcery (OLD s.v. 1) as well as 'poison, but in this
case the two senses are not separable.
371
Ar. Lvs. 982, *= B` 8r #+; GF#8%` P3H%4G-. > '-3+*",-.; For the idea cI. 92.9, 'habet enim
inguinum pondus tam grande, ut ipsum hominem laciniam fascini crederes ('his genitals
hung down with such massive weight that you`d have thought the man himselI was a mere
appendage to his prick`). For the Iigure oI speech, cI. Sat. 38.15, phantasia, non homo;
43.3, discordia non homo; 44.6, piper non homo; 58.13, mufrius non magister; 74.13, codex
non mulier; 134.9, lorum in aqua non inguina.
372
16.4, 'nec accusat errorem vestrum [sc. Quartilla] nec punit, immo potius miratur, quis
deus iuvenes tam urbanos in suam regionem detulerit` ('Quartilla neither accuses nor pun-
ishes you Ior your mistake, instead she wonders what god has brought such urbane youths
into her region`).
373
17.4, 'quaenam est` inquit [sc. Quartilla] 'haec audacia, aut ubi fabulas etiam anteces-
sura latrocinia didicistis?` ('What is this audacity,` said Quartilla, that is, where did you
learn to outrival the robbers oI Iables?`).
374
17.46, 'utique nostra regio tam praesentibus plena est numinibus ut facilius possis deum
quam hominem invenire` ('our region is so Iull oI divine presence that you are more likely
to run into a god than a man`).
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 169
seen as a risible joke by the uninitiated;
375
at the time Encolpius may have
undressed and leIt his old tunic Iull oI gold lying in a deserted place, and
was perhaps clothed in a religious garment, Ior as he tells us in the market
episode, he lost his shirt and acquired a splendid pallium with an intricate
design.
376
The supposition that best accounts Ior all oI these bizarre reIer-
ences is the impersonation oI Priapus by the naked Encolpius by virtue oI his
characteristic phallic looks. The reIerence oI Sidonius Apollinaris to Encol-
pius (alias Arbiter) as Hellespontiaco parem Priapo directly connects the
phallic looks oI the protagonist to Priapus and thus Iurther underscores our
reconstruction oI the lost Iirst Quartilla episode.
377

2.2.8 Encolpius Equal to a God?
II we were to look Ior a 'comparison text to aid us in the reconstruction oI
this obscure episode, it would soon become apparent that the motiI oI a man
impersonating a god to exploit the superstition oI some simple person is
common enough in ancient literature. Josephus (AJ 18.66) tells a story Irom
the reign oI Tiberius oI a young knight, Decius Mundus, who Iell in love
with an unassailably virtuous matron, Paulina. When bribing her didn`t
work, he decided to kill himselI. But his Ireedwoman, Ida, discovered a way
to save him. She promised money to the priests oI the temple oI Isis in
Rome, and one oI them went to Paulina to say that the god had Iallen in love
with her and wished to see her. She became proud at the news and with her
husband`s permission went to the temple. When she had dined and the doors
were closed and the lamps removed, in place oI Anubis, Mundus stepped
Iorth and had sex with her the whole night. Later Paulina told her husband

375
17.6, 'imprudentes enim, ut adhuc puto, admisistis inexpiabile scelus` ('I still believe that
you committed your inexpiable crime unwittingly`); 17.7, 'ne scilicet iuvenili impulsi li-
centia quod in sacello Priapi vidistis, vulgetis deorum consilia proferatis in populum` ('I
am aIraid that youthIul license will lead you to broadcast what you saw in the chapel oI Pri-
apus, and reveal the gods` counsels to the mob`); 17.8, 'petoque et oro ne nocturnas re-
ligiones iocum risumque faciatis` ('I beg and pray you not to make a laughingstock oI our
nocturnal worship`).
376
12.5, videbatur ille [sc. rusticus] mihi esse qui tuniculam in solitudine invenerat ('that
rustic seemed to me to be the one who Iound the little tunic where it was abandoned); 12.2,
raptum latrocinio pallium . splendor vestis ('the stolen pallium . the splendor oI the
garment); 14.5, mulier. inspectis diligentius signis ('the woman . having careIully in-
spected the signs).
377
The line oI Sidonius is a pastiche oI 139.2 vv. 89, me quoque per terras, per cani Nereos
aequor / Hellespontiaci sequitur gravis ira Priapi, which as we saw above can only reIer to
the Quartilla episode, and its aItermath in Encolpius` impotence.
2 STORY 170
and women Iriends and all marveled at the incident. But Mundus conIronted
Paulina and told her the truth, and the lady told her husband. Tiberius had
both the priests and the Ireedwoman Ida cruciIied, but not Mundus himselI.
He was sent into exile, since he was considered to have been compelled to
the deed by the sheer Iorce oI his desire. The incident was soon worked into
a mime, Moechus Anubis (Tert. Apol. 15).
The tenth letter oI Pseudo-Aeschines provides Iurther examples oI the
motiI. The main story is set in Ilion. The lad Cnemon hides by the banks oI
Scamander, while the local virgins, and especially Callirhoe, perIorm a ritual
in which they wash in the river and address the river god with these words:
'Take, Scamander, my maidenhood. When Callirhoe pronounces the Ior-
mula Cnemon jumps Iorth Irom the bushes saying: 'I accept with pleasure,
and I take Callirhoebeing Scamander, I shall do many good things, and
then disappears again with the maiden. A Iew days later, Callirhoe recog-
nizes him in a procession and immediately throws herselI Ilat in reverence to
the god, and looking askance at her nurse she says: 'Behold the god Sca-
mander, to whom I gave my virginity. The older woman understood what
had happened and the story got out. Having related this story, Pseudo-
Aeschines then goes on to tell how Cnemon wasn`t aIraid but on the contrary
provided him with other examples oI similar deeds: many stories involving
gods impersonated, stories which told oI how Meander, Heracles, Apollo,
and Dionysus unjustly gained a reputation Ior being 'adulterers (-&N-+).
Relevant to Encolpius` postulated impersonation oI Priapus are what
appear to be similar perversions oI religious rituals in the Iragments oI Lol-
lianos` -&38&'&'( (POxv 1368) and the Iolaos Iragment (POxv 3010). In the
Iormer a group oI young men dress themselves in black or white garments to
play ghosts in preparation Ior what seems to be a raid by robbers. In the lat-
ter, according to the reconstruction oI E.R. Dodds,
378
a young man oI Iull
male potency attempts to gain access to a woman by impersonating a cas-
trated gallus or cinaedus. Such trickery seems to be generic to ancient erotic
Iictions, especially the sensational and criminal ones.
379

It seems thereIore plausible to assume that Encolpius` crime against Pria-
pus may derive Irom the lost incident at the Priapean temple presided over
by Quartilla in the vicinity oI Neapolis by the road to the Crvpta Neapoli-
tana. Accordingly, Encolpius would suspect that Quartilla in revenge se-
cretly poisoned him. He certainly does derive his impotence Irom the curse
oI somebody who wished harm on both himselI and Ascyltos. His inquiry
about Ascyltos` capacity Ior lovemaking would otherwise barely make

378
See Parsons 1974, 37.
379
See also Sandy 1979, 3745.
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 171
sense. Quartilla and her satvrion, however, oIIer a plausible explanation. The
dramatic irony in Encolpius` accepting as evidence oI Ascyltos` impotence
what Giton tells him adds a Iurther comic twist to the episode. Giton is, oI
course, likely to oIIer Ialse assurances only to calm Encolpius who is prone
to rabid jealousy. The wrath oI Priapus is thereIore likely to be no more than
Encolpius` Iantasy, which is not to say that it isn`t important Ior the story.
Besides, in this connection between the episodes beIore and in the urbs
Graeca we have another indication oI the lack oI discreteness oI the epi-
sodes, and a Iurther support Ior our thesis that in the Satvrica we are dealing
with a typical ancient novelistic plot.
2.2.9 Conclusions
To wrap up our reconstruction oI the Satvrica in this and the preceding chap-
ter it will be helpIul iI we summarize the Iour main principles on which our
restoration oI the story is based. The Iirst is geographical and pertains to the
movement oI the ship oI Lichas Irom Massalia, to Ostia, to Puteoli, to Nea-
polis, and Iinally to the bay oI Tarentum, where the ship and its captain meet
their destruction. The linear and regular progression oI the ship was evi-
dently the main organizing principle oI the plot oI the original Satvrica.
Where Encolpius & Co. go aIter Croton is impossible to say, but we are in-
clined to think that Trimalchio`s ambitions to cross over to AIrica Irom Nea-
polis without ever sailing along the conIines oI another man`s property
(48.3), and Eumolpus` posing as an AIrican landowner (117.8, 125.3, 141.1),
might be anticipatory indications in Encolpius` narrative that this is where
the plot is soon heading, when the extant text breaks oII. Perhaps the legacy-
hunters oI Croton decided to go with Eumolpus to inspect his supposed AIri-
can estates, when he didn`t die Irom the sicknesses that he was Ieigning.
More likely, Eumolpus Iades out (as others beIore), perhaps by Scheintod.
380

AIrica and especially Egypt provide the background Ior many an episode in
the extant erotic Iictions. Whatever happened next, Encolpius in any case
seems to tell the story in Rome itselI, just like Lucius in the Metamorphoses,
since his audience is so typically Roman. We should keep in mind that En-
colpius does not have to return to Massalia, any more than Hippothous or
Lucius in the Latin version oI the Ass-Story, return to their homes.
381


380
I see no grounds Ior believing with Schmeling 1992a that the plot is nearing closure in the
Croton episode.
381
The view that Encolpius goes to Lampsacus on the Hellespont is unIounded and derives
Irom a misunderstanding oI the role oI Priapus in the plot.
2 STORY 172
The second principle oI reconstruction is literary and relies on the con-
vention oI ancient narratives oI recapitulating prior happenings oI the story
in later episodes. The most inIormative oI these recapitulations regard En-
colpius` relationship with Ascyltos, who Iirst enters the story aIter Encolpius
and Giton leave the ship oI Lichas. We thus get recapitulations oI the events
that happened between the Iirst and the second voyages, and brieI allusions
to a trial, a stay in the private jail oI Lycurgus` villa, Encolpius` adventure in
the arena, and Iinally his murder oI the host, Lycurgus, and escape Irom the
villa/jail-house, in the company oI Ascyltos and his dear Giton.
A third principle regards the remarkable regularity oI the pattern, Iirst
observed by CiaIIi, oI the threats to Encolpius` love aIIair with the irresisti-
ble Giton. A third individual appears as a savior, only to turn into a rival and
thereIore an enemy oI the protagonist. Thus we can construct a sequence oI
such rivals, which agrees with the reconstruction based on the Iormer two
principles: Tryphaena, (Hedyle), Lycurgus, Ascyltos, Eumolpus, Tryphaena.
The key to the reconstruction is the realization that the driving Iorce oI the
plot is not so much the 'wrath oI Priapus as the jealousy oI Encolpius. His
possession oI Giton is constantly being threatened by rivals and outsiders
who Iancy the pretty boy as well.
The Iourth structural principle is the observation that the episodes oI the
Satvrica are not autonomous or discrete units, as is oIten assumed on the
basis oI more or less isolated readings oI individual parts oI the work,
382
but
no less an organizational whole than the plots oI the Iully extant ancient
Iictions. From the route oI the voyage, the recapitulations and the sequence
oI rivals which threaten Encolpius, it is clear that the Iirst part oI the extant
Satvrica, as well as much oI what preceded it and is now lost, was indeed a
long digression between the Iirst and the second voyages on the ship oI
Lichas. Apart Irom the very Iirst books oI the original Satvrica, which were
set in Massalia and dealt with the year oI preparation Ior Encolpius` excom-
munication Irom the city, the greatest part oI the story is structured around
the voyage on the ship and the complex love aIIairs between the passengers,
the captain and his wiIe. The voyage included several interludes on shore.
Stops were made in Ostia, which allowed Encolpius to visit Rome during the
Saturnalia, but most importantly, Encolpius abandons the ship in Campania
in an attempt to Iree himselI Irom Tryphaena`s monopoly oI Giton`s aIIec-

382
For a recent attempt to modiIy this view, see Schmeling 1994/5, 209, 'The Satvrica appears
to be written in discrete episodes each with its own beginning and end but strung and held
together by one narrator who is also an actor in each. Although true in so Iar as it stresses
the importance oI the narrator Ior the unity oI the narrative, Schmeling`s statement never-
theless ignores the unity oI plot in the Satvrica.
2. 2 RETROSPECTIVE SOLI LOQUI ES AND DIALOGUES 173
tion and perhaps the captain`s interest in himselI. This absence Irom the ship
involved several episodes and the introduction oI new characters. However,
the return to the ship shows that we are not dealing with loosely connected
episodes, but a consistently structured novelistic plot.
On the whole, we have tried to recover as much inIormation as possible
about the missing episodes oI the Satvrica by sketching a pattern oI the miss-
ing pieces oI the jig-saw puzzle that is the reconstruction oI the work, so that
the Iew pieces that we do have are made to cohere in a reasonably structured
composition. Necessarily, this reconstruction is tentative and sometimes
exempli gratia. Although we have occasionally Ior the sake oI argument
insisted on arranging the pieces in one way rather than another, it is clear
that it could at some points diIIer in detail.
2.3 Rewriting the Satvrica (My Turn)
2.3.1 The Nature oI the Summary
The interpretive summary I am about to oIIer will bring together the recon-
struction oI the previous chapters and serve as an aid to the reader in rehears-
ing the Iragmentary and sometimes incoherent story told in what is leIt oI the
Satvrica. The preserved text contains numerous explicit and implicit reIer-
ences to lost episodes. We should interpret these reIerences just as we do
other passages oI the work, and in doing so we inevitably Iorm ideas about
what was in the lost parts. Any complete interpretation oI the Satvrica`s
Iragments includes this sort oI expansion, Ior otherwise we must paradoxi-
cally treat the extant text as an artistic whole, ignoring its original design as
an extended narrative. The episodes related beIore the extant text begins are
naturally speculative, but they are based on my arguments in the previous
chapters. What I have done is merely to map out a minimum narrative that
accommodates i) all reIerences back Irom the text, and ii) all germane Irag-
ments. The reception oI the extant text oI the Satvrica has been a singularly
creative one Irom the very beginning. My creative or speculative summary is
thereIore merely in accordance with the traditional response to the Irag-
mented state oI the tradition. By oIIering a separate restorative summary,
however, I have avoided the graver mistake oI exercising my ingenuity on
the text itselI with arbitrary emendations.
The Iollowing short narrative reads as Greco-Roman erotic Iiction. Sulli-
van, in his major study oI the work Irom the late sixties, stopped short oI
providing a summary, because he thought 'a summary oI the missing (and
extant) episodes gives a misleading impression, the impression merely oI a
picaresque romance or an adventure story.
383
II a summary gives that im-
pression, however, that seems indeed to have been the extended Iorm oI the
work. Not long ago, moreover, in commenting on the introduction oI modern
theory into the study oI the ancient novel, Sullivan himselI made the point
that reader-oriented theories such as that oI WolIgang Iser, because they in
any case make much oI the reader`s role in extracting a meaning Irom the


383
Sullivan 1968, 38.
2. 3 REWRI TI NG THE SATYRICA (MY TURN) 175
text, should be taken by Petronian scholars as an encouragement 'to be more
enterprising in their theories about the missing portions oI the Satvricon.
384

It seems to me justiIiable to go into considerable detail in the summary,
which will make it rather long. This should be viewed as an attempt to intro-
duce my position on numerous points that are in danger oI being lost iI
merely presented as part oI a detailed argumentation.
385
Details oI the Iol-
lowing summary will be presented without Iurther arguments. II, as I hope is
not the case too oIten, my summary contains items that are not supported by
any arguments in the preceding chapters, the reader should take these as
purely speculative. I shall retain the Iorm oI personal recollection Ior the
sake oI Iidelity to the narrative structure oI the original, although this goes
contrary to the general practice in summaries.
386
Though the basic stance oI
the original is thus respected, obviously no attempt will be made to represent
IaithIully the discursive variety oI the work, nor have I Iound a way to repre-
sent IaithIully the psychological or rhetorical posturing oI the narrator, En-
colpius. JustiIication Ior the Iormer license could be Iound in my belieI that
the Satvrica is not primarily a poetic text, but privileges prose as its dis-
course oI choice. (The book numbers are merely approximate.)
2.3.2 Speculative Summary
|ca. 2 books|. Since you insist, I shall stitch together Ior you the varied
tales oI my woes, and tell oI how I traveled to many cities and got to
know the minds oI many people. I`m Irom Massalia, and my name is
Encolpius. My ancient Greek settlement in the middle oI barbaric Gallia
counts among its Iamous citizens certain travelers who explored the
outer-ocean, and came back to tell incredible tales. In this aristocratic
and austere outpost oI civilization I was given an old-Iashioned educa-
tion. But when I was still young, I Iell in love with a boy named Giton, a


384
Sullivan 1990, 91101.

385
It may be argued that the summary I am providing is a reconstruction oI the text since it
postulates one sort oI text rather than another. However, I nowhere intend to restore the lost
text verbatim nor do I claim the status oI the original Ior my speculative summary. Instead,
the summary is entirely supplementary to the received text and Iragments, merely attempt-
ing to explain and interpret those.

