The Gatekeeper: Narrative Voice in Plato’s Dialogues

Brill’s Plato Studies Series

Editors

Gabriele Cornelli (Brasilia, Brazil)

Gábor Betegh (Cambridge, United Kingdom)

Editorial Board

Beatriz Bossi (Madrid, Spain)

Luc Brisson (Paris, France)
Michael Erler (Würzburg, Germany)
Franco Ferrari (Salerno, Italy)
Maria do Ceu Fialho (Coimbra, Portugal)
Mary-Louise Gill (Providence, usa)
Debra Nails (Michigan, usa)
Noburu Notomi (Tokyo, Japan)
Olivier Renaut (Paris, France)
Voula Tsouna (Santa Barbara, usa)

volume 2

The titles published in this series are listed at brill.com/bpss

The Gatekeeper: Narrative Voice
in Plato’s Dialogues

By

Margalit Finkelberg

leiden | boston
The Library of Congress Cataloging-in-Publication Data

Names: Finkelberg, Margalit, author.

Title: The gatekeeper : narrative voice in Plato's dialogues / by Margalit
Finkelberg.
Description: Leiden ; Boston : Brill, 2019. | Series: Brill's Plato studies
series ; volume 2 | Includes bibliographical references and index.
Identifiers: lccn 2018044129 (print) | lccn 2018045307 (ebook) | isbn
9789004390027 (ebook) | isbn 9789004390010 (hardback : alk. paper)
Subjects: lcsh: Plato. Dialogues. | Discourse analysis, Narrative.
Classification: lcc B395 (ebook) | lcc B395 .F493 2019 (print) |
ddc 184--dc23
lc record available at https://lccn.loc.gov/2018044129

Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill-typeface.

ISSN 2452-2945
ISBN 978-90-04-39001-0 (hardback)
ISBN 978-90-04-39002-7 (e-book)

Copyright 2019 by Koninklijke Brill NV, Leiden, The Netherlands.

Koninklijke Brill NV incorporates the imprints Brill, Brill Hes & De Graaf, Brill Nijhoff, Brill Rodopi,
Brill Sense, Hotei Publishing, mentis Verlag, Verlag Ferdinand Schöningh and Wilhelm Fink Verlag.
All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system,
or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise,
without prior written permission from the publisher.
Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided
that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite
910, Danvers, MA 01923, USA. Fees are subject to change.

This book is printed on acid-free paper and produced in a sustainable manner.

FOR ARYEH

∵
Contents

Preface IX
Abbreviations X

1 Introduction 1
1 “Diegesis through mimesis”: Classification of Narrative Genres in
Republic 3 1
2 The Theaetetus Passage 5
3 Plato as Literary Author 9
4 Narrative Voice in Plato’s Dialogues 13
5 Plato’s Narrator and Narrative Theory: Some Necessary
Adjustments 16

Part 1
The Dialogues

2 The Explicit Narrator: Narrated Dialogues 27

1 Introducing the Narrated Dialogues 27
2 A Single Narrator (the Charmides, Lysis, Republic) 31
3 Multiple Narrators (the Parmenides) 38
4 Conclusions 44

3 The Implicit Narrator: Dramatic Dialogues 47

1 Introducing the Dramatic Dialogues 47
2 The Theaetetus as a Test Case 51
3 Bifocality or a Single Focus of Perception? (the Euthyphro, Crito,
Menexenus vs. the Ion and Hippias Maior) 53
4 An Implicit Narrator-Hero (the Cratylus, Meno, Phaedrus,
Laws) 58
5 An Implicit Narrator-Observer (the Hippias Minor, Laches, Gorgias,
Philebus) 63
6 Socrates as an Implicit Narrator-Observer (the Sophist, Statesman,
Timaeus, Critias) 70
7 Conclusions 74
viii Contents

4 The Explicit and the Implicit Narrator Combined: Mixed

Dialogues 76
1 Introducing the Mixed Dialogues 76
2 The Protagoras 80
3 The Euthydemus 84
4 The Phaedo 88
5 The Symposium 93
6 The Theaetetus 97
7 Conclusions 101

Part 2
The Interpretation

5 Plato’s Experiments with Narrative Voice 105

1 Preliminary Remarks 105
2 A Single Focus of Perception 106
3 Multiple Narrative Levels 112
4 Abandonment of the Narrated Form 116
5 Conclusions 121

6 The Limits of Authority 124

1 Preliminary Remarks 124
2 The Narrator’s Text 125
3 Change of Interlocutor 129
4 Metanarrative Comments 132
5 Distribution and Clustering: Three Case Studies 135
6 Conclusions 147

7 The Narrator and the Author 149

1 Poetry and Painting in Republic 10 149
2 The Body of the Dialogue 150
3 The Gardens of Adonis 157
4 Mimesis and Reality 159
5 Representation of Narration 162
6 Conclusions 166

Bibliography 169
Index of Passages Cited 179
General Index 183
Preface

This book grew out of a lifelong fascination with Plato’s dialogues as literary
fiction. For many years my attitude to Plato was mainly that of an admiring
reader, until in 2003 I first gave expression to some of this book’s ideas in a
paper delivered at a panel on Plato as Literary Author, organized by Ruby
Blondell and Ann Michelini at the annual meeting of American Philological
Association in New Orleans. There were also other occasions, but it took more
than a decade until these initial ideas crystallized into a large-scale project
and the study and research of the dialogues began in earnest. This was a hard
experience, but fascinating and always rewarding.
I am greatly indebted to the participants in the conference on Framing the
Dialogues: How to Read Openings and Closures in Plato, held at the University
of Cyprus in December 2015, for their stimulating influence, and I would like
to thank Eleni Kaklamanou, Maria Pavlou, and Antonis Tsakmakis for organiz-
ing it. I am deeply grateful to Gabriel Danzig for his expert advice, to Aryeh
Finkelberg for his helpful and constructive criticism, and to Murray Rosovsky
for his dynamic and thoughtful copyediting. I am also grateful to the anony-
mous reader of my manuscript for his/her encouragement.
Abbreviations