386
For a summary oI the Asinus Aureus or Metamorphoses oI Apuleius which likewise re-
spects the narrative Iorm oI the original, see Winkler 1985, 3I. For other and sometimes diI-
Ierent summaries oI the plot, see Ernout 1923, xliivii; Sage 1929, xxxvxxxvii; Waters
1902, xxvxxxi; Maiuri 1945, 5185 (only the Cena, but very detailed); Bcheler 1862,
2336; Sullivan 1968, 3480; van Thiel 1971, 2565; and now Schmeling 1996, 461469.
2 STORY 176
neighbor oI mine, and the most handsome and intelligent lad in the
whole oI Massalia. UnIortunately, my poverty caused me to have little to
give him in return Ior his Iavors. However, a plague had aIIlicted the city
and the priests were looking Ior someone who was Iit to play the scape-
goat in return Ior being Ied Ior a whole year at public expense. And so,
out oI need, I oIIered myselI as the scapegoat, and was accepted on ac-
count oI my anatomy. During this period oI easy living I grew Iatter and
healthier every day Irom eating Iigs and other pure Iood that people gave
me. The rumor about my large penis soon Iilled the city, and Ianned the
Ilames oI desire in a certain lady oI good standing whose name was Try-
phaena. This beautiIul and wealthy woman approached me through her
maids, and great intimacies began between us.
This year oI good living, however, was soon ending and the much-
dreaded ritual drew nigh. On the appointed day, I was dressed up accord-
ing to custom with branches and in sacred attire, and thus exposed to the
rage oI the citizenry, I was led through all the streets oI the city, while
they cursed my person and seemed to direct their attention particularly at
my private parts. This awIul experience, which was not just painIul but
extremely humiliating to myselI, Iinally ended in my escape Irom Mas-
salia and into exile.
Tryphaena Iinally came to my rescue, Ior she knew the captain oI a
very large merchantman Irom Tarentum anchored in the harbor oI Mas-
salia. His name was Lichas and he had his wiIe with him on the ship.
When I reIused to leave without Giton, Tryphaena consented to his com-
ing with me Ior she knew that we couldn`t live without each other. She
herselI decided to sail oII into voluntary exile, to Ireely pursue her own
pleasures in the world. With her came a large retinue oI slaves and many
other possessions.
|ca. 3 books|. Although the ship was ultimately bound Ior Tarentum,
where the captain had his wealthy estate, we had to make many stops on
the way. We Iirst set sail Ior Ostia, the harbor city oI Rome itselI. Once
at sea, Tryphaena began to show greater interest in Giton than in me, and
to my distress the boy seemed not much to mind her attention. The cap-
tain, however, made little attempt to hide his interest in my person, and
under the circumstances I could not reIuse him certain Iavors in token oI
my gratitude. AIter a Iew days sailing a violent storm Iell upon us, and
suIIering much hardship we made it into Portus, the harbor oI Ostia.
From there we went on land to the world`s capital. During our stay the
Romans celebrated an ancient Iestival in commemoration oI the licen-
tious equality that prevailed on earth in the golden age oI Saturn. For
2. 3 REWRI TI NG THE SATYRICA (MY TURN) 177
several days unrestrained liberty prevailed in the streets oI Rome; slaves
ridiculed their masters, and spoke with license on every subject; priests
made sacriIices with their heads uncloaked; criminals were spared;
schools were closed; all was riot and debauchery.
|ca. 2 books| We barely escaped Irom the madness, boarded the ship
again and headed Ior Campania. Now, the sweet harmony that had
reigned among crew and passengers was beginning to turn bitter. I could
no longer tolerate Tryphaena`s monopoly oI Giton`s aIIection. The two
were attempting to exclude me, and the unIairness oI it made my blood
boil with rage. Besides, on board was also the captain`s wiIe, Hedyle,
who Irom the start did not like her husband`s interest in me, and had now
caught him in his Ilagrant marital violations. Predictably, she wanted
somehow to take revenge on him and at the same time to satisIy her own
desires. Upon arriving in another Roman harbor, Puteoli, we leIt the ship
and went on Ioot to nearby Baiae. In this wanton bathing resort, the ma-
tron Hedyle, Giton and I secretly shared some licentious moments in the
company oI her maids, and Iinally we struck a pact to Ilee Irom the ship.
I Iirst oI all wanted to get Giton away Irom Tryphaena to restore our old
relationship. When the others were still at Baiae, the three oI us went
back to the ship and I stole the sistrum and Iine robe oI the ship`s pro-
tecting deity, Isis, because, as I hoped, it would Ietch a Iine price later
when sold. In the olden days travelers used to be Ied by pious people, but
in our times nobody helps a stranger without a proIit. When we were still
in the portico oI Hercules in the harbor area oI Puteoli, we ran into
Lichas and Tryphaena who had come to suspect what was aIoot. A vio-
lent conIrontation ensued where I denounced Tryphaena`s wicked ways
and said that she was to blame Ior our elopement. Hedyle appealed to
nearby sailors Ior help, claiming her husband was a common pirate who
was trying to abduct her Iriends and herselI. The last we saw oI them
was that our noble sailors were beating and making ready to mug a
clamoring and Iurious Lichas.
|ca. 2 books|. It slowly dawned on me that I was embarking on a
criminal career, and my Iear oI being caught made me intensely uneasy.
At Iirst we stayed with Hedyle, but it seemed to me that she was turning
into another Tryphaena, and so I began to look Ior ways to get rid oI her.
What I wanted primarily was to be alone with Giton. In a market in
Puteoli, I tried to sell Ior our sustenance the Iine robe and rattle oI Isis.
However, an advocate named Euscion caught us and accused us oI hav-
ing simply stolen the expensive items. The cruel man loudly screamed
that we were Iugitives and couldn`t possibly be the rightIul owners oI so
2 STORY 178
many regal spoils. A true Cerberus oI the courts, he kept barking his ac-
cusations until we were Iorced to hand over the goods and had to prom-
ise that we would show up in court the day aIter. We were badly shaken
and our chances oI escaping were slim, but as it happened a rich gentle-
man by the name oI Lycurgus spotted us and took an immediate liking to
our persons. Lycurgus was an inIluential man, and he sent his steward to
make a deal with the magistrates, who upon being paid handed us over to
his custody.
|ca. 3 books|. And so it was that I got rid oI Hedyle and against all
odds I escaped trial. We, oI course, were very happy to have Iound a sav-
ior, and gladly went with him to his sumptuous Campanian villa. Soon,
however, it became clear what he intended Ior me, because, as I discov-
ered, he had a private jailhouse on his estate. Though I had escaped the
harshness oI public justice, the hardships I suIIered in the private para-
dise oI this cruel man were worse. Giton, however, got diIIerent treat-
ment, Ior in the daytime Lycurgus made him do the job oI a woman in
the jail, but at night he brought the boy to his bed. This arrangement
greatly upset my Ieelings. In this slave-gang Ior Iree men, I met Ascyl-
tos, a young man oI a very similar Iortune to myselI. He didn`t run Irom
my sexual advances, and so we became lovers in the garden, where we
otherwise oIten suIIered great hardships Irom being bitten by Ilies, but
mainly because we were still young and unhardened and certainly not
accustomed to working. In Iact, because oI my soIt nature, I was not oI
much use to Lycurgus, until he decided that he could use me in his
gladiatorial company. I wasn`t even trained, but when the day arrived
they dressed me in a delicate tunic, and Iorced me into the center oI the
spectacle, where I was shameIully matched with a woman, who almost
killed me. As the laughter Iilled the amphitheater, Fortune saw Iit to save
my liIe by bringing on a great earthquake, which shook down the theater.
Many died but I escaped by hurling myselI into the latrine. AIter the dis-
aster, when night had Iallen, I returned again to the villa oI Lycurgus.
When I met Ascyltos again, he commented on the awIul stench that
came Irom my person, but decided to join me in my plans to escape. I
had been careIul to hide Irom Ascyltos the true nature oI my relationship
with Giton. Ascyltos and I broke into the villa and Iound Giton sadly
sleeping in the arms oI Lycurgus. I was so enraged that I killed my en-
emy and so saved my dearest Giton. The boy helped me to Iind a treas-
ure oI gold coins that the miser Lycurgus had sowed into his shirt. This
we took with us and some other articles that we could carry.
2. 3 REWRI TI NG THE SATYRICA (MY TURN) 179
|Book 13?|. The three oI us, Ascyltos, Giton and I, had now Iormed
a company and decided to share our belongings. It was our plan to take
reIuge in Naples, a Greek city nearby, which was reputed to have schools
oI declamation and inhabitants who loved eloquence and the liberal arts.
We hoped there to reap beneIits Irom an environment so congenial to the
educated. At night we moved southwards along the highway, but as we
emerged Irom the Crypta Neapolitana we came upon a shrine sacred to
Priapus. The cult was engaged in a wild orgy, and when I suddenly ap-
peared to them during the celebration halI naked and out oI nowhere,
they immediately Iell to my Ieet and worshipped me as iI I were Priapus.
There I stood impersonating the god, stick and all, and those supersti-
tious people Iirst completely undressed me then clothed me in an expen-
sive robe. I was still wearing this outIit when I ran away and accidentally
leIt behind the gold coins in the old shirt. From a distance I saw a rustic
Iind it where I had leIt it unguarded. Ascyltos did not believe me and
suspected that I had stolen the money myselI. Late that night we Iinally
arrived at our destination, where we checked into a lodging house. That
night I still couldn`t enjoy the pleasures oI Giton because Ascyltos kept a
wakeIul eye on the boy.
|Book 14|. When the sun came up the next morning, Ascyltos and I
put our plan into practice and went straight to the local school oI rhetoric
where we Iirst encountered the antescholarius Menelaus and then lis-
tened to a suasoria delivered by the rhetor Agamemnon. In the portico
outside the school we introduced ourselves to him and he promised to
take us along to a dinner-party that night. The discussion turned to elo-
quence and I tried to impress him by delivering a short tirade against the
vices oI modern declamation. Agamemnon appreciated my sound judg-
ment, but deIended his proIession and extemporized a little poem on
morals and education. Ascyltos disappeared in the middle oI the poem,
and when I noticed this, I too went aIter him because his recent interest
in Giton caused me not to trust him. But I got lost in the city and when I
asked Ior directions to my home, a street-vendor tricked me into a
brothel where she said I must live. There I ran into Ascyltos who said a
man who wanted to have sex with him had similarly tricked him. We
barely managed to escape Irom this place. Resuming my erratic search
Ior the inn I Iinally caught a glimpse oI Giton in a mist, standing in the
street. When the boy had showed me our lodgings, I asked him about
lunch, but he started sobbing and told me that Ascyltos had been there
earlier and had attempted to rape him. I was Iurious and quarreled bit-
terly with Ascyltos, who deIended himselI by reminding me that he had
2 STORY 180
allowed me to do with him in the garden what he had attempted to do
with the boy at the inn. I wanted to break up our Iriendship right away, in
order to resume my old relationship with Giton. Ascyltos, however,
pointed out that Agamemnon had invited us that night to go with him to
a dinner-party, and so, since I was both hungry and penniless, I reluc-
tantly agreed to postpone the breakup until the day aIter. AIter checking
to see whether Ascyltos had leIt I hurried to take my pleasure with Giton,
but beIore I was completely Iinished, Ascyltos came back and inter-
rupted our lovemaking, beating and taunting me Ior what I was doing
with Giton.
As a preparation Ior dividing our belongings we went to the Iorum to
sell the stolen robe Irom the cult oI Priapus. Through Fortune`s won-
drous play, the man and the woman who came up to us and showed in-
terest in the garment had with them our shirt with the money. I realized
that the man was the rustic who had Iound the shirt, and that he was try-
ing to sell it without knowing its actual worth. I wanted to claim my
property, but as I knew Irom previous experience the courts are both cor-
rupt and unIavorable to strangers, and so we decided that it was better to
buy it back, except we had no money. Suddenly, the woman recognized
the pallium Irom its signs and called out Ior the bystanders to grab the
thieves! We were each holding on to the others` garment, when some
shady jurists came up to us. They claimed both parties were suspected oI
stealing and thereIore we had to deposit the pallium to be exhibited in
court the day aIter. We, oI course, demanded likewise that the others also
deposit the shirt. But the rustic grew so indignant at being accused oI
stealing the seemingly worthless shirt that instead oI depositing it he
threw it in Ascyltos` Iace. And so by chance we Iully recovered our
treasure. But on the way back we were Iollowed by the woman, who
turned out to be the maid oI Quartilla, the priestess oI the cult oI Priapus,
which I had tricked the night beIore. She herselI soon arrived at the inn
and demanded that we help her Iind a remedy Ior her tertian Iever, which
she said was caused by the cold which resulted Irom her religious awe
aIter the sacrilegious trick played on her. She now took over the inn, and
her maid immediately went to work on my penis, which was totally
Irigid. When Ascyltos and I tried to struggle we were tied up and taken
back to the shrine oI Priapus where I was given the aphrodisiac satyrion
and we told each other erotic stories. AIter I had drunk most oI the sa-
tyrion, I became all Iired up and Quartilla and I had intercourse. Later
she had us tortured and assaulted by a cinaedus. The whole party Iell
asleep out oI exhaustion, but we soon woke up again, and Quartilla made
2. 3 REWRI TI NG THE SATYRICA (MY TURN) 181
sure we stayed awake most oI the night, claiming that the orgy was an
all-night vigil Ior Priapus. She and her maid Psyche staged the marriage
oI poor Giton and 'Nightlong (Pannychis), a slave girl oI hers who
looked no older than seven. Neither boy nor girl seemed to mind getting
married, and Quartilla dragged me close to her to watch their childish
play through a hole in the door. Towards morning Ascyltos, Giton and I
were allowed to return to our lodgings, aIter promising to return Ior a last
dinner oI liberty the day aIter.
|Book 15|. Now came the second day oI our stay in the city, when
the tertian attack oI Quartilla would come, iI our remedy hadn`t worked.
We were terriIied oI the prospect oI returning to her place, and I was
especially worried about her threat to have Giton as an erotic appetizer
that evening. But one oI Agamemnon`s slaves came by to remind us oI
another dinner invitation Ior scholars, this time to the house oI the Ireed-
man Trimalchio. So we Iorgot about Quartilla and walked to the baths,
where we met with this extremely wealthy and eccentric person, Trimal-
chio, and later went home with him. His house was a maze oI bizarre de-
cor and his servants behaved like public entertainers and tricksters rather
than the household oI a pater Iamilias. The Iood was likewise, and every
dish turned out to be made oI something other than what it had seemed at
Iirst. Trimalchio was constantly showing oII his love oI the liberal arts
and readily put on display his jumbled knowledge, but we, the scholas-
tics that we were, were Iorced to stay silent except to Ilatter him, or else
never be invited again. Meanwhile the Ireedmen, who were his guests,
rambled on in an unschooled manner, even attacking us Ior our arro-
gance and saying that we had become silly Irom literature. Habinnas, a
stonemason working on the host`s tomb, was a late arrival to the party.
When Trimalchio recklessly wanted us to take a bath on a Iull stomach
we tried to escape Irom his house Ior the Iirst time, but the doorkeeper
and his hellish watchdog turned us back. The place was beginning to
seem like the underworld, as we were unable to leave. When the house-
hold was rehearsing the Iuneral oI their drunken master the noise was so
great that the night watch thought the villa was on Iire and broke in.
Then we took the opportunity, said good-bye to Agamemnon, and Iinally
made our escape.
|Book 16?|. Through Giton`s intelligence we at last made it back to
our lodgings in the dark oI night, and once there I tried to make love to
the boy, but Iell asleep in the middle oI the action out oI sheer drunken-
ness. While I was unconscious, Ascyltos, the scoundrel, stole Giton away
Irom me and brought him to his own bed. When I woke up the morning
2 STORY 182
aIter and Iound my bed empty I Iirst planned to kill both oI them in their
sleep but decided not to and instead beat Giton out oI the bed and de-
manded that Ascyltos leave right away. He didn`t protest and we pro-
ceeded to divide the spoils. But then he insisted that we split Giton as
well, as iI he too were our common property. I thought he was joking,
but he drew his sword and threatened to cut oII his share. I grabbed my
sword and we prepared to Iight, but Giton begged us to calm down, and
said he alone deserved to die, because he had violated the oath oI Iriend-
ship. Ascyltos suggested that the boy be allowed to choose whom he
wished to Iollow, and I eagerly consented to this plan, thinking that my
long-standing relationship with the boy had Iorged unbreakable ties be-
tween us. Given a choice the boy didn`t even hesitate but straightway
preIerred Ascyltos and leIt with him. I was thunderstruck with sorrow
and jealousy and moved to another lodging where I locked myselI in Ior
three days. I realized the mistake I had made. In getting rid oI Lycurgus I
had taken into my trust Ascyltos, who was even crueler. I decided to
avenge my disgrace and kill the couple, and so I armed myselI with a
sword and went out to look Ior them in the porticoes. But a soldier I met
on the way stopped me and conIiscated my sword. Perhaps he was just a
thieI. Later I was glad that he had Irustrated my murderous plan.
I came to a temple where there was a gallery Iull oI old paintings. A
poet named Eumolpus entered the temple and introduced himselI to me
as I stood there admiring pictures oI mythical lovers who, unlike me, en-
joyed their loved ones without competition. Eumolpus tried to persuade
me that his ragged looks were a prooI oI his talent, since it showed him
to be better than the parasites who compromise artistic integrity by Ilat-
tery at the tables oI the rich. Rich men, he said, hate and persecute the
lovers oI letters in order to make them, too, seem venal and subservient
to money. I told him about Giton and my erotic sorrows, and to cheer me
up he told me a tale oI his conquest oI a Pergamene boy. His story was
intended to show how all boys put up resistance at Iirst and have to be
bought, but later cannot get enough oI a good thing. Then we discussed
the decline oI the classical arts and that led to his improvising on the Iall
oI Troy, which was the subject oI the painting I had been admiring in the
gallery. Eumolpus had to stop reciting when those walking in the gallery
pelted him with stones. He Iled the temple and I Iollowed him down to
the sea Iearing that I too might be taken Ior a poet. When we were out oI
reach oI their missiles, I told him he was crazy and that I too would
throw stones at him iI he didn`t stop talking like a poet instead oI a man.
He acknowledged that he had oIten met with such negative applause but
2. 3 REWRI TI NG THE SATYRICA (MY TURN) 183
promised to control himselI in the Iuture. We decided to have dinner to-
gether at my place.
We went to the baths and there I Iound Giton alone guarding the
clothes oI Ascyltos and looking unhappy. I took him back to the inn and
leIt Eumolpus reciting at the baths. My love Ior Giton quickly caused me
to Iorgive him. He told me he had only chosen Ascyltos because he was
stronger, out oI concern Ior my welIare, because otherwise I would have
suIIered at his hands. When night came and dinner was served Eumolpus
returned. I was aIraid he would bring back Ascyltos, but when he turned
out to be alone I let him in. He told oI how someone by the name oI
Ascyltos had lost his clothes in the baths and had been leIt standing na-
ked, while a crowd gathered to admire his enormous member, which
made the boy himselI look like a mere appendage. Finally, a Roman
knight who eagerly came to his assistance escorted the young man home.
The poet Iinally contrasted this Iine appraisal oI a penis to the disgrace-
Iul rejection oI his poetry. When Eumolpus saw Giton, he immediately
became interested in him, while I Ieared that I had taken into my trust yet
another rival Ior Giton`s love. I was Iurther irritated by Giton`s positive
admiration Ior the old man. I showed Eumolpus the door, but he tricked
me by locking me in and going aIter the boy. I decided to hang myselI,
but was stopped by the two oI them upon returning. Giton said he would
die beIore me and proceeded to cut his throat with a razor. I too grabbed
the same weapon and tried to kill myselI but it turned out to be an espe-
cially blunted razor Ior beginners and so neither oI us was harmed. A
lodger now entered the room complaining about the noise and soon Eu-
molpus was Iighting with a whole group oI lodgers outside. I was so an-
gry with him that I enjoyed watching him being beaten. When Giton
wanted to help the old man, I hit him on the head and made him cry. Fi-
nally, Bargates the caretaker oI the block saved Eumolpus, Ior they were
acquainted and he wanted to employ the poet to compose invective
against his mistress. At this moment Ascyltos arrived with a public her-
ald to proclaim the loss oI Giton and oIIer rewards Ior his restitution. I
made Giton hide under a bed and sought to soIten Ascyltos` anger by ly-
ing to him that I had not seen Giton, and claimed that his search was just
a pretext to kill me. He assured me that he was still my Iriend, and in the
end he leIt with the herald. Eumolpus now rushed into the room and
threatened to Ietch Ascyltos again to collect the reward. But I pleaded
with him, saying that the boy had run away, but then Giton sneezed and
gave himselI away. The boy nursed the old man`s injuries and together
we appealed to his humanity and culture. He said it was his custom to
2 STORY 184
live dangerously and promised to spare us, but demanded that we either
Iollow him or lead the way to somewhere else.
|Book 17?|. It was dark night when we Iollowed him aboard an un-
known ship and spread our blankets in a secret place on the deck. To our
extreme horror we soon discovered that we had inadvertently returned to
the ship oI Lichas and that Tryphaena was still aboard. When we told
this to Eumolpus, he begged us to believe that although he had used the
ship beIore he hadn`t known about our enmities with the captain and his
passenger. In our conIusion we sought a solution to our problems by ar-
guing pro and contra as they do in the schools but we Iound declamation
useless Ior solving real problems. Finally, we shaved our heads and Eu-
molpus inscribed our Ioreheads with ink to give credence to his story that
we were his branded runaway slaves. As we were shaving, however, a
seasick passenger saw us and reported our ominous behavior at sea. Both
the captain and Tryphaena, who had been desperately looking Ior us, had
mysteriously dreamt that we had come on board the ship again. When we
were dragged beIore the captain and his Iemale passenger, she and her
handmaidens recognized Giton, and he himselI recognized my penis, and
addressed it by my name, Encolpius. Eumolpus boldly pleaded our case,
but without success. It was only when the ship was in a state oI mutiny
because oI us that Eumolpus was able to broker an armistice with a Ior-
mal treaty, according to which Tryphaena and Lichas were required to
pay good cash in compensation iI they sexually assaulted Giton or my-
selI.
Festivity took over Irom hostility. Eumolpus entertained us with a
satire about a certain widow Irom Ephesus, who was so IaithIul to the
memory oI her dead husband that she even Iollowed him into his tomb
and stayed there Ior days. The story was intended to show that although
women may seem chaste at Iirst, putting up austere deIenses, they Iall in
love all too easily, and in the end Iorget their own children in their mad
desire Ior a complete stranger. The sailors laughed at the story, but not
Lichas who was reminded oI his wiIe Hedyle. Tryphaena blushed with
shame and buried her Iace in Giton`s neck. My jealousy was rekindled at
being excluded Irom their sweet caresses, though the dear boy was
probably just being cautious not to upset a newly brokered armistice.
Now a storm came on and when Lichas was asking me with hands out-
stretched Ior the stolen robe and rattle oI the ship`s protective goddess,
he was suddenly carried oII by the wind and disappeared into the sea.
Tryphaena escaped saIely into the ship`s dinghy along with her Iamily oI
slaves. Giton and I tied ourselves together to be united in the moment oI
2. 3 REWRI TI NG THE SATYRICA (MY TURN) 185
death. We driIted ashore in the storm-tossed wreck. Eumolpus, however,
did not even notice what was happening, since he was inspired below
deck writing out a poem on a huge parchment. We had to drag the Ire-
netic poet out oI the wreck. I also Iound Lichas` body lying on the beach
and improvised on the theme oI how human planning is regularly Irus-
trated by Fortune. Later we burnt the corpse on a pyre and Eumolpus
improvised an epigram.
We headed on Ioot Ior Croton, originally an illustrious and warlike
Greek colony, but now a mere ghost town and obsessed with legacy
hunting. It was Eumolpus` plan to pose as a shipwrecked AIrican land-
owner with great estates, who had recently been bereIt oI his only son
and was himselI in bad health. We were to pose as his slaves. This was
intended to trick the locals into providing Ior us in the hope oI inheriting
Eumolpus` alleged property. Here the gold Irom the villa oI Lycurgus
and the clothes I had stolen were useIul to give Eumolpus the appearance
oI aIIluence. On the road to Croton, the poet recited his unIinished po-
litical epic about the civil war between Caesar and Pompey, the one he
was inspired to compose on the storm-tossed ship. He eIIused this com-
position with whirling Iluency until he Iinally stopped, just in time Ior us
to arrive at our destination. Once there, we executed successIully the in-
spired plan oI Eumolpus.
|Book 18?|. For a long time we lived in great luxury and stuIIed our-
selves with Iood, though like most outlaws I anxiously waited Ior the
punishment I had in store. In Croton, where I had taken up the pseudo-
nym Polyaenos, a beautiIul young matron, appropriately called Circe, so-
licited me. As I learned Irom Chrysis, her handmaiden and go-between,
the young woman typically burned with desire Ior condemned criminals
and Ilogged slaves. When we met in the grove by the shrine oI Venus, I
beheld a beauty even beyond the best marble statues oI the great ancient
sculptors. I was naturally eager to please her, but I Iound myselI unable
to have an erection. As it turned out, I couldn`t even succeed with Giton.
I tried various remedies beIore seeing Circe Ior the second time but I was
still unable to please her. I went home in utter shame, having been not
unjustly Ilogged and spat upon by her slaves. Though I Iailed in my at-
tempt to cut oII my useless and deceased member, at least I succeeded in
verbally berating it. AIterwards, I Ielt a little embarrassed Ior talking to
my penis, but as I reasoned to myselI one can Iind epic precedents Ior
this kind oI behavior; ancient heroes never thought much oI conversing
with parts oI their body, the heart Ior example, or the eyes.
2 STORY 186
It dawned on me that I might have been rendered impotent by the
aphrodisiac Ascyltos and I drank in such quantities during Quartilla`s
orgy back in the Greek city. So I checked with Giton to Iind out whether
Ascyltos, too, had abstained Irom lovemaking that night when he carried
the boy oII to his own bed while I was asleep. The way I Iigured it, this
would Iurnish me with a prooI. Giton swore that Ascyltos had not Iorced
him. Armed with this evidence Ior poisoning, I went to the temple oI
Priapus and kneeling on the threshold I apologized to the god Ior my
crime, which, as I explained, was not all that serious, and had in Iact
been committed out oI need, and, besides, with only one part oI my
body. Finally I promised him a sacriIicial banquet when things got bet-
ter. When I was still at the temple, Proselenos, an old hag who had tried
beIore to cure my impotence, dragged me into the priestess` cell and
started beating me Ior having excited the wrath oI the gods against her.
Oenothea, the priestess oI the cult, now arrived in her cell, and upon
hearing the story oI my Iailure she declared that she could Iix any prob-
lem with her magic, and mine especially iI I only spent a night with her.
And so I became a guest in the priestess` squalid cell. She was in the
middle oI preparing a disgusting dinner when she Iell oII a chair, landed
on the stove, toppled the pot and put out the Iire. While she was away
Ietching Iire at a neighbor`s, the sacred geese oI the temple attacked me,
whose leader I heroically slew in a Iierce battle. I Ieared that Oenothea
would become angry at Iinding the goose dead and so I tried to leave, but
as I was about to go, she returned with the Iire. I pretended that I had
been waiting Ior her there on the threshold. I told her what had happened
and showed her the goose and she panicked at beholding such sacrilege.
When Proselenos arrived she too acted as iI I had killed my own Iather.
They were completely calmed, however, as soon as I oIIered them two
gold pieces Ior the goose. II you have money you can get away with any-
thing, including murder. We now cooked the goose and drank wine, and
the drunken and libidinous women tortured me with their useless medi-
cine Ior impotence until I Iled the temple with them in pursuit. Next,
Chrysis, the maid oI Circe, Iell in love with me. She had obviously
changed her mind Irom the time she told me that she only Iancied upper-
class men. But I was still useless Ior lovemaking.
A certain matron whose name was Philomela put her two children in
the care oI Eumolpus, ostensibly Ior their education but really to prosti-
tute them Ior a share in his presumed legacy. Eumolpus straightway took
advantage oI the situation and copulated with the daughter, although he
tried everything to preserve the pretense that he was in bad health. I tried
2. 3 REWRI TI NG THE SATYRICA (MY TURN) 187
to have the son when he was peeping on his sister and Eumolpus, but
Iound the god still against me.
Finally I told Eumolpus that Mercury, who oIten plays this role, had
conveyed that particular part oI my body back Irom the dead. I liIted my
tunic and Eumolpus approved all. He was taken aback at Iirst, but in or-
der to better believe it, he handled the Iavor oI the gods with both his
hands. I warned him that the Iortune-hunters were tired oI his promises
and that they were growing less generous. In his last will, Eumolpus
stated that each one oI his heirs, except we his Ireedmen, would be re-
quired to eat a piece oI his dead body beIore receiving any inheritance.
One oI the legacy-hunters, Gorgias by name, showed himselI willing to
satisIy the condition .


Part 3
Genre







3.1 Ancient Narrative in personis!
3.1.1 The Form oI the Satvrica
AIter the preceding three chapters, dedicated to the summary and reconstruc-
tion oI the Satvrica, we are now ready to consider in more detail the narra-
tive Iorm, including its ancestry and place in the Iamily oI Greco-Roman
literary Iorms. As we saw in the Iirst two chapters, discussions oI this topic
in the scholarship usually begin Irom the premise that the Satvrica is con-
structed Irom two or more established primary Iorms, beginning with the
prosimetric or 'Menippean satire and the Greco-Roman novel and continu-
ing with a long list oI known genres (ancient and modern) that are thought to
have lent components to the amalgam that is the genre oI the Satvrica. Ac-
cepted as a premise since the late nineteenth century, this diIIicult notion oI a
Iormal duality or plurality has had the consequence oI undermining the unity
oI the work as a whole. II the genre is not one, but two or more, then the text
belongs to no genuine genre, but is synthetic with respect to genre, whatever
that may imply, and the eIIorts oI the critic end up being directed at the
hopeless task oI Iinding a viable metaphor to describe this complex state oI
aIIairs, without losing sight oI the Iact that the Satvrica is a single composi-
tion, a single work oI literature. In the supposed absence oI another similar
text, the Satvrica is compared by scholars with other very diIIerent works oI
literature and thus Iound to belong, in terms oI genre, to the same or a simi-
lar category as the 'comparison text oI choice Ior each scholar. As we have
seen above, the language and arguments used to Iorge such connections be-
tween dissimilar texts are oIten ingenious.
And yet the Satvrica has a logic oI its own and shows distinct Iormal
qualities. Its distinctive character is not, however, to be Iound in the mixture
oI prose and poetry or the work`s common traits with the Greek novel, mime
or indeed in its common traits with other genres, although these are certainly
important Ieatures. The one distinct Ieature oI the Satvrica that outweighs all
others is, oI course, the Iact that Encolpius` narrative is conducted in the
'Iirst person, or to use my own terminology, the narrative he delivers is his
personal recollection. Here is a clear Iormal Ieature which has surprisingly
not Ieatured much so Iar in the discussions oI the Iormal characteristics or
the genre oI the Satvrica. More speciIically, Encolpius` personal recollection
3 GENRE

192
includes a series oI extended and repeated narrative impersonations which,
as we saw in section 1.2, warrant our deIining it as narratio in personis,
based on the deIinition oI that term in ancient rhetorical manuals. As we
shall see this Iormal quality oI the work is an identiIying signature, which
squarely places the Satvrica within a single known ancient genre. In modern
times it has been known since the late nineteenth century that the Satvrica
could well be classiIied with similar extant and lost works. But through a
great eIIort oI scholarship, motivated by an ideology strictly speaking irrele-
vant to the understanding oI ancient literature, our text has been divorced
Irom its proper genre. A single text with no equal is a generic problem, but
two or more texts that share qualities to such a degree that they begin to pos-
sess generic sameness, that is the norm in ancient literature and what we call
a genre. But beIore we look at the arguments in support oI the classiIication
oI the Satvrica as Milesian Iiction, let us look at how, so Iar, the 'Iirst per-
son Iorm has been interpreted in the scholarship.
3.1.2 Understanding the Ancient First Person Narrator
Interpreting the narrative Iorm oI the Satvrica in modern times has been
made diIIicult by excessive analogy with the modern novel. The history oI
this interpretation can be traced at least back to 1889 when Elimar Klebs
made attempts to describe Encolpius` tone and narrative stance in the Sa-
tvrica by comparison with contemporary realistic novels. What may seem
surprising to progressive classicists, who are prone to lament the innate con-
servatism oI the discipline, is that Irom the very beginning oI the modern
reception oI the Satvrica critical concepts developed Ior the study oI the
modern novel have been prominent. Over a century ago Klebs employed
critical terms Irom the emerging science oI modern narratology to advance
his once inIluential thesis that the Satvrica was an 'almost modern realistic
novel oI the 'Ich-Roman type, only with an epic structure borrowed Irom
the Odvssev which supplied it with esthetic and artistic unity.
387
The purpose
oI the parodic comparisons oI the protagonist`s experience with the lot oI
epic heroes, according to Klebs, was to express, by way oI irony, the narra-
tor`s awareness oI his pathetic humiliation. This irony was supposedly both
sophisticated and selI-conscious and resembled the complicated narrative
posturing oIten assumed in modern realistic novels.
A decade later Klebs` interpretation was countered by Heinze, whose
understanding oI the narrative structure oI the Satvrica was considerably


387
Klebs 1889, 631I.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

193
more rooted in Greco-Roman literary history.
388
Heinze revises the modern
conception oI the stance oI Encolpius as an ironic selI-conscious reIlection
upon his own humiliation. He sees the selI-deprecation oI the narrator as part
oI the ancient comic stance. He compares the narrative stance oI Encolpius
to the narrators oI the Greek erotic Iictions and observes major diIIerences:
whereas the tone oI the latter shows that the trials and tribulations oI its he-
roes are meant to be taken 'very seriously (bitter ernst), the work oI
Petronius sports an attitude which implies that whatever pain and sorrow is
endured by Encolpius and his comrades is 'Iully deserved (verdienen sie
reichlich), and 'can only, and should only arouse laughter in the reader (die
der Leser nur :um Lachen rei:en knnen und sollen). The diIIerence be-
tween these two narrative types, he claims, is akin to the diIIerence in drama
between tragedy and comedy or Iarce. An even closer parallel would be
tragedy and satyr plays, or tragic parody. Although Heinze showed an under-
standing oI the problems involved in determining the tone and stance oI the
narrator, his eIIort to deIine the diIIerence between the Satvrica and the Iive
Iully extant Greek novels never advanced much beyond saying that Encol-
pius` pose was comic and parodic in comparison with them, and in order to
make his argument more convincing he was certainly led to exaggerate the
'seriousness oI tone in the Greek novels.
The problem resurIaced with a vengeance in the scholarship oI the six-
ties and the seventies oI the twentieth century, a period particularly inter-
ested in questions oI psychology and character. Veyne, Sullivan, Walsh,
Rankin and George saw Encolpius as an inconsistent and Iragmented per-
sonality, who was Iurther complicated by being merely a 'transparent mask
Ior the historical author, who accordingly was the real narrator.
389

It was as an attempt to reconcile 'the discrepancies in Encolpius` charac-
ter and to clearly diIIerentiate the narrator`s persona Irom that oI the his-
torical author that Roger Beck presented his study oI the narrative structure
oI the Satvrica. As in the previous century the discipline oI modern narratol-
ogy is the theoretical background, and Beck explicitly reIers to the new and
improved Iormulations in this Iield, especially as they had then recently been
applied to the Ancient Novel by Tomas Hgg in his Narrative Technique in
Ancient Greek Romances (Stockholm 1971). In Beck`s words '|the| key to
the solution is |.| a realization that in dealing with Encolpius one is con-
cerned not with a single person but with two. Not only are they two distinct
persons separated by what is presumably a considerable span oI time (the


388
Heinze 1899, 503.

389
Veyne 1964, 301324; Sullivan 1968, 119; Walsh 1970, 81; Rankin 1971, 19; George
1966, 349I.
3 GENRE

194
narrator is looking back on his own past adventures) but they are also two
very different characters.
390
AIter his exposition oI the structural relation-
ship between narrator and protagonist in the Satvrica, Beck proceeds to es-
tablish the sophistication oI the narrator versus the navete oI the protagonist.
As it happened, it was this Iurther elaboration oI the thesis which would
be incorrectly construed as the touchstone oI the validity oI his basic descrip-
tion oI the narrative structure. In an article Irom 1987 which directly ad-
dresses Beck`s proposed solution, F. Jones at Iirst appears to accept the basic
Iormal distinction between Encolpius qua narrator and Encolpius qua pro-
tagonist. But then he proceeds to cast doubt on the extent oI the maturity and
sophistication oI the narrator, although he, too, sees him as having aban-
doned the bombastic rhetorical style oI his youth. Nevertheless, Jones con-
cludes by doubting the extent oI the gap between the two Encolpii, as a cau-
tionary measure aIter what he considers Beck`s Iailure to positively deIine
the complicated diIIerence between their characters: '|s|ometimes |.| the
narrator seems to get so involved in the act oI narration that his distance
|Irom the protagonist| vanishes.
391
Instead oI the Iormal diIIerence between
narrator and protagonist, we are leIt with a 'clear connection between the
narrator and his Iormer selI, and a 'solidity and continuity to his character,
though some development in his personality is noted.
392
In a sense this con-
clusion is a natural consequence oI the modern method and premise oI such
studies. Once you begin to read the Satvrica with a method developed pri-
marily Ior the study oI the Bildungsroman, perhaps the most obvious type oI
modern 'Iirst person novel, certain ideas and assumptions are necessarily
carried over Irom the 'comparison text. Without that association, the criti-
cal issue ceases to be the demonstration oI character development or matur-
ity, through pinpointing those character traits oI the narrator which will put
him at a secure psychological distance Irom the protagonist. This Iormula-
tion, moreover, risks conIusing psychological distance, or diIIerence in
mood and personality, with temporal and situational distance, or diIIerence
in personae. Despite the inviting etymology, the ancient rhetorical term per-
sona` is a very diIIerent concept Irom the modern psychological term per-
sonality`.
393

Perhaps sensing that this approach was ultimately an interpretive dead
end, Niall W. Slater proposed a Iresh start by introducing a brand oI reader-

390
Beck 1973, 43. The arguments are Iurther elaborated in Beck 1975.
391
Jones 1987, 819.
392
Jones 1987, 819.
393
On this topic, see Gill 1990 (ed.) and Gill 1996.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

195
oriented criticism into the study oI the Satvrica.
394
His Reading Petronius
(1990), a book which is both neat and attractive in presentation, argues Ior a
complete Iusion oI narrator and protagonist. This uncompromising stance
may, perhaps, be seen as a logical consequence oI the premise which he
adopts Ior his study, Ior iI the reader is the ultimate reIerence, in matters oI
temporal distance as well as in other respects, the only meaningIul present in
narrative is the moment when the reader casts his eyes upon a particular
page, and the only meaningIul past is the reader`s recollection oI pages read.
In essence, this idea stresses the reader`s linear experience oI the text, when
Iirst read, as radically diIIerent Irom what happens in the reader`s mind dur-
ing a subsequent reading, when it would under normal circumstances be
known where the story as a whole is leading.
395
When applied, however, by
Slater to the incomplete text oI the Satvrica, the concept oI the 'Iirst read-
ing is transIormed into something akin to an argument ex silentio to the
eIIect that since the modern reader has not 'so Iar experienced the prologue
oI the Satvrica, where we must assume the narration oI Encolpius was intro-
duced, there remains no reason Ior him to distinguish between the present oI
the narrating act and the past oI the story:
The notion oI a split in Encolpius between narrator and actor, old per-
sona and young persona, |is not| supported by anything the reader has so
Iar experienced in the text. Even iI there were in the lost beginning
something to set up the notion oI Encolpius recounting his past adven-
tures, we should expect some reinIorcement oI this temporal perspective
in the text. There is none.
396

Slater`s attempt to level out the temporal dimensions oI the Satvrica leads
ipso facto to the obliteration oI the narrator, which leaves the protagonist
alone, as it were, acting out the narrative. From here there is a direct route to
interpretations oI Petronius` text as the 'narrative equivalent oI other me-
dia, such as a play on the stage or a movie on the screen.
Such interpretations, Iresh and interesting though they are, clash with the
basic premise oI narrative Iorm as understood by ancient and modern stu-
dents. As Gerard Genette has shown in his admirably comprehensive discus-
sion oI the moment oI narration with respect to the time oI the story itselI,
there are Iour basic types oI narrating acts to be reckoned with:

394
Slater 1990, 467.
395
Slater 1990, 140 n.2, acknowledges his debt to Winkler`s discussion oI reading and critical
method, especially the notion oI 'Iirst reading. Winkler`s reader oriented criticism is most
evident in his Auctor & Actor (1985), a study oI Apuleius` Metamorphoses.
396
Slater 1990, 47 with n.26.
3 GENRE

196
subsequent (the classical position oI the past-tense narrative, undoubt-
edly Iar and away the most Irequent); prior (predictive narrative, gener-
ally in the Iuture tense, but not prohibited Irom being conjugated in the
present |.|); simultaneous (narrative in the present contemporaneous
with the action); and interpolated (between the moments oI the ac-
tion).
397


From this schema it is evident that a real Iusion oI narrator and protagonist
can only occur when the time oI narration and the time oI the story itselI are
'simultaneous. It is easy to demonstrate that the narrative structure oI the
Satvrica is unrelated to this avant-garde modern type. The dominance oI the
preterite in the Satvrica`s narrative clearly excludes it Irom being identical in
structure with Robbe-Grillet`s early novels, or the so-called French 'New
Novel. In these works the narrator is the protagonist, and literally tells the
story as it happens. Furthermore, since the Satvrica is not an interrupted past
tense narrative, such as the 'novel by letters or the 'private diary, nor in
any sense predicting Iuture events, we are unsurprisingly leIt with only one
slot to Iit it in, namely the classical past-tense narrative.
Much has sometimes been made oI the 'Iact that the narrator oI the
Satvrica never alludes to what lies ahead in the work.
398
Not only is this
claim discrepant with the evidence, as is shown by the Iollowing statement
close to the middle oI the Cena episode: nec adhuc sciebamus nos in medio
lautitiarum, quod aiunt, clivo laborare ('we didn`t yet know that we were
struggling, as they say, halI way up the slope oI delicacies, 47.8), but the
very assumption that such proleptic statements are necessary in narrative in
the Iorm oI personal recollections is not well Iounded. They are indeed rare
in the Greco-Roman erotic Iictions as a wholeiI we exclude indirect pro-
leptic statements in prophetic oracles and dreamsand this is one oI only
two in the extant Satvrica.
399
Anticipating what comes later is hardly a requi-
site oI any story, although the Iigure may oI course prove useIul in certain
narrative circumstances. From the standpoint oI narrative Iorm, the preterite
in the Satvrica maintains throughout an unmistakable temporal gap between
the time oI narration and the time when the events oI the story supposedly
took place. Even where the narrator uses the present tense Ior narrating past
events (a Iigure which momentarily creates an impression that the two are
existing at the same moment), this does not constitute a true Iusion oI the

397
Genette 1972, 217.
398
Jones 1987, 816; and Slater 1990, 46, who concurs in this view.
399
Another technically proleptic statement in the Satvrica is 70.8, pudet referre quae secuntur
('One is ashamed to tell what Iollows), see discussion in 1.2.4.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

197
two personae oI Encolpius.
400
This use oI the present tense to intensiIy the
discursive report is part and parcel oI the art oI narration and does not aIIect
the basic rule oI the preterite.
Most recently critics have returned to the narratological analysis oI
Beck`s article (1973), and have thereIore made renewed attempts at under-
standing the psychology oI the Satvrica`s 'Iirst person Iorm. Gareth
Schmeling deIines the tone and stance oI Encolpius as that oI the 'glorious
conIessor who Irom some psychological compulsion, deIined in modern
times, admits to crimes that he hasn`t committed.
401
For Gian Biagio Conte
the character oI Encolpius is deeply rooted in the mandarin educational sys-
tem oI antiquity, which has made oI him a literary 'mythomaniac who iden-
tiIies so strongly with the artistic representations oI mythological heroes that
he loses all touch with the reality oI the present day.
402
However, as we have
seen in section 1.2, the term persona accommodates conceptually not just the
psychological character, but even more the social type and situation (includ-
ing the audience to be addressed) in which the speaker Iinds himselI at the
moment oI uttering the speech. When considered thus, the diIIerence be-
tween narrator and protagonist in the Satvrica does not depend on our suc-
cess or Iailure in diagnosing the psychological conditions oI these present
and past selves oI the same individual.
There is a much more Iundamental and pragmatic diIIerence: While En-
colpius qua narrator is telling his audience a story Irom memory, and pre-
tends at least to remember almost everything that has happened to him
within a certain period oI his past, Encolpius qua protagonist is an agent
stuck in a given moment in time, both completely ignorant oI what lies
ahead, and generally not very resourceIul in dealing with people and events.
There need be no doubt that the recollections oI Encolpius, towards the end,
reached their conclusion by picking up theme phrases or, at least, ideas Irom
the prologue in order to resume the present occasion oI narration. The Iorm
oI the classical travelogue, which originates in written Greco-Roman litera-
ture with Odysseus` Phaeacian tales, dictates that Encolpius survived to re-
turn and tell his tale; and it is likely, although impossible to prove, that at the
end oI our Latin Satvrica our 'hero reached Rome, the most obvious loca-
tion oI his narrative perIormance in Iront oI a recognizably Roman audi-
ence.
403


400
As Jones 1987, 819, seems to indicate. Dowden 1982, 2930 and 45n., argues Ior a similar
Iusion oI Lucius, the auctor, and Lucius, the actor, in Apuleius` Asinus Aureus.
401
Schmeling 1994/1995.
402
Conte 1996.
403
The only exception to this rule is the narrative oI Achilles Tatius` Clitophon, who notori-
3 GENRE

198
In Greco-Roman prose Iictions, as a rule, the prologue is used to provide
important inIormation about the external circumstances under which the
story is told, and thus we most sorely miss this part oI the Satvrica. In the
prologues to the two Iully extant ancient personal recollections, Clitophon`s
narrative in Achilles Tatius and Lucius` narrative in the Latin Metamor-
phoses, the narrators are clearly presented as distinct Irom their younger
selves. Both the time and location oI the narrative versus the events oI the
story leave no doubt that the narration oI the story takes place aIter the
events are Iully over. In Lucius` case, the Latin oI the narrative Iurther dis-
tances his telling oI the story Irom the events themselves, as they supposedly
took place in a Greek-speaking environment.
3.1.3 Recollection vs. Action
Narratives which take the Iorm oI recollection diIIer in one important way
Irom narratives in which the relation oI the storyteller is less clearly deIined
with respect to the characters and events oI the story. Since a recollection
narrative is based on the memory oI one oI the characters, it cannot, without
accounting Ior it especially, introduce material which was not available to
that person as data at the time.
404
Although this may seem to put great con-
straint on the narrator, there is a hidden advantage. By divulging to the
reader no more inIormation than he himselI had to act on at a given moment,
the narrator leaves the reader to speculate continuously about the meaning
and direction that events are taking. This, in turn, makes the reader more
likely to appreciate the complexities oI the moral and practical problems
which the hero must Iace, and thus creates an audience disposition advanta-
geous to the narrator, whose past Iollies are oIten present liabilities.
405
The
audience cannot judge the protagonist on the basis oI inIormation that they
do not have, which assures a Iairer trial Ior the hero (and thereIore narrator)
who is necessarily under their scrutiny.
406


ously does not return to the initial moment oI narration. However, Clitophon is Irom the
point oI view oI narrative Iorm a subordinate narrator, since the 'author introduces him be-
Iore he begins his story, a Ieature which made problematic the Iull return to the initial mo-
ment. See Most 1989, 11433.
404
See Reardon 1994, 8182, whose analysis oI the narrative method oI Achilles Tatius, i.e., oI
Clitophon`s narrative within that work, reveals the same basic structure.
405
Heinze 1915 speaks oI 'a Ieeling oI uneasy tension (ein Gefhl unruhiger Spannung)
caused by the restrictions imposed on the authorial perspective during the delivery oI the
personal account oI Aeneas in the Aeneid, books 23.
406
See Most 1989, 11433. Recollections are, oI course, related to such genres as Iormal
3. 1 ANCI ENT NARRATI VE I N PERSONI S

199
In Encolpius` love story, the narrator`s continuing blindness with respect
to his own obsessive behavior, not just towards Giton but also towards those
who threaten to take him away, could potentially diminish the fides oI his
audience. For example, a moral critic oI Encolpius` account could argue that
what introduces the many rivals Ior Giton`s love is not at all the boy`s
treacherous nature, but Encolpius` own sexual opportunism (at least in the
cases oI Tryphaena and Ascyltos). He is simply hypocritical, and violates
that Iidelity which he expects in Giton and sets as the model oI their rela-
tionship. Both narrator and protagonist remain quite oblivious to this disso-
nance. Indeed, Irom a modern psychological standpoint, it`s one oI the re-
spects in which Encolpius remains immature.
II we adopt the point oI view oI reading oriented criticism, the style oI
narration practiced by Encolpius is a deliberate gradual Ieeding oI inIorma-
tion to the audience as the story progresses, never making them privy to his
Iull knowledge about the outcome oI events, until each scene has been nar-
rated Iully and in detail. We might call this the principle oI adhuc.
407
An
example oI this would be when the narrator represents his youthIul selI as
proclaiming an emotionally involved speech about the human condition over
an unknown corpse he Iinds on a beach: adhuc tamquam ignotum deflebam
('I was still crying over him as somebody I did not know, 115.11). There is,
however, no absolute necessity compelling the narrator to do this, and as we
have seen, at least once in the Satvrica the narrator makes an exception
which proves the rule (47.8). Here, exactly as the narrator says, the young
protagonist and his Iriends did not know at that time that they were only halI
way through the dinner party, although by the time he tells the story Encol-
pius knows this Iull well, because he has long since leIt that dining room and
thereaIter besides done many other things (some oI which we can read about
in the latter part oI the extant Satvrica). At this particular point Encolpius
uncharacteristically wishes the audience to share in his knowledge oI the
hero`s Iuture, which is nevertheless in the narrator`s past, since this knowl-
edge can be used to give a sense oI the excessive quantity oI Iood oIIered at
the party. The guests were already bursting, when halI oI the Iood was
served. Limited as this inIormation may be, it still constitutes a revealing
exception Irom the constraints on narrative inIormation in the Satvrica, an
exception that proves the rule.
Just as the narrator can, technically, leap ahead into the Iuture oI the
protagonist (at least into that part oI it which is still in the narrator`s past), he

apologies and the deIendant`s speech.
407
CI. the use oI the adverb adhuc, still`, in 11.2, 13.1, 17.1, 33.3, 54.3, 70.4, 96.3, 99.5,
106.2, 113.7.
3 GENRE

200
can also jump back in time and reIer to a past which is shared both by him-
selI and the protagonist. We might call this the principle oI the paulo ante.
An example oI this would be when the narrator represents his youthIul selI
as standing on the beach over that same corpse aIter he recognizes that it is
the body oI his enemy Lichas: agnovique terribilem paulo ante et implaca-
bilem Licham pedibus meis paene subfectum ('and I recognized Lichas, who
a short while ago (paulo ante) had been so IrightIul and unIorgiving, now
lying prostrate, as it were, beIore my Ieet, 115.11).
408
The ability oI the
narrator to leave the present oI the protagonist and go back in time seems,
however, less interesting than his leaps into the Iuture, because it does not
distinguish him Irom the protagonist, who can and does accomplish such
Ieats as well (115.12). It is important to note that here the phrase paulo ante
reIers to a point in time just beIore the protagonist`s present, and yet it is the
narrator and not the protagonist who is speaking. The Iixation on the past is
the rule in the classical past-tense narrative. In modern narratology this obvi-
ous Iixation on a particular character and moment in the past is usually re-
Ierred to as a 'point oI view, or as 'Iocalization through that character,
both oI which are metaphors taken Irom the modern art oI photography.
To observe directly the contrast between young Encolpius` continuous
heuristic progress Irom ignorance to knowledge as opposed to the narrator`s
prior knowledge oI the outcome oI events, we can look at the passage,
shortly aIter the opening oI the extant Satvrica, where the narrator tells us
that he leIt Agamemnon reciting in the portico and ran aIter Ascyltos, whom
he suspected oI not being the best oI Iriends. Exhausted and completely lost
in an unknown city, as a last resort, he approached an old street vendor to
ask Ior directions, not truly expecting that she could tell him where he lived:

'rogo inquam 'mater, numquid scis ubi ego habitem? delectata est illa
urbanitate tam stulta et 'quidni sciam? inquit consurrexitque et coepit
me praecedere. divinam ego putabam et subinde ut in locum secretiorem
venimus, centonem anus urbana reiecit et 'hic inquit 'debes habitare.
cum ego negarem me agnoscere domum, video quosdam inter titulos nu-
dasque meretrices Iurtim spatiantes. tarde, immo iam sero intellexi me in
Iornicem esse deductum. execratus itaque aniculae insidias operui caput
et per medium lupanar Iugere coepi.
409



408
CI. the use oI the phrase paulo ante, a short while ago`, in 16.3, 49.3, 74.5, 80.8, 96.1,
137.12.
409
Sat. 7.14.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

201
I said, 'Listen, mother, you wouldn`t happen to know where I live?
She, charmed with such stupid wit, answered: 'And why shouldn`t I?
then she stood up and began to lead the way. I took her to be some sort
oI apparition, and when we came to a locale more out oI the way, the
humorous old woman drew back the curtainand said: 'This should be
where you live.While I was saying that I didn`t recognize the place as
my home, I see some men tip-toeing Iurtively amid written signs and na-
ked whores. Slowly (or rather now when it was too late) I understood
that I had been led to a whorehouse. I immediately cursed the trickery oI
the old lady, covered my head, and took to Ilight through the center oI
the bordello.