AJP American Journal of Philology

BICS Bulletin of the Institute of Classical Studies
BMCR Bryn Mawr Classical Review
CJ Classical Journal
CQ Classical Quarterly
CP Classical Philology
G&R Greece & Rome
GRBS Greek, Roman and Byzantine Studies
HThR Harvard Theological Review
JHS Journal of Hellenic Studies
NLH New Literary History
OSAP Oxford Studies in Ancient Philosophy
PCPS Proceedings of the Cambridge Philological Society
Ph&Lit Philosophy and Literature
PT Poetics Today
RMM Revue de métaphysique et de morale
RIPh Revue Internationale de Philosophie
SSR G. Giannantoni, Socratis et Socraticorum reliquiae, collegit, disposuit,
­apparatibus notisque instruxit G. Giannantoni. Napoli: Bibliopolis, 1990.
TAPA Transactions of American Philological Association
YCS Yale Classical Studies
 Chapter 1

Introduction

1 “Diegesis through mimesis”:

Classification of Narrative Genres in Republic 3

In Book 3 of the Republic, Plato introduces the first-ever classification of forms

of narrative discourse (diēgēsis). He proposes the three following categories:
(a) the “plain” discourse (haplē diēgēsis), which works “through the report (di’
apaggelias) of the poet himself” and is best exemplified by first-person cho-
ral poetry; (b) the mimetic discourse (hē dia mimēseōs), which works through
impersonation and is best exemplified by dramatic poetry; (c) the “mixed”
discourse (di’ amphoterōn), which employs both mimesis and the voice of the
poet himself and is most adequately represented in Homer.1
Plato’s classification, in the form of an opposition between diegesis and mi-
mesis, has been enthusiastically embraced by contemporary literary theory.
Since the Greek diēgēsis roughly corresponds to what we today call “narrative,”2
the opposition in question has been widely used to highlight the distinction
between narrative genres (such as the novel) and the non-narrative (such as
drama). Paul Ricoeur seems to have been the first to point out that reading
Republic 3 in terms of an opposition between mimesis and diegesis issues from
misconstruction of Plato’s original meaning:

To avoid any equivocation, however, it should be recalled that in the

Republic, iii, 392c, Plato does not oppose diegesis to mimesis. Diegesis is
the only generic term discussed. It is divided into “plain” diegesis when
the poet narrates events or discourse with his own voice or diegesis “by

1 Resp. 392d5–6, 394b9–c5. The discussion of mimesis in Book 10 of the Republic addresses
both verbal and non-verbal forms of mimesis (such as painting and sculpture), and pursues a
different agenda: while Republic 3 treats mimesis from the perspective of narrative voice ad-
opted by different forms of discourse, Republic 10 explores the ontological status of mimetic
art as a whole. For a discussion of Republic 10 see Chapter 7.
2 Cf. Halliwell 2009: 18: “I shall employ ‘narrative’ to designate what Plato’s text, at 392d and sub-
sequently, treats as the genus diēgēsis, roughly equivalent to temporally plotted discourse.”

© koninklijke brill nv, leiden, ���9 | doi:10.1163/9789004390027_002

2 Chapter 1

imitation” (dia mimēseōs) when the poet speaks as if he were someone

else …3

Rather than being opposed to diegesis, mimesis is unambiguously identified

by Plato as its subcategory, that is, as a form of narrative. Although in the con-
text of Republic 3 the identification is applied to mimetic poetry, it obviously
bears on the way Plato’s own mimetic compositions should be approached.4
As for the mode of presentation, Plato’s dialogues fall into three groups:
(a) the narrated dialogues, presented as first-person narratives (the Lysis, the
Charmides, the Republic, with the Parmenides forming a special case);
(b) the dramatic dialogues, presented as direct speech exchanges (the major-
ity of the corpus);
(c) the so-called mixed dialogues, cast in both direct and reported speech; in
these dialogues, the first-person narrative is preceded by a direct speech
exchange between the prospective narrator and the prospective narratee
(the Protagoras, the Euthydemus, the Phaedo, the Symposium, with the
Theaetetus forming a special case).
Although this grouping of the dialogues bears no relation to the triple clas-
sification of narrative genres Plato introduced in Republic 3, it has often been
approached as such.5 But the identification is misleading, for Plato never
speaks for himself in his dialogues. As L.A. Kosman put it, “in each of them
[Plato’s dialogues] the narrative voice is itself mimetic.”6 This would eliminate
both the “plain” and the “mixed” narrative as classificatory options. Indeed,
as Plato’s narrator is never identical with the author, his narrated and mixed
dialogues should be considered as no less mimetic than the dramatic ones. At
the same time, considering that, as we have seen, Plato unambiguously classi-
fies mimetic genres as a subcategory of diēgēsis, his dramatic dialogues should
be approached as no less diegetic than the narrated ones. Therefore, as far
as Plato is concerned, all his dialogues should be identified as belonging to a

3 Ricoeur 1985: 180 n. 39. See also de Jong 1987: 3; Halliwell 2002: 54 n. 42; Morgan 2004: 357–59;
Hunter 2004: 24; Halliwell 2009: 18–19.
4 Cf. Halliwell 2009: 21. Mime and the Socratic dialogue were the only genres of prose fiction to
exist at the time; both are subsumed under “poetry” in Aristotle’s Poetics; see further Section
3 below.
5 This tendency can be traced to ancient classifications, see Clay 1992: 116.
6 Kosman 1992: 82. According to Genette (1990: 764–67), dissociation between author and
narrator is the primary criterion whereby fiction is distinguishable from non-fiction; cf.
Schaeffer 2013: 1 [2]. On the dissociation between author and narrator as “touchstone of
fictionality” see also Cohn 1990: 791–800. Cf. Schur 2014: 31, on “the literary face of Plato’s
discourse – defined as the obverse of authorial discourse, which is a sine qua non for philo-
sophical interpretation,” and Schur 2014: 70–73.
Introduction 3