The Iundamentally diIIerent cognitive status oI narrator and protagonist
emerges clearly Irom this recollection oI a past incident (the present tense in
video is without temporal signiIicance). The narrator`s description oI the old
woman`s reaction to the question, and especially the mocking qualiIication
oI that question as urbanitas tam stulta, signals that the mood and under-
standing oI the narrator is just the opposite oI that oI the desperate and help-
less protagonist. The split continues: While young Encolpius like an epic
hero thinks the old woman is a god in human Iorm (divina) because she
knows where he lives and is willing to show him the way, the older Encol-
pius knows Iull well that something else than divine protection or human
altruism is behind the apparently good deed. His knowledge is signaled by
the anticipatory qualiIication oI the old woman as urbana. The contrast be-
tween the absolute navete oI the one and the knowing amusement oI the
other could not be clearer. The realization oI what is happening comes pain-
Iully slowly to the youth, and only gradually does he Iully recognize that he
is being led on and insulted (tardo, iam sero intellexi). For some reason,
however, we detect little or no resentment in the narrator`s account oI this
humiliating incident. There seems rather to be in him a clownish enjoyment
oI how easily he himselI was taken in and how silly he was to trust that old
practical joker.
Another passage, this time Irom the Cena, provides Iurther illustration oI
the gap between the narrator`s knowledge and the protagonist`s ignorance oI
events in the Iuture oI the latter. This passage is the narrator`s account oI his
own puzzled reaction to one oI the many deceptive articles oI Iood oIIered at
Trimalchio`s table:

|.| gustantibus adhuc nobis repositorium allatum est cum corbe, in quo
gallina erat lignea patentibus in orbem alis, quales esse solent quae incu-
3 GENRE

202
bant ova. accessere continuo duo servi et symphonia strepente scrutari
paleam coeperunt erutaque subinde pavonina ova divisere convivis. con-
vertit ad hanc scaenam Trimalchio vultum et 'amici, ait 'pavonis ova
gallinae iussi supponi. et mehercules timeo ne iam concepti sint; temp-
temus tamen, si adhuc sorbilia sunt. accipimus nos cochlearia non mi-
nus selibras pendentia ovaque ex farina pingui figurata pertundimus. ego
quidem paene proieci partem meam, nam videbatur mihi iam in pullum
coisse. deinde ut audivi veterem convivam: 'hic nescio quid boni debet
esse, persecutus putamen manu pinguissimam Iicedulam inveni piperato
vitello circumdatam.
410


. we were still busy with the hors d`oeuvres, when a tray was brought
in with a basket on it, in which there was a hen made oI wood, spreading
out her wings as they do when they are sitting. The music grew loud: two
slaves at once came up and then hunted in the straw. Peacock`s eggs
were pulled out and handed to the guests. Trimalchio turned his head to-
wards this perIormance, and said: 'I gave orders, my Iriends, that pea-
cock`s eggs should be put under a common hen, and by Hercules I`m
aIraid they might now be addled. However, let us see iI they can still be
sucked. We took our spoons, halI-a-pound in weight at least, and ham-
mered at the eggs made out of flour and fat. I almost threw away my por-
tion. I thought a peachick had already Iormed. But hearing a practiced
diner say, 'What treasure have we here?, I poked through the shell with
my Iinger, and Iound a very Iat Iig-pecker, rolled up in spiced yolk oI
egg.

I quote these two passages in Iull, because the linear experience oI reading
them is not easily described and must be experienced. Again we notice that
the narrator communicates the essential Iacts to the audience (ovaque ex
farina pingui figurata pertundimus) several lines beIore the protagonist has
Iound out that the eggs are not real. Once this inIormation has been divulged,
dramatic irony kicks in and the painIully slow understanding oI the hero is
made all the more evident. This is clearly the intention, as can be seen Irom
the phrase: 'I almost threw away my portion, which selI-consciously exag-
gerates young Encolpius` clownishness beyond what actually happened at
the time.
It would try the reader`s patience iI we kept repeating the obvious, and I
Iear that I have demonstrated all too well that the strongest argument against

410
Sat. 33.38.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

203
the close scrutiny oI narrative technique is how Iundamentally mechanical
the method is revealed to be when applied to a speciIic literary text. In Iact,
the method does not allow us to get much beyond its a priori premise, unless
we introduce the question oI literary genre.
3.1.4 Narrative in personis as the Mark oI Genre
Surprising as it is, the scholars who have most studied the implications and
Iormal Ieatures oI the 'personal narrative in the Satvrica rarely bring this
quality to bear on their discussion oI the literary genre. Moreover, it is a
common assumption that all 'personal recollection narratives are essentially
the same, which justiIies the Irequent comparison oI the work with modern
texts which display some Ieatures in common with it. As we showed in
section 1.2, however, the 'personal recollection narrative oI Encolpius is
signiIicantly diIIerent Irom any modern narrative, and modern novels (as
well as the method developed Ior reading them, modern narratology) are
thereIore oI very limited use as 'comparison texts Ior the student oI the
Satvrica.
What remains is the Iact that the closest analogous literary work in the
whole oI extant literature, and one that has almost exactly the same narrative
Iorm as the Satvrica, is, and always has been, Apuleius` Metamorphoses.
Other long 'Iirst person narratives in ancient epic or Iiction do not display
the same structure, and they are always subordinate to a main narrative
which is conducted in the name oI the author. Although the novel oI Achilles
Tatius is practically a personal recollection narrative, iI we leave out the
initial introduction oI the narrator by the author, that text does not display
the variety and extent oI impersonations oI subordinate narrators so peculiar
to the Satvrica and the Metamorphoses.
411
Let us remember that in the Sa-

411
Which is not to say that Achilles Tatius has not been inIluenced by the same structure.
Clitophon`s own introduction oI his narrative, 'you are poking up a wasp`s nest oI narra-
tive, my liIe has been very storied (1.2, '!f3-. J38/8+%8&., 8rG8, ,F/43l #V /V% 0V
=H-&. Z-&'8), includes the mention oI a plurality oI ,F/-& and =H-& which is reminiscent
oI the fabulae oI the Latin works. However, no such plurality oI stories and impersonations
is oIIered in the narrative oI Clitophon. In Achilles Tatius we do have occasional separate
stories and speeches (the slave`s report oI Charicles` death in 1.12; Satyros` Aesopic Iable
in 2.2122; Menelaos` love story, in 2.34, Iollowed by the debate on the relative merits oI
male- and Iemale-directed love, in 2.3538; Clinias` account oI himselI in 5.910, and his
court speech in 7.9, and Thersandros` reply in 7.11; and Iinally, the priest`s account oI the
secrets oI the syrinx, in 8.6, and the speeches at the end oI that book), but these are mostly
short and more closely related to the intrigue oI Clitophon`s and Leukippe`s love story.
3 GENRE

204
tvrica this elastic and yet highly structured genre can accommodate the long
poem oI Eumolpus, the so-called Bellum Civile, and in the Metamorphoses
it can accommodate the long Iable oI the old hag, the so-called Cupid and
Psvche. As a mark oI genre, the perIect match oI the structure oI personal
recollection in personis, common to both works, should be viewed as more
reliable than the prosimetry oI the Satvrica, which we know too little about,
and is Iound in too many types oI works to be a helpIul marker oI genre.
Recently Gian Biagio Conte has reiterated the Iutility oI assuming an identi-
Iication between prosimetry and 'Menippean satire, an identiIication which
presupposes that prosimetry is a reliable marker oI one and only one genre,
and not a mode oI discourse possibly common to several literary types, as
seems the obvious conclusion to draw especially since the discovery oI the
Iolaos and Tinouphis papyri.
412
By replacing the prosimetry with the narra-
tive Iorm as the identiIying generic Ieature oI the Satvrica we accomplish
nothing less than to eliminate the Iruitless notion oI generic duality or plural-
ity which has paralyzed the study oI this text and led to its continued disinte-
gration, as it were, in the hands oI scholars who have earnestly attempted to
make sense oI the problem. I stress that by classiIying the Satvrica and the
Metamorphoses in the same genre I am not attempting to perpetuate the
nineteenth-century notion oI a Roman national novel, to be contrasted to a
Greek one (see below Ior a Iull discussion oI this topic). As a matter oI Iact
it is my intention to argue that the genre is as Greek as were the lost
:8#"-%);*8&., the Greek text adapted by Apuleius when he wrote the
Latin Metamorphoses.
BeIore we go any Iurther, however, we need to lay out the narrative
model in the Metamorphoses, just as we did Ior the Satvrica at the beginning
oI our discussion oI the Iorm. As beIore, the name/mask oI the main narrator
is marked by caps and the names/masks oI the subordinate narrators by quo-
tation marks. This work oI course is preserved complete and so the thread oI
the narrative is neither broken at the beginning nor the end.

LUCIUS (1.15) 'Aristomenes LUCIUS (1.202.5) 'Byr-
rhena LUCIUS (2.56) 'Lucius LUCIUS (2.711)
'Lucius LUCIUS (2.13) 'Milo LUCIUS (2.1521) 'The-
lyphron LUCIUS (2.313.3) 'anonymous prosecutor
LUCIUS (3.4) 'Lucius LUCIUS (3.814) 'Photis
LUCIUS (3.194.8) 'anonymous robber LUCIUS (4.2227)
'old woman LUCIUS (6.2528) 'Charite LUCIUS (6.2931)

412
Conte 1996, 140I.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

205
'anonymous robber LUCIUS (6.327.1) 'anonymous robber
LUCIUS (7.25) 'Haemus the Thracian LUCIUS (7.1020)
'boy in charge oI the ass LUCIUS (7.22) 'one oI the coun-
trymen LUCIUS (7.2427) 'boy`s mother LUCIUS (7.28
8.1) 'Charite`s slave LUCIUS (8.1519) 'old man
LUCIUS (8.219.16) 'old hag LUCIUS (9.2224) 'baker
LUCIUS (9.2610.8) 'physician LUCIUS (10.1214) 'broth-
ers LUCIUS (10.1511.1) 'Lucius LUCIUS (11.34) 'Isis
in Lucius` dream LUCIUS (11.714) 'priest oI Isis LUCIUS
(11.1624) 'Lucius LUCIUS (11.2529) 'Osiris in Lucius`
dream LUCIUS (11.30).

As we can see the general rule is that the subordinate and more or less sepa-
rate stories told by Lucius in the Metamorphoses are narrated in personis. A
Iew, however, are inspired simply by the places visited by Lucius and in rare
instances he does not bother, apart Irom associating them with a particular
spot on the journey, to account Ior their origin. Here there is no impersona-
tion. An example oI this would be the brieI story oI the bailiII`s punishment
Ior adultery (8.22); another is the retelling in Lucius` own person oI the tale
oI the cuckold and the corn-jar, but this one he heard in the inn oI the town
where it supposedly happened (9.57). The story oI the crazed estate-owner
is also told by Lucius in his own voice on the basis oI an eyewitness report
he heard in the past (9.3538). Another such case is a crime recorded by
Lucius (10.212), although that story does Ieature the impersonated speeches
oI the physician (10.89; 11), despite Lucius` statement that in his manger
he was not in a position to hear the main speeches oI the prosecutor and de-
Iendant, and so could not write them down Irom memory Ior the enjoyment
oI his reader (10.7). And Iinally the story oI the crime oI the condemned
woman (10.2328), although said to have been 'heard by Lucius, is simi-
larly told without impersonation. All oI these, however, are woven into the
travelogue oI Lucius and in one way or another come to his attention while
he is on the road.
Lucius with his long asinine ears is like a vacuum cleaner that eagerly
sucks up the 'ancient smut he encounters on the trip, or at least such has
been the opinion oI generations oI concerned moral critics.
413
His character-

413
A rare appraisal oI the 'antike Schmutz in the Satvrica is oIIered by Nietzsche in a post-
humously published Iragment, where the philosopher, aIter comparing Iavorably the ex-
perience oI reading the Satvrica to that oI reading the New Testament, poses the Iollowing
question: 'ist nicht der antike Schmut: noch mehr werth als diese gan:e kleine anmaaliche
Christen-Weisheit und -Muckerei?` See Nachgelassene Fragmente, Herbst 1887 bis Mr:
3 GENRE

206
istic 'curiosity is thereIore not just a moral Iault, it is also what makes him
Iit Ior his role oI gathering this store-house oI narratives to record Ior his
reader (9.15). The scrupulous reader in turn is assumed to be just as inquisi-
tive as Lucius. This is clearly indicated by his initial promise to caress the
reader`s ears with a delightIul string oI stories. As narrator and writer oI the
Metamorphoses Lucius represents his past selI as having been constantly on
the look-out Ior material that could later be used Ior the book (10.2). As he
relates how he rushed back to the house oI Milo to meet his Iate at the hand
oI the witches, he represents himselI as excitedly uttering these words to
himselI: 'O Lucius, come now, |.| the opportunity you have been waiting
Ior has arrived; you can have your heart`s Iill oI marvelous stories, as you
have always wanted.
414
It would appear that he is not so much driven by a
quest Ior the Iorbidden knowledge oI magical transIormations as simply
hungry Ior stories.
A certain pact is established by Lucius with his reader Irom the very
start. Not a serious contract this one, but a game where the rules are that
Lucius can lie as much as he wants, so long as he respects certain norms oI
conduct. One oI these rules is that he must roughly account Ior the source oI
his stories, since this is seen to be a sort oI guarantee oI their authenticity. At
one point in the narrative Lucius as an ass is conIined within the walls oI a
bakery, and yet he claims that despite the limitations put on his Ireedom Ior
snooping around, he still was able to gather important material. This calls Ior
Lucius` eIIort to tackle an anticipated protest Irom the reader that he has
perhaps broken the pact and is now Ireely inventing stories, instead oI
merely reporting what he heard on his travels (9.30). By raising this issue the
narrator re-establishes the rules oI the game and reaIIirms his commitment to
the general constraints on inIormation that apply in recollection narratives.
II we take a brieI look at how ancient narrators oI Iiction account Ior
their relationship with the characters and material oI their stories, we see that
a variety oI well deIined stances was developed. While writers oI modern
Iiction are usually under little constraint to account Ior the source oI their
stories, the ancient Iabricator oI Iiction Ielt that he owed the reader an expla-
nation. The author oI the Apocolocvntosis, which takes place in heaven and
hell, somehow had to account Ior his knowledge about aIIairs in places
which were naturally inaccessible to him. He thereIore wittily pretended to
have a 'source Ior his inIormation in a man, an historical individual, who
had become notorious in the times oI Caligula Ior swearing to the senate that

1888 10 |93| (213); in Colli and Montinari 1970, 8:2, 1756.
414
Met. 2.6, 'O Luci, [.] Habes exoptatam occasionem et voto diutino poteris fabulis miris
explere pectus.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

207
he saw Julia, the emperor`s sister, ascend to heaven. Lucian, in the prologue
to his Ireely invented Jera Historia, similarly exposes the convention in a
meta-literary joke. He starts by expounding the long tradition in Greek letters
oI mendacious story-telling, Iounded by Homer in the Odvssev, and then
makes the conIession once and Ior all that the subsequent story presented in
the Iorm oI an autobiographical travelogue is pure invention in accordance
with this venerable tradition! Everywhere in ancient Iiction we meet with
this requirement to acquire a license or establish a source Ior the Iictional
inIormation.
415
Mysterious documents, Iound in an ancient tomb, are said to
guarantee the story truthIulness, as in Antonius Diogenes` tale The Wonders
Bevond Thule. The same trick is used by 'Cornelius Nepos, the Latin
'translator oI the contemporary Phrygian history oI the Trojan War written
in the hand oI one Dares who actually Iought in that Iamous conIlict. Some-
what related is the discovery in Daphnis and Chlo oI a pictorial history, an
authentic relieI that illustrates how everything happened, the narrative oI
Longus being simply the interpretation oI the pictures oIIered by a local
guide. Less obvious and more conceited are those story-tellers who hide
behind the imposing authority (and general lack oI accountability) oI ancient
historiansquis unquam ab historico iuratores egit? (Apoc. 1)as does
Chariton in his Callirho, which might be seen as a Iictional digression Irom
Thucydides` account oI the Sicilian expedition. The inventor oI such tricks
(according to Lucian) was Homer himselI, and iI we look at the epic conven-
tion, we see that there it is the assistance oI omniscient divine Muses which
guarantees the accuracy oI the story. When particularly detailed passages
occur, the goddesses are re-invoked in order to Iorestall doubt or disbelieI
among the audience. In more pragmatic ways memorv (the mythological
mother oI the Muses) is also the source oI narrative authority in autobiogra-
phies, real or Iictive. When Encolpius, Lucius and Clitophon casually reIer
to their memory during the narration oI their stories, this too is an attempt to
induce the reader to accept the account as real and reliable.
But as eager as he may be to Iill his mind with stories, Lucius in the
Metamorphoses is not interested in just any story, but has a clear preIerence
Ior the titillating, wondrous and horriIying (2.6). Thus he does not like his
mean-spirited host, a staunch realist who is completely Iree oI either adven-
turous or entertaining impulses. In an easily recognizable example oI
Menippean humor, Lucius describes Milo`s unsatisIying, loquacious and

415
Even Odysseus must account Ior his inIormation about what was spoken by gods in heaven:
'This I heard Irom Iair-haired Calypso, and she said that she herselI had heard it Irom the
messenger Hermes (Od. 12.389I.). The same requirement to account Ior inIormation con-
ditions the elaborate beginnings oI Plato`s Parmenides and Phaedo.
3 GENRE

208
Iamished banquet (loquax et famelicum convivium), where instead oI Iood he
dined on nothing but stories (1.26, cenatus solis fabulis). On another occa-
sion it is the waste oI time and the disillusioned style oI the series oI inop-
portune Iables brought on by his host which make him groan and become
angry with himselI Ior not having leIt his company earlier (2.15).
Not every story accordingly is pleasing to everybody, though the rule
applied to storytelling in the Metamorphoses is clearly that stories are or at
least ought to be entertaining. In the prologue oI the Jera historia oI Lucian,
the justiIication oIIered by the author Ior his Iabricated story is that just like
athletes need to relax their bodies in between strenuous exercises, so phi-
lologists need to intersperse their serious studies with light reading that will
relax their mind and prepare it Ior Iuture labor. But although certainly light,
Lucian adds that such reading should also contain some Iood Ior thought,
something entertaining too Ior the Muses. In Macrobius` commentary on
Scipios Dream, which provides the only positive generic classiIication oI
the Satvrica that has come down to us Irom antiquity,
416
Petronius` literary
creation is put side by side with Apuleius as a writer oI entertaining erotic
Iiction. In the absence oI other ancient classiIications oI the work, modern
critics are required to take this one very seriously. Macrobius is trying to
argue that Plato`s (and by extension Cicero`s) philosophical Iictions are not
just lies (as the Epicureans had said), but diIIerent and more noble than mere
Iiction as entertainment. He sets out a division between types oI Iiction to
explain what is acceptable in a philosophical treatise:

There are two kinds oI Iables (the word indicates an admission oI Ialse-
hood): the sort that aims only at pleasing the ears, and the sort that is in-
vented also as an exhortation to virtue. In the class that aIIords pleasure
to the listener we put comedies oI the sort that Menander and his imita-
tors produced on the stage, and stories oI everyday liIe (argumenta)
crammed with the Iictitious Iortunes oI lovers, at which (Petronius) Arbi-

416
The other generic classiIication oI the Satvrica to survive Irom antiquity does not assign the
work to any known genre. John the Lydian, in Mag. 1.41, lists Petronius aIter Turnus and
Juvenal claiming that all three have violated the *"#$%&'Q. 3F-.. Both the comparison
with these two writers oI Latin hexameter satire and the term used by John the Lydian are
puzzling. The generic class *"#$%&'Q. 3F-. cannot mean just satire`, Ior then Juvenal`s
satires could not be said to violate the genre! AIter all, according to the modern understand-
ing, Juvenal`s work, perhaps more than any other writer`s, can be said to deIine the genre oI
Roman satire. The Greek term *"#$%&'Q. 3F-. must thereIore reIer to a Greco-Roman lit-
erary tradition considerably wider than what we are accustomed to call satire`, and include
Greek genres such as satyr plays. Even so, we cannot really answer the question how the
Satvrica relates to John the Lydian`s broadly deIined genre.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

209
ter exercised himselI greatly and in which Apuleius, amazingly enough,
occasionally indulged. This whole species oI Iables, which oIIers mere
delights Ior the ears, a philosophical treatise expels Irom its sacred shrine
and banishes to the nursery room. However, those stories which exhort
the reader`s intellect towards some Iorm oI virtue .
417


Macrobius rejects the Satvrica and the Metamorphoses as mere aural titilla-
tion, because he is concerned with the genre oI the philosophical treatise and
its uncompromising search Ior truth and exhortation to virtue. This is a diI-
Ierent standard altogether Irom Lucian`s requirement that Iictitious stories
should not Irom urbanity and humor oIIer mere entertainment (1 F3-3 0'
#-T J*#8+-$ #8 '"R N"%+83#-. m&,13 G"%AW8& #13 m$N"/4/+"3), but also dis-
play some vision that is not uneducated (J,,V #&3" '"R H84%+"3 -e' P-$*-3
0G&B8+W8#"&). It is easy to lose sight oI this distinction between what is mor-
ally ediIying (and deIensible in a philosophical treatise) and what is regard-
less oI that austere standard highly enlightening and educated. The presence
oI jest and lack oI moral preaching does not oI course preclude a masterIul
artistic vision, notwithstanding the concerns oI sclerotic moralists.
II Macrobius rejects the moral or philosophical value oI Petronius` and
Apuleius` Iictions, modern readers have long sought ediIying, or at least
truthIul, statements in these works.
418
Let us do the same Ior the Satvrica. To
establish the genre oI the ancient 'personal novel, we must go beyond the
Iorm, and seek satiric or satyric content.




417
Macrob. Comm. 1.2.78. Fabulae, quarum nomen indicat falsi professionem, aut tantum
conciliandae auribus voluptatis, aut adhortationis quoque in bonam frugem gratia repertae
sunt. auditum mulcent vel comoediae, quales Menander eiusve imitatores agendas dede-
runt, vel argumenta fictis casibus amatorum referta, quibus vel multum se Arbiter exercuit
vel Apuleium non numquam lusisse miramur. hoc totum fabularum genus, quod solas
aurium delicias profitetur, e sacrario suo in nutrium cunas sapientiae tractatus eliminat. ex
his autem quae ad quandam virtutum speciem intellectum legentis hortantur .
418
Traditionally there are three schools oI thought regarding satire in the Satvrica. Scholars
who tend to emphasize the earnestness oI the moral satire include Highet 1941, Bacon
1958, Reith 1963, Arrowsmith 1966, and Zeitlin 1971 and 1971a. Scholars who stress the
comic and non-moralistic nature oI the work include Sullivan 1971 and Walsh 1974. The
middle ground is taken by Sandy 1969 and Beck 1982, who like myselI believe that the Sa-
tvrica is a comic satire which does not preach its message directly. For a survey oI early re-
actions to Apuleius` levity or seriousness in the Metamorphoses, see Harrison 2002.
3 GENRE

210
3.1.5 The Moral oI the Genre
Unlike the rest oI the Satvrica, the Cena is oIten supposed to illustrate a
veritable scandal oI the early empire, which was the excessive wealth oI
Ireedmen. That the depiction oI Trimalchio`s dinner-party is intended as a
scandalous account is undoubtedly correct; however, as we have seen
through the reconstruction, the same could be said oI several other episodes
which satirize such popular causes oI complaint as the sorry state oI rhetoric,
the venality oI the courts, the abuse oI religious cults Ior Iinancial and sexual
exploitation, the abuse oI the private ergastula, the abuse oI the right to Iorm
guilds, husbands` complaints over their wives` supposed sexual attraction to
strangers. It seems that the Cena was not so diIIerent Irom the rest oI the
work aIter all. A scandalous and sensational narrative is usually entertaining
as such and the emphasis is obviously on making the most oI that element oI
entertainment, rather than delivering an outright moralistic condemnation.
Passing a judgment with moral authority is leIt up to the ideal audience, who
are eminently capable oI doing so, having been created as the implicit 'nor-
mal subjectivity which witnesses the narrator`s comic act.
One is certainly struck by the limited apology oIIered by the narrator Ior
the pathetic perIormance oI his past selI. Narrating his experience, he seems
particularly conscientious when reporting the verbal abuse to which he was
subjected by other characters. We recall, Ior example, his quarrel with
Ascyltos early on, Irom which he emerges an exposed hypocrite and branded
as homo stultissimus (9.2-10.3). Far Irom trying to cover up the disgrace oI
this deIeat by attempting to ennoble his intent, the narrator announces
bluntly that the only reason why his young selI so hastily sought divorce
Irom his Iriendship with Ascyltos was 'lust Ior Giton (10.7, hanc tam
praecipitem divisionem libido faciebat). Another example oI such deleteri-
ous testimony about himselI comes a Iew pages later, when Encolpius re-
ports that Quartilla mocked him as 'a brilliant guy and a real source oI
homegrown wit (24.2, homo acutus et urbanitatis vernaculae fons). Several
times does he describe in detail his stupeIied astonishment at Trimalchio`s
house.
419
Once he tells his audience that his Iriends laughed at him Ior pan-
icking at the sight oI a painted dog and the sign which said cave canem
(29.1-2). An obviously clownish reaction to a common enough phenomenon,
as any modern visitor to the archeological sites oI southern Italy can testiIy
(although we should keep in mind that Encolpius qua hero is an exile Irom
Massalia and unIamiliar in the area). Later, when he must conIront the real

419
Beck 1975, 277-278.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

211
dog, he again exaggerates the extent oI his terror at this inIernal beast, by
promptly reminding his audience that he had even been aIraid oI a painted
dog. He likewise tells oI how Irightened he was by the 'majestic entrance
oI the stonemason Habinnas, a reaction which earned him again the title oI
homo stultissimus (65.5), this time Irom Agamemnon himselI.
How should we explain the narrator`s selI-deprecation and jokes at his
own expense? What is it in the narrative situation which makes this clownish
posturing expedient? It is conceivable that Encolpius has now improved his
social standing, but this would hardly explain why he consistently humiliates
himselI in Iront oI his audience. On the contrary, one would expect any so-
cial elevation to express itselI in a more assertive tone oI voice. It seems that
we can only explain why Encolpius puts himselI down iI we keep in mind
the considerable social inequality inherent in the narrative situation. The
narrative persona oI Encolpius, a (by now) Romanized Greek oI poor origin,
is very much inIerior to his aristocratic Roman audience. It is usually the
social superior who indulges himselI in long-winded narratives about his
own past to an inIerior, and not the other way around. For narrative authority
to work, there must be some such power at play, whatever the basis oI the
reader`s or audience`s respect Ior the author. In addressing himselI to a dis-
tinguished audience, which by Iar outranks himselI socially and morally, no
matter how much he may have improved his lot subsequent to his adven-
tures, Encolpius would have to adopt a humble and clownish persona in
order not to oIIend or bore his discriminating listeners. The Irequently outra-
geous nature oI his story, and the ignoble past which it reveals, would make
such comic posturing even more necessary.
There is, however, another side to his posturing. Throughout the Cena
episode, Ior example, Encolpius tries to earn some points with his audience
by implying that the reason why he made so many dumb mistakes was only
his ignorance oI such vulgarities as took place in that house. This supposedly
noble simplicitas, coupled with his much emphasized disgust with the social
monster Trimalchio, may best be explained as a rhetorical ploy to seduce his
noble audience into believing that he and they, despite everything, share
certain ethical principles. Implicit in this is also a reIerence to Encolpius`
Massaliotic origins. I do not want to exaggerate the narrator`s sophistication
here. He is merely trying what any speaker would try under the circum-
stances, namely to secure the benevolence oI his audience. What allows En-
colpius to 'get away with his satire is his Ioreign provenance and selI-
deprecation, his careIul deIinition oI himselI as a comic Iigure, partaking oI
the inadequacies oI the characters in the narrative and thus no threat morally,
any more than socially, to his presumed elite and cultured audience. We Iind
3 GENRE

212
something similar in Horace`s stance in the Sermones, where mixed with the
sometimes biting social satire the narrator reminds his audience oI his hum-
ble origins as a Ireedman`s son. We should, however, resist the idea that this
ambiguous posturing is distinctly 'Roman, since the generic stance oI sat-
ire, according to the Roman satirists themselves, originated with Greek com-
edy and a type oI popular philosophical diatribe.
There are two obvious social concerns with which Encolpius attempts to
color his narrative. First, the well known anxiety oI Roman aristocrats about
the moral implications oI a widespread study oI Greek rhetoric and literature.
Secondly, the even greater anxiety and threat Ielt by these people in the Iace
oI moneyed individuals Irom the lower classes. These two concerns, which
are prominent themes in the satires oI Horace, Persius and Juvenal, are Iound
in Attic comedy and a wide range oI Greek literature as well. It would be
overly modern to think that the critical distinction to be made here was
merely that between 'Roman and 'Greek values. The study oI national
literatures is a late development in European humanism. Greek aristocrats
were no less apprehensive about the democratic arts oI public speaking and
education Ior the common man than were their Roman counterparts. Neither
is this aristocratic ideology unIamiliar to the Greek narrator, who evidently
received his education in rhetoric and classical letters in his home city, Mas-
salia. As a rule, however, Encolpius is notoriously elusive and un-committed
to speciIic positive values. This attitude clearly derives Irom his marginal
and socially ambiguous status as he communicates his report Irom the un-
derworld across the unbridgeable gap between ancient social strata. He is
speaking to an audience which is Iundamentally diIIerent Irom himselI, and
his only means oI retaining their interest and willingness to listen is to shape
his discourse in conIormity with their values and anxieties. This is his dis-
cursive strategy and the resulting narratorial stance comprises a major liter-
ary conceit in the Satvrica.
Although noble, his audience betrays signs oI decadence and Irustration.
They are willing to believe that the world is going to pieces, precisely be-
cause it has lost the noble values perceived to be traditional in their own
class. In an interesting passage Erich Auerbach attempts to tackle the com-
plexities oI the authorial stance in the Satvrica. It is worth quoting Ior the
insight it contains into the social stratiIication which comes into play in this
text:

Petronius . looks Irom above at the world he depicts. His book is a
product oI the highest culture, and he expects his readers to have such a
high level oI social and literary culture that they will perceive, without
3. 1 ANCI ENT NARRATI VE I N PERSONI S

213
doubt or hesitation, every shade oI social blundering and oI vulgarity in
language and taste. However coarse and grotesque the subject matter
may be, its treatment reveals no trace oI the crude humor oI a popular
Iarce. Scenes . exhibit, it is true, the basest and commonest ideas, but
they do so with such reIined cross-purposes, with such an array oI socio-
logical and psychological presuppositions, as no popular audience could
tolerate.
420


The importance attached to the audience oI the Satvrica by Auerbach is his
genuine contribution to the understanding oI the work. He sees the narrative
as being addressed to a sophisticated audience, radically diIIerent Irom any
oI the lowly characters which appear in the story. Auerbach is not entirely
successIul in positioning the author with respect to the narrator and audience,
but he is clearly sensitive to the problems involved ('Petronius is signiIi-
cantly said to 'look Irom above at the world he depicts). Encolpius and the
audience are not oI the same social rank. It is the 'sociological and psycho-
logical presuppositions oI the audience, rather than the narrator, which sup-
ply the premise oI the social criticism in the Satvrica while the non-assertive
and roundabout way oI delivering it is caused by the narrators inIerior posi-
tion towards them.
3.1.6 Jerkehrte Sprache
As the extant text opens, the narrator is impersonating his youthIul selI, de-
claiming against declamation (1.12.9). What he says is that declaimers are
possessed by an alien kind oI madness (alio genere furiarum declamatores
inquietantur) as they shout their declamations on exaggerated subject-matter
in empty and noisy sententiae (rerum tumore et sententiarum vanissimo
strepitu). Students, moreover, are stultiIied by these exercises, since they
hear and see none oI the customs that are in general use (nihil ex his quae in
usu habemus aut audiunt aut vident). As a result they are not only incapable
oI producing anything but sermo vitiosus ('Iaulty speech, see discussion in
section 1.2.7), they also Ieel as iI they had been transported to another world
when they visit the real courts oI law in the Iorum (cum in forum venerit,
putent se in alium orbem terrarum delatos). The distinction here made be-
tween hearing and seeing those things which are in general use and the vir-
tual disease (veluti pestilenti quodam sidere) oI certain contrived Iorms oI

420
Auerbach 1953, 47.
3 GENRE

214
speech is at the heart oI the conservative 'restraint (mens bona) stylistics
which Encolpius tries to emulate in addressing his audience.
These ideas, however, occur in a speech uttered in the name oI young
Encolpius, and the narrator seems to deliberately undercut his younger selI
when he adds at the end oI the excited and inspired tirade: non est passus
Agamemnon me diutius declamare in porticu quam ipse in schola sudave-
rat'Agamemnon didn`t allow me to declaim any longer in the portico than
he had himselI sweated in the school (3.1). Declamation is exactly what the
young man had criticized most, and yet his older selI reIers to that very criti-
cism as 'declamation. Moreover, when Agamemnon, in response to the
criticism, extemporizes a satire in the style and meters oI Lucilius, attempt-
ing to correct his own and others` parasitic vices and Iailures as educators (5
vv. 122),
421
young Encolpius cannot resist the lure oI the rhetor`s metrical
verbosity (6.1, dum hunc diligentius audio . et dum in hoc dictorum aestu
motus incedo'while I`m listening to him with close attention . and while
I was transported in excitement over this Ilood oI words). So much so that
he Iails to notice that Ascyltos has run out on him (non notavi mihi Ascvlti
fugam). Only later, during his quarrel with Ascyltos (9.1010.3), is he made
to realize that the poetry oI Agamemnon was nothing but 'broken glass and
dream interpretations (vitrea fracta et somniorum interpretamenta), in the
words oI his young Iriend, and that his motives Ior listening to it were less
than noble (multo me turpior es tu hercule, qui ut foris cenares poetam lau-
dasti'by Hercules! you are much less honorable than I am. You praise a
poet to be invited to dinner). He seems thereIore in the past neither to have
had the power to speak without Ialling into the vices oI contemporary dec-
lamation and versiIication nor to have possessed any resistance to the decep-
tive attractions oI these 'perverted arts.
Clearly the narrator is not presenting his younger selI as any sort oI
credible reIormer, whether in stylistic or moral matters. As Ior himselI at the
time oI narrating, he is content with not letting his characters get away with
boastIul claims without proving them wrong immediately. Encolpius is mak-
ing Iun oI himselI in the past and he is Iar Irom excluding himselI Irom the
criticism that he has leveled against the scholastics. He is a speaker who is
willing to give deleterious testimony about his own ineptitude, but he does
this on one extremely important condition, the condition that this be viewed
as symptomatic oI universal decline. As he treats the scholastici, so does he
treat every other type oI people he meets in his story. They are all inept in

421
Note that the whole poem is structured in the Iigure oI correctio (i.e., non x, sed v), and
switches Irom scazons to hexameters exactly at the sed which introduces the antithesis. For
a general description oI the Iigure, see e.g. Lausberg 1960, 3867.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

215
comparison with the audience, who aIIord the only example oI apparent
virtue and positive moral values in the Satvrica. In Aristotelian terms, En-
colpius and all his characters are comic in the sense that they are worse than
the audience. Their only redeeming Iactor is perhaps that they at times seem
to know that they are comic Iigures.
When Encolpius and Ascyltos realize how shameIul their quarrel is, they
burst out laughing (itaque ex turpissima lite in risum diffusi), which is also
the reaction oI the scholastici when they come out Irom the extemporal dec-
lamation oI the speaker who took over Irom Agamemnon (iuvenes sententias
rident ordinemque totius dictionis infamant). In Trimalchio`s house, laughter
is also a spontaneous reaction to the way the host speaks and conducts him-
selI, although it must be suppressed Ior reasons oI Ilattery.
422
As Trimalchio
is Iirst carried into his dining room to background music and placed on IluIIy
pillows, the sight 'squeezes a laugh Irom the imprudent (32.1, expressit
imprudentibus risum). AIter he gives a speech on the topic oI bowel move-
ments and the importance Ior health oI unrestrained Iarting, even in the din-
ing room, his guests politely thank him Ior his concern Ior them, while 'hid-
ing their laughter in the cups (47.7, castigamus crebris potiunculis risum).
Laughter in the Satvrica is thus oIten the only sign oI moral rectitude in the
characters oI the story. It is a relieving and reassuring sign oI sanity, in an
otherwise mad world, and it can never be completely suppressed. During
moments oI laughter the characters and the audience unite, as it were, in
their understanding oI the moral implications oI the story.
423


422
52.7, excipimus urbanitatem iocantis [sc. Trimalchionis], at ante omnes Agamemnon qui
sciebat quibus meritis revocaretur ad cenam ('We praised the urbanity oI Trimalchio`s
joke, but none more than Agamemnon who knew how to earn another dinner invitation).
423
This repeated 'background laughter may be a generic Ieature oI comic personal recollec-
tions. Since the protagonist is oIten the butt oI the jokes, reports oI spontaneous outbursts oI
laughter among the characters is clearly a good way Ior the narrator to signal to the audi-
ence when something is intended to be Iunny. Perry 1925, 40 n.3, lists instances oI this
same Iigure in the epitome oI the Greek Ass-Story. Conte 1996, 7374, argues in a some-
what similar manner Ior the role oI the reader in revealing the intention oI the 'hidden au-
thor oI the Satvrica. The reader, he says, 'assumes Ior himselI the ironizing attitude oI the
author, until '|t|he reader`s smile . makes explicit the author`s implicit voice, a voice
that would otherwise be bound to silence in a text in which the narrator`s 'I ostensibly
conducts the entire narration. The diIIerence is that whereas I read passages oI laughter in
the text as directive signs to the reader about the satire oI the Satvrica, Conte uses the mod-
ern reader`s laughter as such a directive sign, without allowing Ior the necessarily historical,
cultural and individual nature oI the reader`s laughter. Plaza 2000, 163164, argues that
studying laughter and derision in the Satvrica is important Ior determining the genre oI the
Cena. Indeed, she claims that laughter and derision 'are an essential Ieature oI both satire
and the Iarcical theatre, i.e. mime and comedy, the genres that compete Ior superiority in
3 GENRE

216
Three oI the Ireedmen express their Iears that the scholastici are laugh-
ing at how they speak. Echion knows well that in their eyes his manners are
ridiculous, and so he takes it upon himselI to deIend his point oI view un-
provoked. He pretends to sense a critical attitude in Agamemnon himselI
towards the way he speaks, and he imagines the rhetor as saying: 'What is
that boring man blabbering about? (Quid iste argutat molestus?). He then
quickly provides an answer to this hypothetical criticism: 'Well, it`s because
you, who know how to speak, don`t say anything. You are not like us, and so
you laugh at poor men`s words, but we know that you`ve become silly Irom
literature (46.1, Quia tu, qui potes loquere, non loquis. Non es nostrae fas-
ciae, et ideo pauperorum uerba derides. Scimus te prae litteras fatuum esse).
Niceros, likewise, when asked by the host to tell a story, is aIraid oI the
scholastici, who he thinks will laugh at his words (61.4, timeo istos scholas-
ticos ne me rideant). As we have seen above, the narrator, to accentuate the
shortcomings oI such an incompetent story-teller, introduces his badly told
ghost story in pompous epic language, haec ubi dicta dedit . exorsus est
(61.5). The same phrase is also used in Eumolpus` exaggeratedly epic epic
poem (121.l).
AIter the pittaciajokes are read aloud to the guests, they all laugh Ior a
while (56.10, diu risimus), though Ior diIIerent reasons. Ascyltos, who has
not mastered Agamemnon`s art oI Ilattery, cannot hold back any longer and
throws up his hands in a gesture oI general dismissal and laughs until his
tears start Ilowing (57.1, ceterum Ascvltos, intemperantis licentiae, cum om-
nia sublatis manibus eluderet et usque ad lacrimas rideret). At this Her-
meros is roused to his host`s deIense, and tries to restore order by suppress-
ing this unwanted laughter. He argues that Ascyltos is alone in Iinding Tri-
malchio Iunny, and that Agamemnon, his senior as scholasticus, does not
think the Ireedmen are ridiculous (57.8, Tibi soli ridiclei uidemur, ecce mag-
ister tuus, homo maior natus. placemus illi). He naturally does not delve into
the reasons behind the rhetor`s acceptance oI their manners. At this on-
slaught Giton, likewise, 'indecently lets out a long suppressed laugh (58.1,
post hoc dictum, Giton . risum iam diu compressum etiam indecenter ef-
fudit). At this Hermeros, who assumes Giton is a slave, turns his attentions
towards him and prides himselI Ior not knowing the nonsense oI liberal edu-
cation (Non didici geometrias, critica et alogas naenias). He Iurthermore
insists that Giton`s master has wasted his money on the boy`s rhetorical edu-

this episode. With regard to genre, Plaza clearly views the Satvrica as 'synthetic, and in
other episodes oI the work laughter suggests to her the genres oI comedy and erotic poetry.
See below section 3.2.4 on the origin oI this approach to the problem oI genre in the Sa-
tvrica.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

217
cation (58.78, Iam scies patrem tuum mercedes perdidisse, quamuis et
rhetoricam scis), Ior it is his classical education which has bred such arro-
gance in a common slave. Forming a contrast to Giton`s useless education is
Hermeros` own simple and practical instruction and the trade that he learned,
the basis oI his Iinancial prosperity.
424

In this manner the dinner party at Trimalchio`s can be seen as one long
match between declaimers who practice Iormal oratory as iI they were pos-
sessed by alien Iuriesand who have thereIore lost all sense oI the reality oI
Roman institutions and liIe as it should be according to aristocratic values
and the Ireedmen who have not at all learnt to speak well, and Ior whom the
only knowledge worth acquiring is their lowly trade. In practice, the Ireed-
men may be said to win the match, despite being constantly subjected to
ridicule, because the scholastici are not even allowed to open their mouths in
their own deIense, and because they are parasites at the Ireedmen`s table.
What their deIense would have been, however, is no mystery because Aga-
memnon has ultimately blamed the madness oI declamation on his students
and their parents, by deIining them as lunatics and claiming that learned
teachers were simply Iorced to play along with their madness (3.2, nil mirum
in his exercitationibus doctores peccant, qui necesse habent cum insanien-
tibus furere |'No wonder teachers are at Iault by employing these exercises,
Ior they have to play at being insane to please the madmen|). The parents oI
Agamemnon`s pupils seem to be oI the same social class and have the same
values as Trimalchio and his guests.
From the standpoint oI Encolpius` audience, the comic value oI this
encounter between two social types lies in the Iact that the two groups have
undertaken a mutual deception, which exposes both as deprived and hypo-
critical. While the scholastici attract their young students and earn invita-
tions to dinner parties with honey balls oI words all spiced up with poppy
and sesame seeds (mellitos verborum globulos et omnia dicta factaque quasi
papavere et sesamo sparsa), Trimalchio, in Encolpius` language, uses simi-
larly spiced up dormice (glires melle ac papavere sparsos) to attract the
scholastici to his dinner table. The scholastici trade compliments Ior Iood,
whereas the Ireedmen trade Iood Ior compliments (35.1, laudationem insecu-

424
As is evidenced by the concluding part oI Hermeros` speech, 58.14, 'Ego, quod me sic
uides, propter artificium meum diis gratias ago ('I thank the gods Ior giving me the trade
which made me what I am`); and the speech oI Echion, 46.8, artificium numquam moritur
('a trade never dies`); cI. as well Trimalchio`s words, 56.1, 'Quod autem, inquit, putamus
secundum litteras difficillimum esse artificium? Ego puto medicum et nummularium
('What trade`, he said, do we think most diIIicult aIter that oI letters? I think doctor, or
cashier`).
3 GENRE

218
tum est ferculum). The whole encounter is like that oI two groups oI human
satyrs led by their Sileni, Agamemnon and Trimalchio. It is in contrast with
a noble audience that is beyond criticism, in which both these groups are
Iound inIerior and ridiculous.