single ­category, namely, mimetic narrative – or, in Plato’s own words, “diegesis
through mimesis.”7
It is worth dwelling at some length on the way Plato’s Socrates explains to
Adeimantus, his interlocutor in this part of Republic 3, how the classification
that he has introduced works. In an oft-quoted passage he turns into indirect
speech the spoken sections of the Chryses episode in Book 1 of the Iliad, the
foremost representative of the so-called mixed genre, thereby turning it into
“plain discourse” (393d1–394b1). Much less attention, however, has been paid
to what immediately follows the passage in question. “Or you may suppose,”
Socrates tells Adeimantus, “the opposite case – that the intermediate passages
in the voice of the poet (ta tou poiētou ta metaxu tōn rhēseōn) are omitted, and
the exchange of words (ta amoibaia) alone is left.” “You mean, for example, as
in tragedy?” Adeimantus asks, and Socrates replies: “You have conceived my
meaning perfectly” (394b3–8, cf. 393b8).
Would Socrates’ answer have been different had Adeimantus named the
Sōkratikoi logoi instead? As the following discussion demonstrates, in the lat-
ter case a further qualification, not necessarily literary, would be required.
Consider Socrates’ specification of the kind of mimetic discourse that will be
admitted in the ideal state:

Suppose, I answered, that a just and good man in the course of a nar-
ration (en tēi diēgēsei) comes on some saying or action of another good
man – I should imagine that he will like to speak in his voice (apaggel-
lein), and will not be ashamed of this sort of mimesis: he will be most
ready to imitate the good man when he is acting steadfastly and sensibly;
in a less degree when he is overtaken by illness or love or drink, or has
met with any other mishap. But when he comes to a character which is
unworthy of him, he will not wish to apply himself seriously to assuming
the likeness (apeikazein) of a bad man; he will do such a thing, if at all, for
a moment only, when that man is performing some good action, but he
will be ashamed of what he is doing, especially as he is not experienced
in imitating such persons nor would he like to fashion and frame himself

7 Cf. Morgan 2004: 358: “The only safe conclusion to draw from the passages discussed above
[Republic 3 and Theaetetus 143b-c] is that all Platonic dialogues are conceived as narratives. I
propose to adopt the same approach: even dramatic dialogues are by implication narratives.”
Cf. also Halliwell 2009: 19: “in terms of Socrates’ classification, drama is itself a species of
narrative diêgêsis, i.e. wholly mimetic diêgêsis (cf. 394c).” On Theaetetus 143b-c see the next
section.
4 Chapter 1

after the baser models; for his mind despises such things, unless they are
being done for the sake of play (paidias charin).8

Such a reputable imitator would adopt the model of Homer in his narrative
(diēgēsis), “and his speech (lexis) will take part in both, in the mimesis as well
as in the other form of narrative (tēs allēs diēgēseōs), but there will be very little
of the former in a long discourse (logos)” (396e4–7). He is the only kind of mi-
metic artist who will not be expelled from the ideal state:

For we mean to employ for our benefit the rougher and severer poet or
story-teller, who will imitate the speech (lexis) of the decent man and will
use the discourse (ta legomena) which follows the models we prescribed
at first when we began the education of our warriors. (398a7–b4)

The explanation is obviously self-referential,9 but not immediately. Had in-

deed Socrates authored the Republic, the above explanation would have ac-
counted perfectly well for both his position as the dialogue’s narrator and the
Republic’s narrative mode. But it does not account for Plato’s position or the
narrative mode of the Republic as authored by Plato and narrated by Socrates.10
That is, if Plato had been the dialogue’s narrator, then like Homer he would
have been speaking in his own voice, mixing “plain narration” (first-person
discourse) with purely mimetic passages. But the Republic’s narrative voice
does not belong to Plato, and this prevents us placing the dialogue with the
so-called mixed genres. Plato’s position is that of a reputable imitator who in
his discourse impersonates another good man, Socrates, and who ought not
“be ashamed of this sort of mimesis.” Accordingly, the Republic can only be
approached as “diegesis through mimesis.” The same obviously applies to the
other narrated dialogues and to the narrated parts of the mixed dialogues.
Still, the classification of narrative genres set out in Republic 3 does not suf-
ficiently clarify what precisely Plato had in mind when identifying dramatic

8 Resp. 396c5–e2. Unless otherwise indicated, the translations of Plato are those by Benja-
min Jowett, adapted where necessary.
9 Note especially the use of the expression “for the sake of play” (παιδιᾶς χάριν) at 396e2.
The expression is repeated verbatim in the Phaedrus’ programmatic enunciation of the
limitations of writing; it is used there to justify the writing of dialogues as a philosopher’s
pursuit in his leisure time (Phdr. 276d2, cf. 276e1–3). Cf. also Resp. 602b8: “mimesis is some
kind of play rather than serious pursuit (ἀλλ’ εἶναι παιδιάν τινα καὶ οὐ σπουδὴν τὴν μίμησιν).”
On Plato’s concept of paidia see below, Chapter 7.
10 Similarly Ferrari 2010: 20–22. But I do not share Ferrari’s contention that the Republic itself
cannot be subsumed under Plato’s classification; see below in this chapter.
Introduction 5

genres as narratives. Even more pertinently, it does not clarify how this identi-
fication is supposed to work when applied to Plato’s own dramatic dialogues.
If the dramatic dialogues are to be approached as narratives, whose narratives
are they? Ultimately, of course, they are Plato’s, but we saw that in the Republic
a fictional narrator rather than Plato himself delivers the story, as is the case
in his other narrated texts. No narrator, however, fictional or not, is explicitly
present in the dramatic dialogues. In view of this, should we still follow Plato
in blurring the boundary between the narrative and the dramatic discourse? Or
should we ignore his formula “diegesis through mimesis” as a purely theoretical
exercise? Fortunately, we possess a piece of evidence – a clue, as it were – that
allows us to form a clearer idea of how Plato himself would have approached
these issues.

2 The Theaetetus Passage

For Plato’s concept of narrative voice overall, Theaetetus 143b5–c5 is of utmost

importance. A moment before Euclides’ slave starts reading aloud his master’s
account of Socrates’ narration of his encounter with Theaetetus, Euclides says
to his narratee:

Here is the roll, Terpsion; in fact, I wrote this account (logos) in such a
way that Socrates appears in it not as narrating (diēgoumenon) to me
as he actually did, but as conversing (dialegomenon) with those he said
he had conversed with – Theodorus the geometer and Theaetetus. And
lest in the process of writing I should get into trouble with such inter-
mediary pieces of narrative (hai metaxu tōn logōn diēgēseis) like “I said,”
“I remarked,” whenever Socrates speaks of himself, or again, “he agreed,”
or “he disagreed,” whenever he speaks of his interlocutor, I let them con-
verse among themselves, having omitted such pieces altogether.11

The passage repeats Plato’s proposal, first made in Republic 3, to omit the mark-
ers of indirect speech12 thereby turning a narrated piece into a dramatic one,
and the dialogue implements it. But it also takes this experiment several steps
further.

11 Tht. 143b5–c5; my translation.

12 Cf. Resp. 394b4 τὰ τοῦ ποιητοῦ τὰ μεταξὺ τῶν ῥήσεων as against Tht. 143c1 αἱ μεταξὺ τῶν
λόγων διηγήσεις.
6 Chapter 1

As Gérard Genette has pointed out, the Theaetetus passage presents a nar-
rative situation that economizes on at least one narrative level, for the second-
ary narrative is presented here as the primary one: as a result, the origin of the
narration is blurred.13 To express the fact that, contrary to appearances, the
Theaetetus narrative belongs to the second rather than the first narrative lev-
el (the level that he calls “diegetic”), Genette dubbed this form of narrative
“pseudo-diegetic,” namely a kind of narrative that occurs when “a narrative sec-
ond in its origin is immediately brought to the first level.”14
Genette’s analysis of the Theaetetus’ switch from reported to direct speech
underscores the fact that Plato views the distinction between the narrated and
the dramatic dialogues as relating less to the mode of narration than to the
number of narrative levels that a given dialogue makes explicit. Approaching
the Theaetetus from the perspective of narrative mode, we would be equally
justified in regarding its “pseudo-diegetic” narrative as “pseudo-mimetic,” for
here a narrated dialogue is disguised as a dramatic one.15
But the Theaetetus passage does not just concur with Republic 3 in point-
ing out that the dramatic dialogues are to be approached as narratives and
demonstrating how the transition from the narrative to the dramatic discourse
may be effected; nor does it merely show that the distinction between the nar-
rated and the dramatic dialogues concerns the number of narrative levels they
exhibit. The Theaetetus passage highlights that insofar as they blur the source
of narration, the dramatic dialogues should be regarded as narratives whose
narrator is suppressed.
The idea that Genette’s pseudo-diegetic narratives should be approached
as “narratives with a suppressed narrator” was first expressed by Irene de Jong.
She also applied it to a broader corpus of ancient Greek literature:

In the case of philosophical dialogues and mimetic idylls or eclogues, we

are dealing with a suppressed primary narrator, and hence when charac-
ters in those poems start narrating they should be considered secondary
narrators.16

13 Genette 1980: 236–37 (= Genette 1972: 245–46).

14 Genette 1980: 240 (= Genette 1972: 248–49). Cf. Pier 2014: 3.2.2: “a narrative second in ori-
gin but which, lacking a diegetic relay, is narrated as though it were diegetic.” On Genette’s
taxonomy of narrative levels see below, Section 5.
15 Cf. Morgan 2004: 358: “The Theaetetus thus insists both that the body of the dialogue is a
narrative, and presents that narrative as a dramatic conversation.” Cf. also de Jong 2014a:
17, and below, with n. 20.
16 De Jong 2014a: 34. Cf. also de Jong 2004: 7–8, 2014b: 3.2.1, and below, n. 3 to Chapter 3.
Introduction 7

At the same time, identifying Socrates of the Theaetetus as a suppressed nar-

rator does not do full justice to his role in the dialogue. After all, we have no
doubt about the identity of the Theaetetus’ narrator. Suppressed though he
formally is, Socrates is nevertheless indirectly present in the text: he is the dia-
logue’s implicit narrator.
A similar strategy of economizing on a narrative level by presenting in
­dramatic form what is supposed to be a piece of narrative can also be sug-
gested for the Sophist and the Statesman, negotiated as continuing Socrates’
narration in the Theaetetus, and for the Timaeus and the Critias, negotiated
as continuing Socrates’ narration of what is supposed to be a version of the
­Republic (below, p. 71). Their dramatic form notwithstanding, these dialogues
are also presumed to be narratives whose implicit narrator is Socrates. But es-
pecially telling is the evidence provided by the Parmenides, a dialogue appar-
ently written in close proximity to the Theaetetus.17
The Parmenides is a narrated dialogue arranged as a series of nested ­boxes;
the characters on the bottom level of the narrative, which constitutes the
body of the dialogue, are the philosophers Parmenides and Zeno, their host
Pythodorus (who narrates the story), and two young men, Socrates and Ar-
istoteles, who have come to Pythodorus’ house to listen and talk philosophy
with the distinguished guests. Pythodorus’ account of the meeting is filtered
through two upper-level narrators, Pythodorus’ narratee Antiphon and the
latter’s narratee Cephalus. When the dialogue enters its second quarter, the
participants ask Parmenides to demonstrate his method of deduction from a
single hypothesis (135d-136e). After some hesitation, Parmenides accepts the
challenge, and Aristoteles, the youngest in the company, steps forward to act
as Parmenides’ respondent: “Ask, and I will answer.” “Let us proceed, he said
(phanai): If there is one, the one cannot be many?” (137c2–5; my emphasis).
The infinitive phanai (“he said”), signaling as it does that we are still within
an account of reported speech,18 is the last indication that the Parmenides is
a narrated dialogue. From now on, the dialogue proceeds, to all intents and
purposes, as a dramatic one.
The Parmenides’ switch from reported to direct speech, often seen as per-
plexing, can hardly be anything other than the same strategy of converting a