VersiIication is another type oI discursive madness, to be contrasted with the
'normal manner oI speech practiced by Encolpius in addressing himselI to
his Iine audience. As we have seen, the Satvrica`s comic discrimination
based on manner oI speech is an essential Ieature oI the work. Rhetoricians
and poets alike are ridiculed Ior their inability to speak human language
(read: the urbane Latin idiom oI well-bred Romans). The underlying conceit
is to recognize no other category oI speech than the conversational language
oI the aristocratic audience, and measure all statements by that standard. On
this criterion young Encolpius is saner than such linguistic madmen as Aga-
memnon and Eumolpus, but much inIerior to the narrator who has acquired
an idiom which is almost that oI his audience. As we saw above (in section
1.2.6) the playIully pragmatic discourse analysis at work in the Satvrica was
originally a part oI Cynic literature. But primitivist attacks like this one on
education and 'science must have had an appeal Iar beyond the ranks oI
Cynic 'philosophers in a society where the noble Iamilies and their emula-
tors Ielt that their traditional monopoly on high culture was threatened.
No sooner is the narrator done with narrating the conIrontation oI the
scholastics and the uneducated tradesmen than he introduces the poet Eu-
molpus. The poet begins by claiming not to be venal (like the scholastici)
and adduces as prooI oI his artistic integrity the Iact that rich men (read: rich
upstarts like Trimalchio) do not like his poetry. But no one likes his poetry,
except perhaps Bargates who speaks with rabiosa barbaraque voce (96.5)
and needs the poet to compose invective against his mistress. Eumolpus is a
compulsive versiIier who with his extemporization in the pinacotheca on the
capture oI Troy provokes ordinary people walking in the temple portico to
drive him away with a shower oI stones as a cursed madman. He himselI
takes this response to his poetry as an inverse compliment, but young Encol-
pius Iears to be taken Ior a poet as well (90.2, timui ego ne me poetam vo-
caret) while he is in the other`s company. SaIely out oI reach, the narrator
reports that he asked the poet what he thought he was up to with this disease
(90.3, Quid tibi vis cum isto morbo?). During the less than two hours that
they had spent together, he says, Eumolpus had more oIten spoken like a
poet than like a man (90.3, minus quam duabus horis mecum moraris, et
saepius poetice quam humane locutus es). Even though the poet promises to
abstain Irom this 'Iood Ior the whole day (90.6, toto die me ab hoc cibo
3. 1 ANCI ENT NARRATI VE I N PERSONI S

219
abstinebo), Encolpius leaves Eumolpus some moments later reciting again in
the bathhouse (91.3, relicto Eumolpo, nam in balneo carmen recitabat).
There he gets his usual hostile reception (92.6, paene vapulavi, quia conatus
sum circa solium sedentibus carmen recitare, et postquam de balneo tam-
quam de theatro eiectus sum . |'I was almost Ilogged, because I tried to
recite a poem to those seated around the bath-tub, and aIter I was thrown out
oI the baths as I was thrown out oI the theater|).
Perhaps the most comic description oI Eumolpus` peculiar madness
comes at the end oI the episode oI the voyage (115.15), where the poet, in a
moment oI inspiration, is completely oblivious to the liIe-threatening storm
and shipwreck they are suIIering; a scene which could be read more gener-
ally as the 'shipwreck oI poetry, adapting the classic 'ship oI poetry topos
to Iit the context.
425
Encolpius describes how the boys heard strange sounds
coming Irom the captain`s cabin, as iI some beast were trying to break out oI
its cage (quasi cupientis exire beluae gemitum). They Iollowed the noise and
Iound Eumolpus in the midst oI covering a great parchment with written
verses. Amazed that he should have leisure to write poetry in such proximity
with death, the boys dragged him out shouting and told him to restrain him-
selI (mirati ergo quod illi vacaret in vicinia mortis poema facere, extrahimus
clamantem iubemusque bonam habere mentem). The poet was merely an-
gered and didn`t want to be disturbed, begging to be allowed to Iinish his
sententia, because, as he said, the poem was struggling towards its end ('si-
nite me` inquit 'sententiam explere, laborat carmen in fine`). Eventually,
Encolpius tells how he asked Giton Ior help and took the 'Irenetic and
'mumbling poet by the hand and pulled him onshore (inicio ego phrenetico
manum . et in terram trahere poetam mugientem). The day aIter, Eumolpus
with absolute consistency oI character was again composing poetry, rolling
his eyes seeking to pick up signals Irom aIar (115.20, oculos ad arcessendos
sensus longius mittit), this time Ior an epigram in memory oI the drowned
ship-owner Lichas.
On the way to Croton, Eumolpus takes the opportunity to recite his un-
Iinished (nondum recepit ultimam manum) poem, which he introduces in a
mock critical preIace, to his Iellow travelers, another captive audience. He
starts by claiming that poetry is more than just versiIying and using poetic
diction, and then proceeds to distinguish himselI Irom another equally undis-
tinguished group, the declaimers, who he says are mistaken iI they think that
poetry is easier than composing controversies painted in vibrating little sen-
tentiae (118.2, controversiam sententiolis vibrantibus pictam). Poetry is

425
Connors 1994, 233.
3 GENRE

220
diIIerent Irom prose history, or the oratory oI the courts with testimonies
Irom witnesses, in that it does not have to establish what really happened
(118.6, non enim res gestae versibus comprehendendae sunt, quod longe
melius historici faciunt . potius . vaticinatio appareat quam religiosae
orationis sub testibus fides). Poetry requires a headlong plunge oI the Iree
spirit (praecipitandus est liber spiritus) into circumlocutions and divine
agency with Iabulously twisted expression oI sententious opinions (per am-
bages deorumque ministeria et fabulosum sententiarum tormentum). The
desired result will be like prophetic madness (furentis animi vaticinatio),
although, according to the poet, the stuII oI the civil war is crushingly heavy
(quisquis attigerit . sub onere labetur), and only Ior someone replete with
letters (plenus litteris) to attempt. Laid down Ior imitation are such estab-
lished institutions as Homer and the Greek lyric poets, but Virgil is included
as well, and Horace, the only lyric Roman worth mentioning. The master
poet must Ilee Irom all vulgarity oI language and choose words removed
Irom the uninitiated common man and he must adopt the Iamous opening
words oI Horace`s third book oI odes as his motto (refugiendum est ab omni
verborum, ut ita dicam, vilitate et sumendae voces a plebe semotae, ut fiat
'odi profanum vulgus et arceo` |'one must Ilee Irom every vileness, as it
were, oI vocabulary, and choose words remote Irom the plebs, taking as
one`s motto: I hate the uninitiated crowd and stay away Irom it`|). Finally,
the rhetorical sententiae should not be obvious and stand out Irom the body
oI the discourse, but should be woven into it and shine with the color oI the
poetic garment (praeterea curandum est ne sententiae emineant extra corpus
orationis expressae, sed intexto vestibus colore niteant).
It is clear that this contradictory programme is not to be taken seriously.
Its Iunction is to be a Iurther sketch oI Eumolpus` poetic madness. The pro-
logue is hortatory in tone and yet it is completely deprived oI authority, com-
ing Irom such a character. Like so many grand statements in the Satvrica this
one Ialls Ilat on its Iace. The style oI the rather long poem oI Eumolpus,
which Iollows, is an obvious, although not overly exaggerated, parody oI
epic conventions.
426
This programmatic statement serves exactly the same
Iunction as the poem oI Agamemnon (5.1), in that it comically makes the
poet preach against vices which are his own in a language that is ridiculous
and absurd. When the verbose and Iantastic poem Iinally ends with the arri-

426
Eumolpus` verse and its postulated targets in the larger context oI Roman literature is an
immense topic which Ialls outside the scope oI this study. We are only concerned with read-
ing the poem in its immediate context and with respect to the personae oI the narrator En-
colpius and his character, the poet Eumolpus. For a recent study dedicated to the poetry oI
the Satvrica, see Connors 1998.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

221
val in Croton (cum haec Eumolpos ingenti volubilitate verborum effudisset,
tandem Crotona intravimus |'Once Eumolpus had poured out these verses in
enormous verbal spin, we Iinally entered Croton|), the poet`s equally ver-
bose and Iantastic conIidence trick is described in language similar to his
poetry (124.3, ex praescripto ergo consilii communis exaggerata verborum
volubilitate, unde aut qui essemus, haud dubie credentibus indicavimus |'In
accordance with the script we had agreed upon together in preIabricated
verbal spin, we told these certainly gullible Iellows where we came Irom,
that is to say who we were|).
3.1.7 Infra pecuniam
We now come to the second dominant moral theme in the Satvrica, namely
how the supposedly non-aristocratic interest in moneymaking has replaced
all other values with the value oI currency. It is here that we shall Iind
accumulative evidence Ior Encolpius` consistent claim that the world suIIers
Irom an over-appreciation oI the value oI money. This theme especially al-
lows Ior a reading oI the Satvrica as a whole as an example oI Saturnalian
literature, a literature that aims at portraying a 'verkehrte Welt.
427
On a
purely syntactical level this tone oI the work is maniIested in the ubiquitous
Iigure oI correctio ('it was not what you would expect, but something en-
tirely diIIerent or 'you should not Ilatter the rich as everybody does, but
study hard and be virtuous), which contributes to the sense oI scandal and
impracticable moralism oI the characters. In addition to the widespread lin-
guistic aberrations analyzed above, everything in the world oI the Satvrica
sooner or later Iinds itselI 'subservient to money, infra pecuniam (84.3), as
Eumolpus puts it.!
Retracing our steps we recall that in the opening passages oI the Satvrica
Encolpius blamed Agamemnon and other teachers oI rhetoric Ior having
destroyed eloquence (primi omnium eloquentiam perdidistis) by teaching
young boys contemporary declamation. Agamemnon`s answer is to deIend
his proIession and explain what he ironically calls the secret art (ars secreta)
oI the rhetorical schools: As the hypocritical Ilatterer who seeks invitations
to the dinner-parties oI the wealthy, the teacher oI eloquence (eloquentiae
magister) must think Iirst oI all oI that which is most pleasing to his stu-
dents. OI course (nimirum), says Agamemnon, the teachers act incorrectly
(peccant) when they make youths practice declamation ostensibly to im-

427
For the 'verkehrte Welt theme, especially in relation to the Cena and the Croton episode,
see Dpp 1993 144177.
3 GENRE

222
prove their eloquence, but really to entertain them and attract them to the
schools. But it is really the parents who are to blame (parentes obiurgatione
digni sunt) who sacriIice their own children, like everything else, to selI-
interest (spes quoque suas ambitioni donant), and hurry them into the courts
beIore their education is Iinished. Despite the Iact that the parents them-
selves proIess to believe that eloquence is a good and noble thing, it is still
their ambitio, which we may translate in this context as 'greed or 'corrup-
tion, that gets the better oI them.
This topic is Iurther illustrated in a later episode (46), which we have
previously looked at Irom another angle. At the dinner-party oI Trimalchio,
Echion the Iireman (centonarius) addresses Agamemnon in a long speech,
which reads as an illustration oI Agamemnon`s explanation oI the decline oI
rhetoric. The son oI Echion, who is a promising student and has the right
ingenium to become an eloquent lover oI classical letters, will not be allowed
to waste his time on such unproIitable and 'polluting pursuits. His Iather,
who is a deliberately insensitive type (non debemus delicati esse), kills his
goldIinches and disapproves oI the boy`s painting (we recall that painting is
another special interest oI Encolpius alongside most Iorms oI literature),
because, according to the Ireedman, these things make the boy idle and un-
prepared Ior liIe`s real goal, moneymaking. Only law counts as a worthy
subject in Echion`s mind, Ior it oIIers the promise oI pecuniary proIit.
The values expressed by Echion Iorm a striking contrast, Ior example,
with Horace`s account oI his education in the sixth satire oI Book I. The
reader will recall that the poet`s Ireedman Iather, a man oI exactly the same
rank as Echion, despite his inIerior social status, wouldn`t send his son to the
local school, run by a nobody called Flavius, where, as Horace puts it, the
sons oI 'mighty centurions went. Instead he took young Horace to Rome
where he would get the best contemporary education with the sons oI
knights and senators. We note the implicit acknowledgment oI a two-tiered
educational system. A good Roman education Ior boys was usually not to be
had outside the capital.
428
The thought oI Echion`s Primigenius as another
potential Horace, destroyed by his ignorant Iather, may be distressing to
some, but Echion has a diIIerent view oI things. What he sees are the local
scholastics, including Agamemnon, who have been reduced to the role oI
parasites, and he doesn`t see this as a desirable Iuture Ior his son.
Any knowledge oI or even interest in the aristocratic schools oI Rome is
completely beyond this simple man. In the world portrayed by Encolpius, the

428
Except when private tutors were used in upper-class households. Why else does Quintilian
make such a lengthy pitch Ior sending youths to school (Inst. Or. 1.2.931)? On the educa-
tion oI Roman boys in general, see Bonner 1977 and Kaster 1988.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

223
Iinal analysis must be that neither Echion`s cruel pragmatism, nor Agamem-
non`s erratic educational programme, can oIIer the slightest hope to young
Primigenius. He is simply a member oI the lower classes, and as such will
not receive a good education, no matter how talented. The same moral can
be read Irom Encolpius` account oI the children oI Philomela, the matron oI
Croton, who prostitutes them to Iurther her own legacy-hunting, while pro-
Iessing to leave them in the care oI Eumolpus Ior the sake oI their education
(140). In Encolpius` report Irom the underworld, there is not a glimmer oI
hope Ior such children. Those who are not born to wealth are utterly ridicu-
lous in their inIeriority and their pathetic striving.
Another incident illustrates how money corrupts justice (1215). Encol-
pius and Ascyltos go to the market in the twilight to try to sell the stolen
pallium. Through Fortune`s wondrous play (o lusum Fortunae mirabilem'),
the prospective buyers oI their stolen cloak are also the possessors oI a small
tunic, tunicula, Iull oI gold coins, which the duo had stolen earlier and then
lost. Young Encolpius, true to his training in declamation, is ready to argue
the case Iormally on the elementary statute that a person reIusing to return
the belonging oI another can be Iorced to do so with the injunction oI a law
court (negavi circuitu agendum, sed plane iure civili dimicandum ut si nollet
alienam rem domino reddere, ad interdictum veniret). But here the narrator
Encolpius suddenly interrupts his narrative to introduce his own recogniza-
bly Cynic point oI view (notwithstanding his attempt in line three to distance
himselI Irom the charge oI preaching a Cynic dogma):

quid Iaciant leges, ubi sola pecunia regnat,
aut ubi paupertas vincere nulla potest?
ipsi qui Cynica traducunt tempora pera
non nunquam nummis vendere verba solent.
ergo iudicium nihil est nisi publica merces,
atque eques in causa qui sedet, empta probat. (Sat. 14.2)

What can laws accomplish, where money alone rules, where poverty
cannot win a case? The selI same men, who go through liIe with a
Cynic`s purse, are not unaccustomed to selling their testimony. Accord-
ingly a lawsuit is nothing but a public auction, and the knight who sits in
the jury delivers a verdict that has been bought.

These lines about the ineIIicacy oI the law in a society where corruption is a
matter oI routine are a rare direct statement Irom Encolpius in another dis-
course type which marks them oII Irom the rest oI the narrative (see my dis-
3 GENRE

224
cussion in section 1.1.1 about the groundless removal oI these verses Irom
where they belong in the text). This moral 'message is dressed up as poetry
to deliberately undercut the seriousness oI the socio-cultural critique. When
Encolpius dons a versiIier`s mask, as he does here, it is with the intention oI
drawing a little bit oI the sting and immediacy oI the explicit social criticism
by transmuting it Irom the real and relevant world to an impotent and tritely
poetic oneentertainment displacing or veiling social commentary. To un-
derstand the subversive eIIect oI this sudden switch Irom conversational
prose to poetic recital one should always keep in mind the aspect oI per-
Iormance. Poetic rhythm and diction, which Encolpius treats, according to
the convention oI prosimetry, as a speech aberration, is a specialized 'num-
ber Irom his bag oI tricks; the audience never knows whom they`re going to
be seeing or hearing next, so they surrender to the entertainer`s charm and
inventiveness, allowing the social message, which is scattered throughout the
narrative and hidden in the talk oI clowns and social inIeriors, to build up by
indirection.
In the past oI the story Encolpius was Iull oI simplistic optimism, but in
the present narrative situation his older selI seems Iully cognizant oI the
degenerate state oI things, although he usually does not allow his 'serious
Iace to obtrude in the prose narrative. The narrator`s selI-deprecating strate-
gies not only protect him Irom his audience, but also protect the audience
Irom the Iorce oI the satire by subverting it. Indeed, he protects himselI pre-
cisely by protecting his audience. The satire needn`t be taken seriously, be-
cause it`s not advanced in a serious way by people (the narrator and his
troupe oI personae) whom the audience has to take seriously. However, the
building up oI the message and the consistency oI the narrator`s ideas about
the power and rule oI money is unIailing and gradually takes on the Iunction
oI a reliable truth about the world oI the Satvrica.
Through the otherwise transitory events narrated, the characters behave
in perIect harmony with this general principle. In the market scene, both
parties to the above quarrel at last want to settle with a simple exchange oI
goods, because both think that this will be most proIitable. But the Iorces oI
law and order, 'advocates and yet little more than thieves (advocati tamen
iam paene nocturni), who want to make a proIit out oI the cloak, insist that
the disputed property be deposited with them, hoping that out oI Iear neither
party will show up in the morning in Iront oI the judge. Their argument is
that the case is not a matter oI simple dispute between two parties oI the type
common in textbook cases oI rhetorical controversia (neque enim res tantum
quae viderentur in controversiam esse), since each accuses the other oI steal-
3. 1 ANCI ENT NARRATI VE I N PERSONI S

225
ing (in utraque parte scilicet latrocinii suspicio haberetur) and neither party
even pretends to be innocent.
None oI the characters truly appreciates declamatory exercises Ior the
simple reason that they are useless in the world oI the Satvrica, where quar-
rels are not resolved in highly Iormalized disputes but through cash pay-
ments. Trimalchio, the uneducated but wealthy host oI the scholars, knows
this only too well, but in order to lend credence to his posturing as a well-
bred lover oI literary studies, he asks Agamemnon to perIorm an exposition
oI the controversia he declaimed that day. Agamemnon starts with a stan-
dard cliche: 'A poor and a rich man were enemies (48, pauper et dives in-
imici erant). But he gets no Iurther Ior the moment because the other inter-
rupts with his clever remark: 'What is a poor man? Ior which he receives
the obligatory praise Irom Agamemnon. The implication seems to be that
Trimalchio, who once was a slave and not even in possession oI his own
body, is now so rich that he can aIIord to ignore the existence oI poverty.
The narrator then continues with the completely indiIIerent phrase, 'and
Agamemnon introduced some controversia or other (nescio quam contro-
versiam exposuit), not even bothering to report the argument oI Agamem-
non`s controversia. AIter the exposition oI the case, it takes Trimalchio two
short sentences to clear up the problem deIinitively: 'II this happened, then
there`s nothing to argue about. II it didn`t happen, it is nothing (hoc . si
factum est, controversia non est, si factum non est, nihil est). II we inquire
into why, according to Trimalchio, a real case between a poor man and a rich
man is not a controversy, the obvious answer is that since the rich man is
able to bribe the jury, he will win in any case, and so there is nothing to ar-
gue about! However, iI it is an imaginary case, it`s meaningless nonsense
that no one but a declaimer would entertain. So much Ior the interest oI Tri-
malchio in the scholastic subtleties oI controversiae.
429

The same topic also comes into play when the narrator introduces Eu-
molpus, the poet, into the pinacotheca and thus into his story: 'Behold! .
an old man grown white entered the gallery, a person with a tortured Iace
who seemed to promise something or other great, but not accordingly well-

429
Ever since Heinsius various commentators and translators have interpreted the words oI
Trimalchio: 'hoc, si factum est, controversia non est`, as meaning that since a rhetorical
controversia may be deIined as a Iictitious case, thereIore a real event cannot be a contro-
versia. This implies both knowledge and genuine interest on the part oI Trimalchio in the
Iormal terminology oI the rhetorical schools. But such learned scholasticism in Trimalchio
seems wholly out oI character. Besides, iI Trimalchio thinks that controversiae must be Iic-
tional cases, what does he then mean when he adds, 'si factum non est, nihil est`? Surely, iI
Heinsius was right, he should have added, 'si factum non est, controversia est`.
3 GENRE

226
groomed or smartly dressed, so that he evidently belonged to that brand oI
literati which is despised by the rich (ut facile appareret eum hac nota lit-
teratorum esse quos odisse divites solent). As iI this were not enough Eu-
molpus introduces himselI with these words: 'I`m a poet and as I hope not oI
the humblest spirit, iI laurels, which Iavoritism indeed also awards to the
unworthy, are to be relied upon. You ask, then, why I`m so badly dressed.
Well, it`s because commitment to genius never made anyone rich (amor
ingenii neminem unquam divitem fecit). Then he delivers six hexameters
about how merchants, soldiers, Ilatterers and adulterers all proIit Irom their
activity when eloquence alone shivers in Irosty rags and calls with 'a penni-
less tongue, inops lingua (83), upon the Iorsaken arts. He then switches
back to prose to elaborate Iurther:

'Non dubie ita est: si quis vitiorum omnium inimicus rectum iter vitae
coepit insistere, primum propter morum diIIerentiam odium habet; quis
enim potest probare diversa? deinde qui solas extruere divitias curant,
nihil volunt inter homines melius credi quam quod ipsi tenent. insectan-
tur itaque, quacumque ratione possunt, litterarum amatores, ut videantur
illi quoque inIra pecuniam positi. (Sat. 84.23)

'There is no doubt that this is the way it is. II a man sets himselI against
every vice and starts oII on the straight and narrow, he`s immediately
hated because oI his diIIerent ways. No one can approve oI conduct diI-
Ierent Irom his own. And secondly, those who are interested in piling up
money don`t want anything else in liIe regarded as better than what they
hold themselves. So lovers oI literature are persecuted by every means
possible so that they too will seem subservient to money.