17 See n. 19 below; on the Parmenides see also Chapter 2, pp. 38–44, and Chapter 5, p. 117.
18 Translated literally, it would be something like “he [Pythodorus] said that he [Par-
menides] said,” presumably spoken by Antiphon. This mode of presentation, resulting
in that all the narrative levels are simultaneously present in the narrative, is especially
characteristic of the Symposium, see Section 5 below. On Antiphon and Pythodorus as
narrators see Chapter 2.
8 Chapter 1

narrated text into a dramatic one that we saw suggested as an experiment in

Republic 3 and implemented in full in the Theaetetus. As R.B. Rutherford put it,

The longer second part [of the Parmenides] is entirely in dialogue, as if

Plato had wearied of perpetually repeating “he said” and preferred to fo-
cus on the argument. The opening exchange in the Theaetetus seems to
meet this difficulty.19

As with Socrates the narrator of the Theaetetus, in the Parmenides too Plato
suppresses the dialogue’s narrator, Pythodorus, who nevertheless continues to
be implicitly present in the text as the ultimate source of the narrative.
Republic 3, the Theaetetus, and the Parmenides attest that in Plato’s view the
difference in presentation between the narrated and the dramatic dialogues
does not affect the status of the latter as narratives. For Plato, the only differ-
ence between the narrated and the dramatic dialogues is the manner in which
they negotiate their narrators: the narrator is explicit in the narrated dialogues
and implicit in the dramatic ones. This is as far as Plato’s theory goes. The ques-
tion however is whether Plato’s theory agrees with his literary practice.
If Plato conceived of all his dialogues, both the narrated and the dramatic,
as narratives, this would impart to his corpus a unity of design that to date has
not been explicitly recognized.20 Yet it is reasonable to suppose that, if indeed
Plato understood the dramatic dialogues as narratives with an implicit narra-
tor, this attitude must find its expression not only in the Theaetetus, where he
openly admits as much, nor only in the Parmenides, whose very design leads
to this conclusion, but also in the wholly dramatic dialogues. Testing the hy-
pothesis of implicit narrator, as well as its literary and theoretical implications,
against the entire corpus of Plato’s dialogues is this book’s principal objective.

19 Rutherford 1995: 274. Cf. Guthrie 1978: 64: “The Parmenides showed a transitional stage,
in which the narrative form is tacitly dropped half way through, and it is a fair inference
that, as has been assumed on other grounds, it slightly preceded the Theaetetus”; Thesleff
1982: 207: “the Parmenides … reflects in a significant way the trend towards abandoning
the reported form”; Brandwood 1990: 251: “… in the Theaet. Plato renounces the use of the
reported dialogue form, which seems to be merely an explicit declaration of a practice
already implicitly adopted early in the Parm. (137c).” On the relative chronology of the
two dialogues see Sedley 2004: 16–17, and below, Chapter 5.
20 De Jong remarks on several occasions that, since philosophical dialogues, alongside mi-
metic idylls or eclogues, can be presented in both dramatic and narrated form, the dra-
matic variety should also be approached as a narrative, see de Jong 2004: 7–8; 2014a: 17,
34, and above, with n. 16; she does not expound further on these remarks. Morgan 2004
arrives at a similar conclusion on the basis of Republic 3 and the Theaetetus (see above,
with n. 7), but does not expound on it either.
Introduction 9

In the following examination of how Plato negotiates the issue of narrative

voice, I shall use the conceptual and analytical apparatus of modern literary
theory.

3 Plato as Literary Author

Together with tragedy and comedy, dialogue is a distinctive contribution of

classical Athens to the set of literary genres at our disposal. Its literariness was
recognized as early as Aristotle. Just a few decades after dialogue had appeared
on Greek literary scene, Aristotle wrote in the Poetics:

Yet another form of mimesis uses speech (logoi) alone, either in prose
or in verse … , but until now it has been without a name. For there is no
common name we could apply, on the one hand, to the mimes of Soph-
ron and Xenarchus and the Socratic dialogues (hoi Sōkratikoi logoi) and,
on the other, to mimetic poetry one would compose in iambic, elegiac, or
any other meter.21

Later in the same passage, Aristotle dismisses the habit of labeling as “poetry”
compositions on medicine or natural philosophy only because they happen
to be put in verse. He concludes: “and yet Homer and Empedocles have noth-
ing in common but the meter, so that it would be right to call the one poet
(poiētēs), the other natural philosopher (phusiologos) rather than poet.”22
Aristotle’s assertion that a philosophical dialogue should be considered “po-
etry,” whereas a philosophical poem should not, may well strike us as peculiar.
We should, however, take into account that, when discussing “poetry” (poiēsis,
“making”), Aristotle only had the mimetic variety in mind.23 Therefore we
have good reason to assume that to all intents and purposes Aristotle’s “­ poetry”

21 Arist. Poet. 1447a28–b13; tr. S.H. Butcher, slighly adapted. The same or a closely similar
view had been expressed by Aristotle in the early dialogue On Poets, see Arist. fr. 72 Rose:
logous kai mimēseis (“pieces of discourse and of mimesis”), relating to the mimes of Soph-
ron and the first Socratic dialogues. See further Haslam 1972; Clay 1994, 34 n.22; Ruther-
ford 1995, 15, 44; Ford 2010.
22 Arist. Poet. 1447b17–20. See also Poet. 1451b2–3: “The work of Herodotus might be put into
verse, and it would still be a species of history, with meter no less than without it.”
23 See esp. Poet. 1447b14–15 “as if [the poets] are not called poets by virtue of mimesis.” It is
symptomatic that Aristotle ignores the entire genre of lyric poetry and hardly even men-
tions the choral odes of tragedy, obviously on account of their non-mimetic character. See
further Finkelberg 1998: 10–11, and 2014: 159.
10 Chapter 1

amounted to what we today call “fiction.”24 The Socratic dialogue, with its mi-
metic narrative voice (above, with n. 6), also belongs to this category. This is
not to say that it is in every respect commensurate with other forms of fiction.
The peculiarity of the philosophical dialogue as a literary genre is that the
subject matter that it has to arrange into a “plot” consists of arguments rather
than events: it is the arguments that form the dialogue’s substance – its “story,”
as it were.25 Consider for example the Symposium. Apollodorus retells Aris-
todemus’ report about how, years ago, he accompanied Socrates to Agathon’s
house and joined the party being held there. Yet Agathon’s party, for all its ups
and downs, is not the “story” of the Symposium: it is a framework for the array
of arguments communicated in the participants’ speeches. It is by means of
Plato’s skillful handling of this framework that the dialogue’s “story” is so struc-
tured as to result in a piece of plotted narrative.26
Not all Plato’s dialogues are as lavishly elaborated as the Symposium, but this
does not mean that a dialogue whose literary elaboration is minimal will pres-
ent its arguments in a pure and unalloyed form. The mere presence of Socrates
and at least one interlocutor, the bare mention of the spatiotemporal setting
which Socrates has come to be associated with, are enough effectively to com-
municate to the reader that a given piece of prose fiction should be identified
as belonging to the genre of Socratic dialogue. “Are you, Alcibiades, on your
way to offer a prayer to the god?” “Yes, Socrates, I am.” Never mind that Alcibi-
ades ii, whose opening words I have just quoted, is apparently not authored
by Plato: the minimal conditions are fulfilled, and just as with fan fiction the

24 See Laborderie 1978, 53–55; Clay 1994, 23–24; Finkelberg 2014, 158–59. Cf. Finkelberg 1998:
195: “[C]onsistent application of the criterion of mimesis would mean admitting into the
realm of poetry the novels of, say, Tolstoy, but excluding from it the poems of, say, Hölder-
lin.” Cf. also McKeon 2002: 30, on “Aristotle’s revolutionary simple abstraction ‘poetry.’”
25 Here and elsewhere, I use the terms “story” and “plot” in the sense they have acquired
in classical anglophone criticism, see e.g. Forster 1972 [1927] or Brooks and Warren 1959
[1943]. Cf. Baldick 2001, s.v. “plot”: “the plot is the selected version of events as presented
to the reader or audience in a certain order and duration, whereas the story is the full
sequence of events as we imagine them to have taken place in their ‘natural’ order and
duration. The story, then, is the hypothetical ‘raw material’ of events which we recon-
struct from the finished product of the plot.” On alternative designations of this kind of
relationship, such as “fabula” vs. “sujet” or “histoire” vs. “discours,” see, e.g., Scheffel 2013:
3.1–3.4.
26 Cf. Kosman 1992: 84: “They [Platonic dialogues] are philosophical dramas in the sense
that the action that takes place, that is represented mimetically, is philosophical argu-
ment,” and Schur 2014: 102: “the main events in Plato’s dialogues are acts of speech and
argument.” Plato’s plotting of the dialogues is discussed in detail in Chapter 6.
Introduction 11

reader is immediately plunged into the fictional universe of Socratic dialogue

and subscribes to its conventions.27
Hence it would be incorrect to assume, as Michael Frede did, that “[a]s a
rule it is an argument which forms the backbone of the dialogue and gives it its
structure.”28 The only group of dialogues to which this assumption may apply
are short elenctic dialogues coextensive with a single argument.29 But even the
elenctic Meno contains no fewer than four interrelated arguments: the open-
ing discussion of the definition of virtue is followed by a demonstration of the
theory of recollection, which is in turn followed by a discussion on whether
virtue can be taught, and finally by another, on knowledge and true opinion.30
The arguments presented in the Meno do not determine the structure of the
dialogue – rather, the structure of the dialogue as articulated in its conversa-
tional dynamics determines the presentation of the arguments.
A standard Platonic dialogue normally comprises more than one argument –
some interrelated, others blatantly heterogeneous. The more complex or var-
ied the assemblage of arguments, the more intensive strategies of emplotment
will be employed. The arguments may be progressive or circular – as expressed,
for example, in the following passage: “And then we got into a labyrinth, and
when we thought we were at the end, came out again at the beginning of the
search, having still to seek as much as ever.”31 They may contain more than one
stage, or several arguments may compete in the same dialogue. To give a few
examples, in the Protagoras the initial discussion on whether virtue can be
taught is followed by three more: – on how virtues are related, on Simonides’
poem on virtue, and on pleasure and the good – each presenting a different
argument. The heterogeneity of arguments paraded in the Gorgias, the Sym-
posium, and the Phaedrus is too well known to be dwelt on here. The Soph-
ist encompasses two main arguments loosely connected – the search for the
definition of the sophist and the inquiry into the nature of not-being, whereas
the Statesman places under one cover the discussion of method and the dis-
cussion of politics. Even the Theaetetus with its ostensibly single argument
comprises both the exposition of Socrates’ maieutic method and the famous

27 Cf. Iser 1993: 12, on the contractual terms involved in the reading of literary texts: “Among
the most obvious and the most durable of such signals are literary genres, which have
permitted a wide variety of contractual terms between author and reader.”
28 M. Frede 1992: 202.
29 The Ion, the Euthyphro, the Hippias Major, the Hippias Minor, the Laches, the Charmides,
the Lysis. On the structure of the Socratic elenchus see Vlastos 1983, M. Frede 1992: 209–12.
30 On the structure of the Meno see, e.g., Blössner 2011, and below, pp. 59–60, 127–28.
31 Euthyd. 291b7–c2. “Going round and round” is one of Plato’s favorite argumentative strate-
gies, see e.g. Euthyph. 15b; Grg. 517e; Meno 74a; Tht. 200ab, 209de; Plt. 286e.
12 Chapter 1

Digression, whose relevance to the dialogue’s conceptual framework has been

debated since antiquity.32
What holds the arguments together and consolidates them into one whole
is the overarching design of the dialogue, which we have every reason to call
its plot.33 This does not mean that what I refer to as the dialogues’ plot can-
not be approached as an argument in its own right. But this is not the kind
of argument that is cast in words. In the Symposium, the sublime last para-
graphs of Diotima’s speech are followed by a “satyr drama” sparked by Alcibi-
ades’ arrival;34 in the Phaedrus, the climactic Palinode is followed by a long
discussion on rhetoric and the futility of writing; the Gorgias, the Phaedo, and
the Republic end with eschatological myths, which invite us to view the dia-
logues’ content from a different angle. It is above all this juxtaposition of parts
that turns the design of Plato’s dialogues into a communication tool in its own
right.35 We may of course call this juxtaposition an argument, but it would be
the same kind of argument that transpires, for example, from the design of the
Iliad or Vanity Fair. As G.R.F. Ferrari put it,