It might Ior a moment seem Irom these words that a virtuous individual was
speaking, but this is certainly not the case with Eumolpus. The man who like
a Cynic philosopher calls poverty the sister oI bona mens turns out to be an
expert conIidence man, who is willing to cook up the most elaborate decep-
tions Ior a proIit. Like so many characters in the Satvrica he proIesses to
know what is right, but claims that he cannot practice his virtues because oI
rich men and their corrupt ways. Despite his boastIul claims oI Cynic virtue,
he too is 'subservient to money, infra pecuniam positus (84.2).
Encolpius tells his audience that while they were still standing in the
pinacotheca he asked the poet about the reason behind the slothIul state oI
contemporary arts and especially painting. This discourse is reminiscent oI
the one initiated by Encolpius when he met Agamemnon. The state oI con-
3. 1 ANCI ENT NARRATI VE I N PERSONI S

227
temporary literature and visual arts is a constant preoccupation oI his. Eu-
molpus` answer is prompt: 'the greed Ior money has caused this upheaval,
Ior in the olden times, when naked virtue still had its appeal, the liberal arts
were strong and there was a mighty competition among men not to let any-
thing oI beneIit to the ages lie hidden Ior long (88, pecuniae . cupiditas
haec tropica instituit. priscis enim temporibus, cum adhuc nuda virtus
placeret, vigebant artes ingenuae summumque certamen inter homines erat,
ne quid profuturum saeculis diu lateret).
430
!
Religion has gone the way oI the arts. Eumolpus in the above speech in
the pinacotheca laments that people have stopped praying in temples Ior
such things as eloquence and philosophical wisdom. And they don`t even
pray Ior such obvious blessings as 'good sense, mens bona, and good
health. Instead, they have hardly reached the Capitol when they start promis-
ing giIts to the god Ior arranging the death oI a rich relative, Ior letting them
Iind a treasure, or granting them a certain level oI wealth Iree oI risk. Even
the senate, says Eumolpus, which should be a model oI what is good and
right, regularly promises quantities oI gold to the supreme god oI the state
cult, Capitoline Jove. Thus even this most respectable gathering oI Roman
citizens legitimates the greed oI every one by assuming that even the Iather
oI the gods is 'subservient to money, infra pecuniam positus (84.2). It is no
wonder, then, that the appreciation oI artistic beauty has decreased when to
gods and men a mass oI gold seems prettier than the works oI Apelles and
Phidiasthose 'crazy little Greeks (Graeculi delirantes), he adds as iI to
draw a bit oI the sting oI his criticism oI the Roman senate.
The edge oI this criticism is, oI course, blunted by the absurdity oI the
speaker, a crazy poetas it must be. Encolpius cannot aIIord to preach sena-
torial corruption unIiltered to his Roman elite audience. The cleverness oI
the technique is to let the narrator develop a number oI blurred or limited
perspectives through his character personae and his own central narrator`s
persona on the same social problem. Because none oI these personae has the
authority to address the elite audience as a social and moral equal, their diI-
Ierent voices cannot give oIIense. Collectively, though, their indictment is
both accurate and damning. Again, iI we consider the perIormance aspect,

430
Walsh 1970, 96I., tells his reader that Eumolpus is dead wrong about the prisca tempora.
From a strictly historical point oI view this may be correct, but the narrator believes it and
his Roman audience seems to believe it as well. We cannot thereIore use this supposed his-
torical inaccuracy oI Encolpius` satire as an argument to trivialize its message. Just because
we are skeptical oI the glory oI the olden days doesn`t mean that Encolpius (or Petronius, iI
you will) didn`t believe in it. Let us remember that ancient satire and the modern discipline
oI scientiIic history are worlds apart.
3 GENRE

228
the stance is very eIIective. Instead oI causing anxiety and paranoia among
the ruling elite, as did, according to Tacitus (Dial. 2.1), Curiatius Maternus`
passionate impersonation, while reciting his historical tragedy, oI the stoic
'revolutionary Cato, Encolpius can say worse things indirectly through his
mask oI a silly poet, so long as he does it in an entertaining Iashion.
Likewise in the speech oI the insigniIicant Ireedman Ganymedes (44),
we notice the same preoccupation with the corruption oI religion by money.
Ganymedes is concerned about the misery caused by corruption in his home-
town, the urbs Graeca. There is no bread to be had, he complains, because
the aedile oI the market makes dirty deals with the bakers to Iix the market
price. This wasn`t so, he says, when he Iirst came there Irom Asia as a boy.
Then the magistrates were lions and punished those who imported bad corn
Irom Sicily. Now, however, the aedile is keener to make a buck Ior himselI
than to preserve the lives oI the townspeople. But then, unexpectedly per-
haps, he takes a leap oI Iaith and conjectures that all the misery must be
caused by angry gods. His Iatalism, oI course, undercuts the political mes-
sage without, however, retracting what has been said. No one, says Gany-
medes, regards heaven Ior what it is, no one Iasts and Jove is not worth a
single hair to people, instead with eyes closed they all count their property
(Nemo enim caelum caelum putat, nemo ieiunium servat, nemo Iovem pili
facit, sed omnes opertis oculis bona sua computant). He goes on to describe
a memorable picture oI how the matrons oI old in their best clothes used to
climb the hill bareIoot, hair loose and mind pure, and pray to Jove Ior rain.
In those days, oI course, it started raining by the bucket and they returned
home 'like drowned rats (udi tamquam mures). This nostalgia Ior the good
old times, beIore men grew obsessed with money, works to accentuate the
sense oI despair in the present.
The topic continues with obsessive persistency. Encolpius says that in
Croton he sought a cure Ior his impotence Irom the witch doctor Oenothea.
While she is away renewing the Iire she had unwittingly put out, he Iights a
mock-epic battle with a Ilock oI geese, commemorated in a virtual epic sim-
ile oI at least Iive hexameter lines. He eventually kills their 'leader and
teacher oI cruelty (136, dux et magister saevitiae). When the priestess re-
turns, she inIorms him through shrieks and curses that he has committed a
hideous sacrilege by killing a goose that was sacred to Priapus. However, as
soon as Encolpius oIIers to expiate Ior the crime by paying two gold pieces,
with which they, as he puts it, 'could buy both gods and geese (137, unde
possitis et deos et anseres emere), both Oenothea and her Iriend Proselenos
are immediately calmed and become more than willing to cover up the sacri-
3. 1 ANCI ENT NARRATI VE I N PERSONI S

229
lege. And now, in a typical Iashion, Encolpius delivers Iive elegiac distichs
about the maniIest omnipotence oI money:

quisquis habet nummos, secura naviget aura
Iortunamque suo temperet arbitrio.
uxorem ducat Danaen ipsumque licebit
Acrisium iubeat credere quod Danaen.
carmina componat, declamet, concrepet omnes,
et peragat causas sitque Catone prior.
iurisconsultus 'parret, non parret habeto
atque esto quicquid Servius et Labeo.
multa loquor: quod vis, nummis praesentibus opta,
et veniet. clausum possidet arca Iovem. (Sat. 137.9)

Whoever has money can sail in saIe wind and dilute his Iortune in a pri-
vate mixing bowl. Let him take Danae to wiIe, and tell Acrisius himselI
to believe what he told Danae. Let him write poetry, make speeches,
command the world by snapping his Iingers, win his court-cases and
outdo Cato in moral authority. As a legal expert, let him have his
'Proven or 'Not proven, and be all that Servius and Antistius Labeo
were. In short, whatever you want, with ready cash, make a wish and it
will come true. Your moneybox has Jupiter shut up inside.

The pessimistic argument, 'money is omnipotence, is here considerably
expanded to cover a vast sphere oI inIluence. The rich man can sail in saIe
wind, and does not have to suIIer shipwreck in liIe. He may dilute his Ior-
tune suo arbitrio ('at his own discretion), i.e. he is the arbiter bibendi in
liIe`s drinking party. He can weaken the eIIects oI bad Iortune, strengthen
those oI good Iortune. Like Jupiter himselI he shall have Danae by shower-
ing her with gold, and her Iather will have to swallow his pride and accept
whatever (mythological) pretext given. The rich man can Ireely compose
poetry, too, whatever the extent oI his talent; he can even declaim to the
guaranteed applause oI everyone present. These general statements about the
power oI money are interesting in themselves, but when compared with the
narrated behavior oI Trimalchio and his guests, Ior example, they take on a
special importance Ior the overall design oI the verkehrte Welt oI the Sa-
tvrica. We recall Trimalchio`s distorted compositions (34; 41; 55) and his
astrological philologia (39), all oI which was met with applause bought Irom
his educated audience. On the same principle, the rich man can also play
every instrument, win cases in court and be considered morally superior to
3 GENRE

230
Cato himselI. Should he choose to practice law, money will guarantee the
persuasiveness oI his arguments no less than eloquence once guaranteed the
success oI Servius and Antistius Labeo. In short, everything the rich man
wants is permitted him, because Jove is at his service, locked up in his mon-
eybox.
There is an obvious similarity in the way the narrator introduces these
lines here and the three elegiac distichs on corruption (14, quid faciant leges,
ubi sola pecunia regnat |'What can laws accomplish where money alone
rules|). These poemsiI that is the proper termconstitute the closest
thing to a committed statement Irom the narrator that we shall ever Iind in
the Satvrica. The change oI discourse type, Irom the urbane colloquialism
appropriate to addressing the distinguished audience to the more involved
discourse type oI elegy, is a clear indication that Encolpius does not want to
voice such criticism without undercutting the message with poetic 'mad-
ness. The ubiquity oI this message in the preserved Iragments means that it
must also have been an important aspect oI many a lost episode. But this
does not mean that the satire ever became overwhelming Ior the audience. It
is precisely to avoid being taken too seriously that Encolpius lapses into
verse. It is impossible, however, to argue that Encolpius` obstinate criticism
oI ancient capitalism (money equals greatness) is meant to be merely Iunny,
because the events oI the story itselI conIirm in details the accuracy oI the
idea. At times the plot oI the Satvrica is in such perIect agreement with En-
colpius` satirical theory oI contemporary society that one could perhaps be
justiIied in speaking oI an illustration oI theoretical principles.
Technology is next. Unsurprisingly, we detect the same pattern once
more. The reader will recall the statement oI Eumolpus (88) to the eIIect that
in the 'olden times there was a great competition between men not to let
anything oI beneIit to the ages lie hidden Ior long, but that among his con-
temporaries the greed Ior money had made an end oI this unselIish scientiIic
spirit. Trimalchio himselI in an 'urbane outburst is made to illustrate the
principle with an outrageous anecdote, the Iamous story oI the man who
invented unbreakable glass and was promptly rewarded by the princeps:

'Fuit tamen Iaber qui Iecit phialam vitream, quae non Irangebatur. Ad-
missus ergo Caesarem est cum suo munere, deinde Iecit reporrigere Cae-
sarem et illam in pavimentum proiecit. Caesar non pote valdius quam
expavit. At ille sustulit phialam de terra; collisa erat tanquam vasum ae-
neum; deinde martiolum de sinu protulit et phialam otio belle correxit.
Hoc Iacto putabat se coleum Iovis tenere, utique postquam illi dixit:
Numquid alius scit hanc condituram vitreorum?` vide modo. Postquam
3. 1 ANCI ENT NARRATI VE I N PERSONI S

231
negavit, iussit illum Caesar decollari: quia enim, si scitum esset, aurum
pro luto haberemus. (Sat. 51)

'Mind you, there was a craItsman once who created a glass bowl that
didn`t break. So he got an audience with the Emperor and gave it to him
as a present. Then he made Caesar hand it back to him and threw it on
the Iloor. The emperor was visibly shaken. The Iellow picked the bowl
oII the groundit had been dented like a bronze dishpulled out a
hammer Irom the Iold oI his dress and preceded to Iix the bowl and make
it as good as new. AIter this perIormance he thought he held Jove by the
balls, especially aIter the emperor asked him: Is there anyone else who
knows this process Ior making glass?`But see what happened!When
the man answered No!`, the emperor had his head cut oII, the reason be-
ing that iI it was made public, gold would have become dirt cheap.

This little story, whatever its origin, captured the imagination oI ancient
authors. Pliny (Nat. 36.195) includes it in his encyclopedic work, though,
without giving it Iull credence. According to him a craItsman under Tiberius
invented a method Ior making glass unbreakable. For this his workshop was
destroyed in order that precious metals would not lose their commercial
value. In Dio Cassius (57.21.57) roughly the same story is made to exem-
pliIy Caesar`s jealousy and the moral oI the story is entirely diIIerent. There
the inventor is an architect who had already accomplished the restoration oI
a collapsing portico in Rome, Ior which Tiberius rewarded him with money
and exile. On another occasion when the architect was seeking pardon he
deliberately dropped a crystal goblet and then repaired it, evidently to show
oII his skill at restoration yet again, but this time it cost him his liIe. Isidore
(Etvm. 16.16.6.) has the story in a version similar to that oI Petronius and
attaches the same moral to it, even asserting that it is true that iI glass were
unbreakable, it would be better than gold and silver. John oI Salisbury (Pol.
4.5) relates the same story as that oI Trimalchio in Petronius but adds much
detail, although he preserves the moral oI the story: Caesar has the artisan
killed to prevent gold and silver Irom becoming cheap as dirt. With respect
to our investigation, Caesar in Trimalchio`s story uses his power to kill an
invention 'oI beneIit to the ages, in the words oI Eumolpus (88). And he
does this Ior exactly the same reason that the poet alleges as the cause oI the
demise oI contemporary science, namely an obsessive concern with the ac-
quisition and protection oI wealth.
The last time the topic occurs in the extant Satvrica is at the very end
where the motiI oI captatio or legacy hunting is treated. The comic evils oI
3 GENRE

232
captatio are a well-exercised theme in Greco-Roman satire. Horace (S. 2.5)
treats this theme in a parodic dialogue, which is a reworking oI a motiI Irom
the Odvssev, Book 11, and represents Tiresias advising Odysseus in the
underworld on how to enrich himselI upon arriving in Ithaca a poor man.
The topic is not originally Roman, oI course, since Horace is here borrowing
a Greek comic topos, which exploits the convention oI taking mythological
trips to the underworld to seek inIormation Irom the dead. The comic 3A'$&"
was a Iavorite device oI Menippus and other Cynics.
In the Satvrica, Croton is a city which has completely Iallen to this nasty
vice. Encolpius, Giton and Eumolpus have survived the shipwreck and just
buried Lichas, the owner oI the ship, when they take to the road again. From
a distance they see a city and inquire Irom 'a certain bailiII about the nature
oI the place (116). The ancient glory oI Croton is here deliberately empha-
sized in order to heighten the misery oI its current state. Encolpius usually
sketches the distant past in very positive terms to better highlight disgust
with the present. The overseer, who has no other purpose in the narrative
than to yield inIormation to the vagabond characters, is not unsurprisingly
concerned with the same cultural losses the narrator keeps lamenting. Liter-
ary studies are no longer celebrated in Croton, and neither is eloquence, and
the virtues are in a state oI neglect. No one raises a Iamily either, Ior people
with natural heirs are not likely to be courted by legacy hunters.
This absurd city stages yet another episode oI the Satvrica`s protracted
love aIIair between parasites and hosts. Eumolpus invents a conIidence trick
and poses as a shipwrecked man oI great property, without heirs, oI course,
and in miserable health, in order to stimulate the generosity oI the people oI
Croton. The spin which Encolpius puts on this traditional topic oI satire is
quite interesting. Letting Eumolpus & Co. outsmart the legacy-hunters by
exploiting their greed to their own advantage, the narrator again pitches one
group oI madmen against another and so retains his audience`s good will by
blunting the impact oI the satire.
What seems to be only a simile in the city`s description by the vilicus
vultures eating corpses in a plague-ridden countrysideis later translated
directly into action in the will oI Eumolpus. The eating oI his corpse is obvi-
ously proposed by Eumolpus as a deterring condition Ior those planning to
collect his imaginary inheritance, but not even this arrests the appetite oI the
human vultures oI Croton: 'The enormous reputation oI his money blinded
the eyes and minds oI those miserable people (excaecabat pecuniae ingens
fama oculos animosque miserorum). In the glorious hometown oI Pythago-
ras, the Iamous vegetarian philosopher, the citizens are willing to become
cannibals Ior Iinancial proIit, violating the most sacred taboo oI civilization.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

233
Eating human Ilesh is the proper behavior oI Cyclopes, one-eyed monsters
living beyond civilization and not bound by the laws oI any city. As in the
Iigurative language above, blinding is the punishment oI cannibalistic Cy-
clopes. The episode oI Croton, the uncivilized cityan oxymoron in Greco-
Roman political thoughtcan thus be read as that oI a troupe oI satyrs (En-
colpius, Giton and Corax), led by their Silenus (Eumolpus), who take up
habitation in the caves oI the Cyclopes (Crotoniates). In other words, a narra-
tive obeying in some sense the laws oI satyr-drama, the generic name oI
which was B%O" *"#$%&'F3, or in the plural *"#$%&'(, and the preIerred
plots oI which centered on the Iabulous tales oI the Odvssev.
The mention oI cities and civilization brings us lastly to the Iantastic
poem on the civil war, recited by Eumolpus just beIore entering Croton.
Here the same theme oI greed destroying the City dominates the sensational
description oI the causes oI the civil war between Caesar and Pompey (119).
This unIinished impetus oI a poem (hic impetus . nondum recepit ultimam
manum) oI almost 300 hexameter lines is, as we have seen, introduced by its
Iictional author as composed in the spirit oI canonical epic literature. But
what is begun by involving the gods in a Iabulously exaggerated version oI
Roman history characteristically ends with their Ilight and disgust with hu-
manity. Thus the loItiness oI traditional high poetry collapses into a descrip-
tion oI the sordid Iacts oI liIe.
The poem begins with a description oI Rome at the height oI her power
and wealth. Victorious, the Roman by now possessed the whole world, and
yet he was not satisIied (Orbem iam totum victor Romanus habebat . nec
satiatus erat). II there was still anywhere gold or sellable goods to be Iound,
that land, that people was declared an enemy and war was waged Ior proIit
(318). This public greed leads to shameIul immorality which is as hard Ior
the poet to relate (119 v. 19, heu, pudet effari perituraque prodere fata
|'Alas! one is ashamed to declare and reveal Iuture ruin|) as it was hard Ior
Encolpius to narrate the account oI Trimalchio`s domestic madness (70.8,
pudet referre quae secuntur |'One is ashamed to tell what Iollows|):

'Nec minor in campo Iuror est, emptique Quirites
40 ad praedam strepitumque lucri suIIragia vertunt.
venalis populus, venalis curia patrum,
est Iavor in pretio. senibus quoque libera virtus
exciderat, sparsisque opibus conversa potestas
ipsaque maiestas auro corrupta iacebat.
45 pellitur a populo victus Cato; tristior ille est,
qui vicit, Iascesque pudet rapuisse Catoni.
3 GENRE

234
namquehoc dedecoris populo morumque ruina
non homo pulsus erat, sed in uno victa potestas
Romanumque decus. quare tam perdita Roma
50 ipsa sui merces erat et sine vindice praeda.
Praeterea gemino deprensam gurgite plebem
Iaenoris illuvies ususque exederat aeris.
nulla est certa domus, nullum sine pignore corpus,
sed veluti tabes tacitis concepta medullis
55 intra membra Iurens curis latrantibus errat.
arma placent miseris, detritaque commoda luxu
vulneribus reparantur. inops audacia tuta est.
hoc mersam caeno Romam somnoque iacentem
quae poterant artes sana ratione movere,
60 ni Iuror et bellum Ierroque excita libido? (Sat. 119)

'The same madness is in public liIe, the true-born Roman is bought, and
changes his vote Ior plunder and the cry oI gain. The people are corrupt,
the senate oI the Iathers is corrupt, and their support hangs on a price.
The Ireedom and virtue oI the old men had decayed, their power was
swayed by bribes, and even their dignity was stained by money and trod-
den in the dust. Cato is beaten and driven out by the mob; his conqueror
is unhappier than he, and is ashamed to have torn the rods oI oIIice Irom
Cato. Indeed it wasn`t just the manand this was the disgrace oI the
people and the ruin oI its ethicswho was driven out, rather in his per-
son the power and glory oI Rome were conquered. So Rome in her deep
disgrace was herselI price and prize, and she despoiled herselI without
any one to avenge her. Moreover, the greed oI usury and the handling oI
money had caught the common people in a double whirlpool and de-
stroyed them. Not a house is saIe, not a man but is mortgaged; the mad-
ness spreads through their limbs, and trouble bays and hounds them
down like some disease sown in the dumb marrow. In despair they turn
to violence, and bloodshed restores the good things lost by excess. The
poor are reckless Ior they have nothing to lose. While Rome was plunged
in this Iilth, how could the healthy rationality oI the arts have moved her
Irom her slumber? No, to her only madness and war and lust aroused by
the sword were exciting.

According to Eumolpus` argument, the depravity and loss oI virtue in the
senatorial class is to blame Ior the initial corruption which led to the civil
war. In view oI the character oI Encolpius` audience and the dominance oI
3. 1 ANCI ENT NARRATI VE I N PERSONI S

235
the theme oI money in the work as a whole, this passage seems oI great im-
portance Ior understanding the underlying moral presuppositions oI the nar-
rative. What is more, it seems on the Iace oI it that we are to think oI the
civil war as Iinally putting an end to the arts and glory oI antiquity and inau-
gurating the shameIul contemporary state oI aIIairs, a thoroughly unoriginal
and corrupt Rome. II we add to this the gloriIying description oI the
Younger Cato and the unIavorable treatment oI the mad general Julius Cae-
sar, we are Iorced to recognize here an example oI the rallying cry oI liber-
tas, a nostalgic reIrain in the rhetoric oI the Roman senatorial class under the
principate, sometimes called the Stoic opposition. The Younger Cato was an
unmistakable symbol Ior these sympathies, especially under Nero. He was
looked upon as the last deIender oI the Roman Republic and libertas (not, oI
course, a democracy but an oligarchy in the hands oI the noble Iamilies).
431

Lucan gives an imposing portrait oI Cato in the Pharsalia and Ior Seneca
he is the ideal Roman Stoic. These may be deIlated ideals to us, but in the
early empire such ideas could easily be seen as politically Ilammable. How-
ever, by putting them in the mouth oI the crazy poet Eumolpus, Encolpius
can promote the message comically without committing himselI to anything.
These lines, surprisingly considering the nature oI the audience, but unsur-
prisingly considering the persona adopted by the narrator, contain a criticism
oI the old Roman aristocracy. Eumolpus says that with the loss oI virtue and
the onset oI venality among the populus and the curia patrum, i.e., the ruling
upper classes (1930), the plebs was deIenseless against usury, mortgages
and bondage Irom debt (315). Eumolpus has beIore blamed the senatus Ior
not living up to its role as moral authority (88, ipse senatus, recti bonique
praeceptor). And it seems that iI these opinions oI hiswhich the narrator
has prudently assigned to a character other than his youthIul selIare to be
acceptable to Encolpius` audience, they must be oI unusual moral weight Ior
Roman aristocrats.
3.1.8 Ideal vs. Real Audience
We can see Irom this that the implicit or ideal audience oI the Satvrica is
clearly not a real audience but an idealized construct oI the narrator and his

431
CI. Sullivan 1985a, 117II. But Sullivan who in general treats Petronius as a Ilatterer oI
Nero, and more so than Ior instance Seneca or Lucan, nowhere treats the political implica-
tions oI this passage. My own impression is that the political views oI the narrator oI the Sa-
tvrica are not signiIicantly diIIerent Irom what is expressed by other writers and intellectu-
als oI the time.
3 GENRE

236
text. It is important that we do not believe excessively in the historicity oI
Encolpius` good audience, who really play no more important role in the text
than as intelligent and respectable Ioils to his comic act. The ideal audience
is the locus oI the Satvrica`s moral sanity and eIIectiveness as social and
cultural satire, since the narrator, at least on the Iace oI it, is just a clown and
a Iool. But the sanity oI the audience does not derive Irom any actual per-
Iormance beIore an unusually responsible, clear-sighted and truly cultured
group oI aristocrats. It`s the narrator and his text that create the audience and
its values, and they do so obliquely, by triangulation Irom several Ilawed
perspectives. What the narrator has Agamemnon, Echion, Eumolpus, and his
own younger selI say and do about education postulates, when all put to-
gether, an audience capable oI drawing sane and sophisticated moral conclu-
sions about good and bad paideia and the state oI letters and the arts. But the
satire need not be taken seriously, because it is not advanced in a serious
way by people whom the audience has to take seriously. Even when Eu-
molpus criticizes the corrupt senate, no oIIense will be taken, because the
narrator has previously taken care to marginalize Eumolpus as a crazy poet.
It is not that the historical audience oI the Satvrica was composed oI
people who were unusually tolerant oI criticism or personally upright. The
audience, like audiences Ior satire in all times and places (including, e.g., the
Athenian demos listening to Aristophanes), are ordinary hypocritical human
beings. They tolerate the satirist not because they are truly exempt Irom his
criticism but because the eIIective satirist makes sure the 'who-the-hell-
does-he-think-he-is? moment never arrives. There is also the matter oI Ilat-
terythe real audience wishes to identiIy with the implied audience, which
is beyond moral criticism. From the historical author`s perspective, Petronius
achieves this eIIect, (i) through an inIerior narrator who 'knows his place
and Ilatters his audience, (ii) through the limitations oI the various personae
through whom the narrator speaks, and (iii) through the limitations, triteness,
even absurdity oI the various discourses oI those personae.
3.1.9 The Attributes oI Encolpius the Prick
The Satvrica is a perIormance text, and this makes the act oI telling the story
even more obtrusive than it would otherwise be. In this respect it is very
diIIerent, Ior example, Irom a modern novel, which is typically written Ior
the silent and solitary reader. To give due respect to this perIormance aspect
oI the Satvrica, we need to place beIore our eyes, as it were, the Iigure oI
Encolpius as he relates his adventures to his audience. Although nothing
3. 1 ANCI ENT NARRATI VE I N PERSONI S