A reader of the dialogues could derive Plato’s philosophic views from

the interaction of the characters, from the flow and the outcome of the
discussion, could take action and characterization into account, could
do all this without ever attributing directly to Plato the arguments of a
particular spokesman and still assume that Plato’s intention, no less than
George Eliot’s, was to make his views apparent and to persuade his read-
ers of their validity.36

32 On the Gorgias and the Phaedrus see Chapters 3 and 6; on the Sophist and the Statesman
see Chapter 3; on the Symposium and the Theaetetus see Chapters 4 and 6.
33 Cf. Bloom 1991: xx: “Every argument must be interpreted dramatically, for every argument
is incomplete in itself and only the context can supply the missing links. And every dra-
matic detail must be interpreted philosophically, because these details contain the im-
ages of the problems which complete the arguments. Separately these two aspects are
meaningless; together they are an invitation to the philosophic quest.”
34 Symp. 222d3-4, Socrates to Alcibiades: “This satyric and indeed silenic drama of yours
has become manifest (τὸ σατυρικόν σου δρᾶμα τοῦτο καὶ σιληνικὸν κατάδηλον ἐγένετο).” See
further Clay 1983: 194; Rutherford 1995: 204–5; Sheffield 2001.
35 See also below, pp. 101–102, on the juxtaposition of the dramatic frame and the framed
narrative in the multi-level dialogues.
36 Ferrari 2000. Cf. Griswold 2002: 85: “But the fact of his anonymity as author means that
‘Plato’s meaning’ is not ascertainable in the way that, say, ‘Kant’s meaning’ may be ascer-
tainable in the Critique of Pure Reason.” For a suggestive parallel between Plato’s dialogues
and the classical novel see Murley 1955: 287; for a spirited defence of the literary approach
to Plato see Schur 2014.
Introduction 13

This is why we can only do justice to both the form and the content of Plato’s
dialogues if we approach them as works of literature.

4 Narrative Voice in Plato’s Dialogues

As we have seen, the mimetic character of the Socratic dialogue was suffi-
cient for Aristotle to classify it as a literary genre. The classification as such
still stands,37 and the outstanding literary quality of Plato’s dialogues, the most
prominent representatives of the genre, is universally recognized. At the same
time, the place Plato’s dialogues occupy in the history of literature cannot even
remotely be likened to their place in the history of philosophy. It would be no
exaggeration to say that, outside the field of classical studies, they are hardly
visible as works of literary fiction. Let me illustrate my point.
With the advent of narratology, much attention has been paid to such liter-
ary strategies as multiple narrative levels, frame narrative, embedded narrative,
frame-breaking (metalepsis), and more. Side by side with the study of the ways
these strategies have been applied in modern and postmodern fiction, literary
scholars frequently refer to their pre-modern and early modern antecedents.
The canonical list of the latter consists of The Thousand and One Nights, The
Canterbury Tales, and The Decameron – as if the Symposium and the Phaedo,
the Parmenides and the Theaetetus, all of them highly sophisticated framed
narratives with multiple narrative levels and skillful use of metalepsis, had
never been written. The Wikipedia entry “Frame Story” enhances the standard
list by adding the ancient Egyptian Papyrus Westkar and the Mahabharata.
Plato’s dialogues, again, are conspicuous by their absence.
A notable exception is the founder of narratology Gérard Genette who, as
we saw above, uses the Theaetetus’ switch from the narrative to the dramatic
mode as a prototype of pseudo-diegetic narrative.38 Yet, this is Genette’s sole
reference to Plato’s literary practice. To the best of my knowledge, the only
systematic narratological approach to Plato’s dialogues is Kathryn Morgan’s
contribution to a volume purporting to apply the methods of narratology to
ancient Greek literature, all of it written by classicists.39

37 See, e.g., Baldick 2001, s.v. “dialogue”: “a literary form in prose or verse based on a debate or
discussion, usually between two speakers.” On the “constitutional fictionality” of Plato’s
dialogues see Cohn 2001: 493.
38 Cf. de Jong 2014a: 17 and 2014b: 3.2.1, and above, Section 2.
39 Morgan 2004; only the narrated dialogues and the narrated parts of the mixed dialogues
are discussed. Ferrari 2010, makes effective use of the narratological apparatus in his
subtle analysis of the way the narrator is treated in the Republic; Collins 2012 and 2015:
14 Chapter 1

In classical studies proper, attending to literary aspects of Plato’s dialogues

is of course hardly new. Over the years, not a few readers of Plato, students
of philosophy included, have proved acutely aware of his literary strategies
and their relevance to the philosophical argument.40 Recent decades espe-
cially, have seen an unprecedented surge of interest in Plato’s literariness.
Reading Plato’s dialogues against the background of Attic drama has become
particularly widespread. “Throughout his dialogues, Plato writes with the the-
ater in mind”: – this assertion by the author of a recent book neatly reflects the
drama-oriented turn in the interpretation of Plato’s dialogues.41 The attention
to Plato’s dramatic antecedents is, however, only one aspect of a more general
phenomenon. Since the 1990s, monographs, articles, collections of essays, and
conference proceedings have sparked a stimulating dialogue between philoso-
phy and literature, illuminating such issues as the genre of Socratic dialogue,
Plato’s literary art, his practices of characterization, possible literary influences
on his work, and more.42
At the same time, although several recent studies do concentrate on the
figure of the narrator, first and foremost Socrates, in Plato’s narrated and mixed
dialogues,43 no comprehensive study of narrative voice in Plato’s corpus as a
whole is to be found. The reason is obvious: it is tacitly assumed that the dra-
matic dialogues, which constitute most of the corpus, do not lend themselves
to being approached in terms of narrative. As we have seen, however, this does
not agree with how Plato perceived dramatic texts, including those he himself
authored. As Republic 3 and the Theaetetus, discussed in the first two sections
of this chapter, make clear, Plato viewed even the discourse that is presented
in dramatic form as a variety of narrative — a narrative whose narrator, though