237
absolutely necessitates the geographical location in Rome oI this implicit
setting, the capital itselI would be most in keeping with the nature oI the
audience. There is no need, however, to be too speciIic about this, since it
could be argued that Ior such a generic audience a generic setting is all that
is needed.
The importance oI dress, gesticulation, and the use oI the voice is great
in any perIormative situation, and the Romans in general were no exception
in assigning value to appearances. However, beyond the internal textual evi-
dence oI great liveliness in Encolpius` style oI perIormance (see section
1.2.4), we know little about how the narrator is supposed to be dressed or
groomed when he rides through his tale. On the other hand, inIormation is
abundant about such things as clothing and general appearance when we
look at the young Encolpius traveling through Italy. It is, however, diIIicult
to Iigure out how much oI this description is still valid Ior the storyteller,
some unspeciIied time aIter the adventures are over. We must necessarily
proceed with caution here, and yet it is clear that the inIormation provided by
the narrator about his past selI, beyond describing the protagonist directly,
also Iunctions as an indirect selI-portrayal. His attributes are built up Irom
several sources, both explicit and stated in his narrative, and implicit and
based on well-known cultural and literary stereotypes.
To take an example, when Encolpius tells us oI the practical joke played
on him by the anus urbana (7.2), the implications are not likely to be lost on
the audience. The old woman led him to a brothel which, she said, was
where he 'ought to live (7.2, 'hic` inquit 'debes habitare`). II we gather
the details oI the appearance oI this unhardened youth and his Iriends
(102.12, iuvenes adhuc laboris expertes), we begin to grasp the Iull meaning
oI the old woman`s mockery. As Encolpius tells us himselI, he had long
locks oI hair Ialling over his ears (18.4), the loss oI which caused him much
grieI (108.1, 110.4), a sorrow that was, however, easily cured by the aid oI a
curly blond wig (110.5). For clothes, he wore a short undergarment or tunic,
the so-called N&#;3 (12.5, 19.5), and on his Ieet he had the unsoldierly Greek
slippers, phaecasiae (82.3)
432
Fortunata too wears golden phaecasiae (Sat.
67.5)which quickly ruin his attempt to pass himselI oII as a member oI the
military. This general 'soItness does not seem to have altogether disap-
peared in the necessarily somewhat older Encolpius who tells the story, as is

432
)"&'(*&" were white shoes traditionally worn by Athenian gymnasiarchs, as well as Attic
and Alexandrian priests. Antony wore a pair as a part oI a typically Greek costume, when
he was in Alexandria acting as a private individual and Irequenting the schools and temples
oI that learned city (App. BC 5.11).
3 GENRE

238
indicated by the Iact that the Satvrica is a love story about himselI and the
boy Giton.
Quite a lot may be deduced Irom the Iact that he is a Massaliot, although
he seems to have acquired some highly Roman attributes by the time he tells
the story, attributes which he then unrealistically projects on to his younger
selI, especially his proIiciency in the Latin language and knowledge about
Roman literature. This Greco-Roman blending in Encolpius is, oI course,
made more complicated by the method oI composition apparently used by
Petronius, i.e. his transIormation oI a Greek text into a Roman one (see more
on this below in section 3.2). Nevertheless, a constant oI his identity, at least
as a youth, is his origin in Massalia. Community stereotypes were as com-
mon in the ancient world as they are now (consider Ior example Homer`s
Phoenician pirates and Cretan liars), and some oI Encolpius` attributes are
Ieatures oI his Massaliotic origin. The conservative and old Iashioned Mas-
saliots, according to ancient sources, tied their hair in a knot, wore long and
many-colored (women`s) dresses, used perIumes and were accordingly con-
sidered soIt and delicate, even eIIeminate. As we have seen in section 1.2.5
the Greek phrases 'you might sail to Massalia or 'you are coming Irom
Massalia were proverbial expressions Ior someone who had Iallen into
luxurious and eIIeminate ways.
433
The sources say that the women in this
community were not allowed to drink anything stronger than water, the place
being wanton enough without the women getting drunk on wine.
434
Thus the
phaecasiae (82.3) worn by Encolpius are not necessarily a sign oI his want-
ing to prostitute himselI, but may also be seen as the regular manner in
which respectable Greek men dressed in Massalia. As we have seen above,
his Ioreign style oI dressing is open to two interpretations, it can be seen as
archaic and respectable or luxurious and eIIeminate.
However, as Encolpius` narrative progresses, it becomes ever clearer that
the wicked old woman was not alone in associating the young man with
prostitution. The curious foedus oI peace brokered by Eumolpus on the ship
(109.23) is mostly taken up with the speciIication oI the price to be paid by
the adults Ior having sex with the boys (200 denarii Ior Encolpius, halI that
amount Ior Giton). Furthermore, the handmaiden oI Circe, Chrysis, provides
a detailed description oI this aspect oI our hero:

'quia nosti venerem tuam, superbiam captas vendisque amplexus, non
commodas. quo enim spectant Ilexae pectine comae, quo Iacies
medicamine attrita et oculorum quoque mollis petulantia, quo incessus

433
Suda, 8. 499, 3161; Ath. 12.25.
434
Ael. JH 2.38. The same custom was supposedly also observed in luxurious Miletus.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

239
arte compositus et ne vestigia quidem pedum extra mensuram aberrantia,
nisi quod Iormam prostituis ut vendas? (Sat. 126.12)

'Because you know your sex appeal, you become arrogant, and sell your
embraces, instead oI granting them Ireely. What else is the point oI your
nicely combed hair, your Iace plastered with make-up, the soIt glance in
your eyes, and your walk arranged by art so that never a Iootstep strays
Irom its place? It means, oI course, that you are prostituting your beauty
and selling it oII.

From this passage we learn Ior example that young Encolpius wore make-up.
Should we think oI the narrator, too, as wearing make-up? It is hard to say.
One`s immediate assumption may be that iI the narrator is considerably older
he might no longer wear make-up. However, based on the above description
wearing perIume and perhaps also make-up did not mean to Massaliots what
it means to Chrysis.
Lichas` 'recognition oI Encolpius reveals another important attribute,
Ior the captain identiIies the bald and disguised Encolpius not by his Iace
(and hands), but by his genitals, which he may be seen as addressing directly
by the name oI Encolpius since as he utters the greeting his hand and eyes
are headed in that direction (105.9, nec manus nec faciem meam consi-
deravit, sed continuo ad inguina mea luminibus deflexis movit officiosam
manum et 'Salve` inquit 'Encolpi`). This recalls the earlier witticism oI
Eumolpus about Ascyltos: 'his genitals hung down with such massive
weight that you`d have thought the man himselI was a mere appendage to his
prick (92.9, habebat enim inguinum pondus tam grande ut ipsum hominem
laciniam fascini crederes). The appeal oI the genitals is such that the man
himselI becomes but an extension oI them.
435
From this observation we can
Iurthermore conclude that when Lichas addresses Encolpius` genitals by his
proper name, the implication is that the man Encolpius takes his name Irom
his genitals. Encolpius` phallic identity, oI course, is most clearly evident in
his redende Name. It is thereIore not surprising to Iind that a large part oI the

435
Quartilla is as interested in Giton`s immature genitals as she seems to have appreciated
those oI Encolpius, the two-legged donkey: 24.7, mox manum etiam demisit in sinum et per-
tractato vasculo tam rudi 'Haec` inquit 'belle cras in promulside libidinis nostrae mili-
tabit, hodie enim post asellum diaria non sumo ('Then she slipped her hand inside the
Iolds oI his clothes, and Iondled his very untried penis. This will make a good starter to
rouse our desire tomorrow, since I`ve already had the donkey today, I don`t want small ra-
tions`). Eumolpus also careIully examines with both hands Encolpius` resurrected member
(140.13).
3 GENRE

240
extant narrative consists oI a long episode relating the dysIunction oI Encol-
pius` mentula, a treatment oI this subject which is unique in its Irankness and
scope in all oI ancient literature.
436

The phallic clown is a well-known archetypal Iigure in Greek culture.
His earliest appearance may perhaps be noted in the Homeric underdogs,
Thersites and Iros, and certainly in the parodic epic Iigure Margites and the
philosophical clown and slave Aesop. Other marginal Iigures oI literature
include the satyrs and especially Silenus, whose looks were commonly made
use oI to describe the philosopher Socrates. Another philosopher, Diogenes
the Cynic, may be counted in, and such comic characters as the stupidus oI
mime and the scholasticus oI ancient jokes belong to the type. Beside the
obvious attribute, the phallus, this Iigure was regularly identiIiable by its
shaven head or baldness, pointed ears, snub nose, and a pot-belly. Not all oI
these attributes need be present Ior the Iigure to be recognizable. Baldness is
an especially prominent sign oI the creature`s phallic identity, oIten repla-
cing the phallus itselI. The reason Ior this is that baldness, besides apparently
making the man himselI look more phallic, was intimately associated with
sexual activity in Greco-Roman culture.
According to Aristotle`s Historia Animalium loss oI hair Irom the head
or eyebrows only occurs aIter a man has become sexually active (HA 518a).
Moreover, no boy or woman or castrated man ever goes bald (HA 518a).
Likewise, the hairs oI the eyelashes are said to Iall oII when sexual activity
begins, and the more the greater this activity is (HA 518b). Finally, in those
who are given to sexual activity the congenital hair (the hair on the head,
eyelids, and eyebrows) is more likely to Iall oII (HA 518a, A-$*& BC O,,-3
"g #%+N8. #-9. J)%-B&*&"*#&'-9. "g *$//8389.). Incidentally, Aristotle himselI
is the target oI invective preserved in the Greek Anthologv, which casts him
as the lecherous bald man with the other deIining characteristics oI the phal-
lic clown.
437
Plato`s description oI Socrates` looks in the Svmposium as re-
sembling those oI Silenus relies on the same easy association. The transIeral
oI the attributes oI Silenus (baldness, snubbed-nose and pot-belly) to Socra-
tes became the basis oI his portraits, and a cliche in later literature.
438
In the-
atrical costumes, baldness could represent phallic looks with or without the
accessory phallus, and was used to characterize parasites, slaves, cooks,
(dirty) old men, moneylenders, and in general male characters with strong

436
The topic has been studied recently in detail by John M. McMahon 1998.
437
Anth. Graec. Appendix, Epigr. irrisoria, 11:!!&'%F., )","'%Q., #%"$,Q., D !#"/8&%+#K., /
,"/3Q., G%-/(*#4%, G",,"'"9. *$3KA3-. ('short, bald, stammering is the Stagirite |Aris-
totle|; lecherous, pot-bellied, and the associate oI whores).
438
Var. Men. 490, tam glaber quam Socrates ('as bald as Socrates)
3. 1 ANCI ENT NARRATI VE I N PERSONI S

241
corporal appetites over which they have little or no control. The pimp oI
New Comedy is bald and has a phallic name (e.g., Sannio, Ballio).
439
These
types, moreover, were the Iavorite subjects oI vase painters and makers oI
terra-cotta Iigurines.
440
In mime, the most common character was a bald
clown called the 'stupidus, who was usually a cuckolded husband and
would regularly be beaten during the act.
441

When the shaven and eyebrowless protagonist Encolpius is recognized
by his prick by Lichas, he resembles a cinaedus (Gel. 6.12.5, and Cic. Q.
Rosc. 20). This is perhaps a sign oI what is coming, and Encolpius is cer-
tainly in some sense a cinaedus since he delivers his description oI an at-
tempt at selI-castration in the Sotadean meter (132.8). But he also resembles
the men with shaven heads, memorably described by Lucian (Merc. Cond.
1), who gather in crowds at the temples and spin yarns about their shipwreck
and unlooked-Ior deliverance, which is exactly what the narrator Encolpius
is doing as he gives us this account oI his shameIul 'recognition and subse-
quent unexpected salvation.
The association oI baldness and a phallic nature was equally close in late
republican and early imperial Rome, as is testiIied by a popular verse against
Julius Caesar, Urbani, servate uxores. moechum calvom adducimus ('Men
oI the City, guard your women, we are bringing a bald adulterer),
442
and
another verse by Juvenal, Cum . / . calvo serviret Roma Neroni ('When
. Rome was slave under bald Nero), which is given a sexual interpretation
by Servius.
443
Pliny reports, in language reminiscent oI Aristotle`s Historia
Animalium, that among the hairy tribes oI the Alps and Gallia Comata loss
oI hair is rare in the case oI a woman, unknown in eunuchs, and never occurs
in any case beIore sexual intercourse has taken place.
444
Seneca, the philoso-
pher, believed that baldness and gout in women were recent, and a sign oI
the times. Born to be passive in sex (pati natae), women had violated the law
oI nature by somehow (Seneca does not elaborate this point) becoming ac-
tively engaged in sex through penetrating men (viros ineunt). As a Iitting
punishment, argues the philosopher, they now have lost the privileges oI
their sex and are beginning to suIIer Irom virile diseases (damnatae sunt
morbis virilibus).
445


439
See, e.g., Plaut. Rud. 371, and Pollux 4.145, 'receding hairline or bald.
440
For illustrations, see Nicoll 1963, 4389.
441
Maxwell 1993, 10
442
Suet. Jul. 51; cI. SHA 15.14.2, calvus moechus ('bald adulterer).
443
Juv. 4.38, and Serv. Aen. 4.214.
444
Plin. Nat. 11.131.
445
Sen. Ep. 95.21; see Adams 1982, 190I.
3 GENRE

242
II I am correct in assuming that Encolpius the narrator is a bald man in
conIormity with his age and typea point which will necessarily remain
speculativea good parallel to him would be oIIered by Lucius, the narrator
oI the Metamorphoses oI Apuleius. This luckless young nobleman turned
phallic creature by magic lotions appropriately portrays himselI at the end oI
his narrative as completely bald or shaven, and thus marks himselI oII as
being still a phallic clown at the time oI narrating his story, despite having
regained his human Iorm and lost the donkey-penis (the shaven head oI
Lucius the pastophorus is as ambiguous as Encolpius` white shoes; it is open
to a respectable religious interpretation and a shameIul phallic one). SigniIi-
cantly, in the shape oI an ass, Lucius, like the 'stupidus oI mime, is con-
stantly being beaten. Jack Winkler speculated in this context that an openly
phallic narrator could well be an indication oI a certain lack oI control over
the material oI the story. Such powerlessness would oI course be a narrative
stance and not indicate a lack oI sophistication. Lack oI narrative control can
maniIest itselI in Iormal Ieatures such as mixture oI discourses (an indication
oI inconsistency and lack oI authority in the character), a method oI compo-
sition that may go all the way back to the Homeric poem Margites.
This is so despite the Iact that the phallus may also be thought oI as the
prime symbol oI dominance and patriarchal authority. Although ithyphallic
eIIigies were common at Iestivals, as well as in apotropaic trinkets and wall-
carvings, those who actually exercised patriarchal authority seem to have
rigorously concealed this organ or preIerred to portray it as small and imma-
ture.
446
Thus, Right in the Clouds oI Aristophanes promises a small penis as
one oI the good results oI an old-Iashioned aristocratic education.
447
As a
dramatic device in Old Comedy, the depiction oI men with huge artiIicial
erections was designed to produce laughter. Mocking laughter seems like-
wise to be what met the bald man oI antiquity on stage and oII.
448
These
'deIormities were seen as a sure sign oI a low and shameless nature, and
thereIore justiIied the instantaneous degradation oI the individual in ques-
tion. A male who is so obviously not in control oI his appetites is seen as the
very opposite oI the Iree male citizen, who was required to adhere rigorously
to the ideals oI individual autonomy and restraint. Any sign oI obsessive
behavior or dependency can quickly become a liability when the basis oI the

446
A notable exception Irom this rule is Caesar`s veiled obscene threat to the senate aIter his
military successes in Gallia that proinde ex eo insultaturum omnium capitibus ('Irom then
on he would shame the heads oI all |senators|), reported by Suetonius (Jul. 22.2).
447
Dover 1978, 1256; Henderson 1991, 109.
448
Plut. Mor. 86b92I (88 E 10); Lucian DMort. 1.2; D.C. Hist. Rom. 76.8.4.; Schol. Ar. Nu.
145a; and Syn. Calv. encom. 7.
3. 1 ANCI ENT NARRATI VE I N PERSONI S

243
individual`s citizen-rights derive Irom the assumption that he is a Iree and
rational agent, and when the male citizen`s mastery oI his appetites is seen as
a condition Ior his mastery oI his social inIeriors.
449
We have thereIore the
paradox oI the over-sized phallus which is nevertheless unmanly.
We may speculate that Encolpius the narrator oI the Satvrica is bald
without deriving this attribute Irom anything more speciIic than a stock Iig-
ure oI Greco-Roman culture. Just as Lucius, in the Iorm oI an ass, is con-
stantly beaten and ridiculed, the luckless male prostitute Encolpius is sought
aIter and used Ior the sexual gratiIication oI other characters in the story, and
is thereIore the very opposite oI a dominant male. Although these two narra-
tors share such Iundamental aspects, we need not rely on CiaIIi`s thesis that
Apuleius transIerred some oI Encolpius` traits to his Latin adaptation oI the
Greek Ass-Story.
450

II Encolpius the narrator were bald, every reIerence to the hair that he
was so proud oI in the past would thus acquire special signiIicance. Take Ior
example the discomIort expressed in these words, 'I continually kept hiding
. my Iace because I understood that I was marked by no ordinary ugliness
(110.4, abscondebam . frequentius vultum intellegebamque me non tralati-
cia deformitate esse insignitum). We may compare this with Lucius` long
and loIty eulogy about the importance oI hair on the head oI beautiIul
women (2.89). Coming Irom bald and phallic narrators such emphasis on
the beauty oI hair makes good sense as comic characterization, but would
otherwise be rather pointless. It would also be a Iurther sign oI shameless-
ness Ior Encolpius to deliberately draw attention to this signiIicant (lack oI)
attribute, and make Iun oI it.
A work called Satvrica,
451
which Ieatures the travelogue and erotic mem-
oirs oI a man called Encolpius, narrated in a mixture oI discursive types to
an audience oI his betters, is very likely a continuation oI the general comic
tradition which begins with Margites and continues with satyr plays and the
comic theater. II young Encolpius is a satyr, the narrator may be likened to a
Silenus. They are both oI related nature, but the older one has lost most oI
his youthIul pretensions and posturing (along with his hair), and acquired a
giIt Ior telling stories. The distance between the implied nobility and virtue

449
As is argued by Halperin 1990, 88112, and Edwards 1993, 6397.
450
CiaIIi 1960.
451
The adjective *"#$%&'F. is used, Ior example, by Plutarch to describe the Iun loving crowds
oI the theaters and palaestra, to which the emperor Nero had been so generous: 'men who,
like satyrs, live Ior each day0)X8%-& '"R *"#$%&'-R #-9. j+-&. P3H%4G-& (Galba 16); cI.
the words oI Eumolpus at Sat. 99.1.
3 GENRE

244
oI the audience and the shamelessness and low social status oI the narrator
assures the eIIectiveness oI both the comedy and the satire.

3.2 The Hidden Genre
3.2.1 Origins
In this Iinal chapter, I intend to submit new arguments regarding the origin
and mode oI composition underlying the Satvrica oI Petronius. I have at-
tempted earlier to show that the text under scrutiny is written expressly Ior
perIormance by a single actor, or ancient lector in the sense oI a lively re-
citer. It has also been shown, in the central chapters, that the original Sa-
tvrica was not as radically episodic as is oIten assumed, but rather exhibited
a central plot constructed around the person oI the narrator, and organized by
a technique which on the whole resembles that used in other known Greek
and Greco-Roman prose Iictions. In the preceding chapter, moreover, we
have attempted to deIine and describe the genre oI the ancient 'personal
recollection novel, and discussed the narrator Encolpius, his 'philosophy,
attributes, and comically subordinate social and moral status vis-a-vis his
audience. All these arguments have aimed at restoring to the Satvrica its
original Iorm as an extended Iictional narrative with a unity and logic oI its
own, unrelated to the author`s biography. It is now time to place the work as
described in the context oI Greco-Roman literary history.
By scrutinizing the peculiar mixture oI Greek and Roman linguistic and
cultural elements in the text, I argue that the realism oI the Satvrica is illu-
sory, because its blend oI Greek and Latin cultures was nowhere and never
exempliIied in the real world. To account Ior the hybrid nature oI the work, I
advance a new hypothesis regarding the composition oI the Satvrica, which
can resolve various diIIicult problems in the traditional scholarship without
diminishing the value oI much excellent work in the Iield. According to our
hypothesis Petronius was not merely working in the tradition oI the Greco-
Roman comic novel when he wrote the Satvrica, but was Ireely adapting a
speciIic Greek text, now lost, which was likewise written in a mixture oI
discourse types. Because oI the many Roman elements in the Satvrica, stud-
ies oI its intertextual relationship with other Roman works are important,
even iI the work as a whole has a Greek model. Petronius can, now as be-
Iore, be seen to draw on other genres, both Greek and Latin, Ior various
other eIIects, techniques, themes, and subjects, though the emphasis should
3 GENRE

246
naturally be on the main generic Ieatures oI the narrative.
452
In my view the
genre oI the Satvrica is neither an original invention nor a synthetic amalgam
oI various other genres without a unity and character oI its own, but a deIin-
able and known ancient genre with distinct Ieatures.
The direction oI my argument calls Ior an exploration oI the possible
antecedents oI the Greek work behind the Satvrica, preserved as little more
than titles, although sometimes helpIully described in ancient testimony.
Later texts belonging to the same genre will also be taken into account, in-
cluding the lost Greek model oI the Metamorphoses oI Apuleius and recently
discovered Greek papyri containing Iragments oI prosimetric and non-
prosimetric erotic and criminal Iictions. This evidence should make it possi-
ble, in conjunction with our explication oI the narrative structure oI the Sa-
tvrica, to provide a rough description oI the literary Iorm oI the Satvrica, a
Iorm that in antiquity even had a name oI its own.
3.2.2 The Analytic Rigor oI Nationalism
The greatest obstacle to revising our understanding oI how the writer oI the
Satvrica went about composing his work is not so much the lack oI ancient
sources as the peculiar place occupied by Petronius Arbiter in modern narra-
tives oI Roman literary history. In the last decades oI the nineteenth century,
Petronian scholarshipunder pressure Irom pervasive ideologies, relating to
the consolidation oI national states in Europe, and the general upheaval
caused by the revolutionary progress in science and technologybegan to
invest unstintingly in a vision oI Petronius as a national writer and a great
innovator. According to this new interpretation, Petronius had, in the Iashion
oI contemporary writers oI Naturalist documentary novels such as Emile
Zola, invented a new Iorm oI literature Ior describing the daily liIe and man-
ners oI his ancient Italian Iellow countrymen. This conception oI Petronius
as the ancient Roman master oI a modern literary genre coincided with the
tendency oI politically active German philologists to see in the uniIication oI
ancient Italy under Roman rule a classical model Ior their own project oI
building a modern national state.

452
Petronius can be seen, with Sandy 1974b, as drawing on contemporary Iarce as continuing
metaphor, and, with Panayotakis 1995, as drawing on the same genre to Iacilitate the
imaginative representation or 'staging oI certain episodes oI the narrative; likewise he can
be seen, with Bcheler 1862, as drawing on Varronian satire Ior the application oI
prosimetry in Latin, and, with Beck 1982, as drawing on Lucilian satire in the adaptation oI
the narrative persona.
3. 2 THE HI DDEN GENRE

247
Although a modiIied version oI this romantic idea oI our author has until
recently held a great deal oI currency in the scholarship, it cannot be taken
Ior granted that today`s students oI the Satvrica are generally Iamiliar with
the books and articles where this interpretation was Iirst presented with the
backing oI learned scholarly argumentation. To name a distinguished exam-
ple, Winkler and Stephens, in a commentary on the Iolaos Iragment, claim
that the problem oI composition in Petronian scholarship has been inIlu-
enced by the 'prejudice oI Hellenists that Petronius must have had` a Greek
antecedent.