45–166, applies narratalogical categories, including metalepsis, to the Euthydemus and

the Phaedo; Collobert 2013 proposes a taxonomy of narrative modes and narrators in Pla-
to’s dialogues; Kaklamanou and Pavlou 2016 use narratalogical methods in approaching
the prologue of the Theaetetus. See also Halliwell 2009, for what I see as an unjustifiably
negative assessment of the applicability of narratology to Plato’s theory and practice.
40 See esp. Taylor 1960 [1926]; Friedländer 1958–69 [1928–1930]; Bloom 1991 [1968]; Guthrie
1975 and 1978; Laborderie 1978.
41 Puchner 2010: 5. On Plato’s affiliation with dramatic poetry see esp. Nussbaum 2001[1986],
122–35; Heath 1989: 12–27; Arieti 1991; Nightingale 1995: 60–92, esp. 67–69; Blondell 2002,
1–29; Finkelberg 2006: 67; Charalabopoulos 2012, but see already D. Tarrant 1955.
42 General: Rutherford 1995, Kahn 1996, Griswold 2002, Michelini 2003, Erler 2006, Rowe
2007, McCabe 2008, Capuccino 2014, Schur 2014, Fritz 2016, but see already Thesleff 1967,
Laborderie 1978; genre: Clay 1994, Kahn 1996: 1–35, Ford 2010, Rossetti 2011 [2003], cf.
Laborderie 1978: 43–51; characterization: Blondell 2002, Cotton 2014; literary influences:
Clay 1994, Nightingale 1995, Ford 2008, Capra 2014, cf. Laborderie 1978: 13–43.
43 Morgan 2004; Bowery 2007; Ferrari 2010; Schultz 2013; but see already Murley 1955.
Introduction 15

suppressed, is nevertheless implicitly present in the text. However, Plato’s con-

cept of narrative voice that transpires from Republic 3 and the Theaetetus has
never been tested against his own literary practice. The present book is an at-
tempt to fill this gap.
The book falls into two parts. Part One offers a close reading of the narra-
tor’s text and its equivalents in Plato’s narrated, dramatic, and mixed dialogues.
I shall start with an analysis of the narrated dialogues, the only group whose
narrator is explicit all along (Chapter 2). Though not numerous, the narrated
dialogues are instrumental in providing us with a reliable set of criteria which
allow us to map the way Plato negotiates the figure of the narrator – his func-
tions, his manifestations, the scope of his authority. The analysis leads to the
conclusion that Plato’s narrator is thoroughly defined in terms of perception
and knowledge. The narrator is he who focalizes the story in its entirety to
the exclusion of any other character; as a result, the dialogue is permanently
filtered through the narrator’s individual perception.
The results obtained for the narrated dialogues serve as the basis for ap-
proaching the dramatic dialogues, the largest group in the corpus (Chapter 3).
Applying to the dramatic dialogues the same criteria whereby the narrator of
the narrated dialogues has been identified shows that, except for three dia-
logues (the Euthyphro, the Crito, and the Menexenus), all Plato’s dramatic dia-
logues exhibit a single focus of perception regularly associated with one of the
characters. This allows us to suggest that these dramatic dialogues are con-
ceived as narratives whose narrator is implicitly present in the text. Like the ex-
plicit narrator of the narrated dialogues, the implicit narrator of the dramatic
dialogues is construed as the first-person narrator who focalizes the dialogue
in its entirety.
The results of the analysis of both the narrated and the dramatic dialogues
will be further applied to the mixed dialogues, the most structurally complex
group in the corpus (Chapter 4). Since the mixed dialogues share their charac-
teristics with both the dramatic and the narrated ones, the principles on whose
basis these two groups have been approached will be applied to the respective
parts of the mixed dialogues as well. The mixed dialogues’ multi-level structure
will also be analyzed in detail; special emphasis will be placed on Plato’s strat-
egies of metalepsis, or breaking the boundaries between different narrative
levels.
Part Two is devoted to synthesis and interpretation. In Chapter 5, I shall try
to identify Plato’s major contributions to the poetics of narrative voice and
to map their diachrony. In a sense, Plato’s entire literary career can be pre-
sented as a series of attempts to replace the literary form of narrated dialogue
(which he apparently inherited from his predecessors) with that of dramatic
16 Chapter 1

dialogue (which he apparently introduced himself) – yet without abandoning

the latter’s identification as a form of narrative. It seems to follow that Plato
strived for his dialogues to behave like narratives without looking like them. To
achieve this goal he eventually abandoned the explicit narrator, but kept the
signals of his authority intact. It was obviously important to him that his dia-
logues be filtered through a single focus of perception. This key role has been
delegated by him to the implicit narrator.
In more than one sense, the two concluding chapters stand apart from the
rest, in that they depart from the discussion of narrative voice in the strict
sense of the word. In Chapter 6, I shall try to outline the scope of the author-
ity at the narrator’s disposal by following material manifestations of narrative
voice and delineating their role in the dialogues’ structure. This leads to the
conclusion that, although the authority vested in Plato’s narrator is far from
negligible, it does not spread onto the plot of the dialogue. The latter is crafted
by the author, whose authority lies beyond the narrator’s domain. To illustrate
how the narrator’s text behaves in the context of the dialogue as a whole, three
dialogues (the Lysis, the Gorgias, and the Phaedo) are analyzed in detail.
Finally, proceeding from the renewed discussion of mimesis introduced in
Book 10 of the Republic, I shall try to define the narrator’s place in the fictional
universe created by the author and the nature of the relationship between the
two (Chapter 7). The author fashions a plotted narrative out of the arguments
he purports to exhibit: he is the demiurge of a fictional universe. Yet, the reader
can only enter this universe through the narrator’s mediation. The narrator is
the dialogue’s gatekeeper, a filter through whose agency it is communicated to
the reader. His main function is to control the way the dialogue is presented to
the reader by sustaining a certain perspective of it. There is little room to doubt
whose perspective it ultimately is.

5 Plato’s Narrator and Narrative Theory:

Some Necessary Adjustments

Throughout this book I approach the issue of narrative voice in Plato’s dia-
logues from the standpoint of classical and post-classical narratology, with spe-
cial emphasis on such topics as narrative levels, focalization, narrative frame,
and metalepsis. Let me start by clarifying the principal concepts involved.
I use the term extradiegetic to designate the narrator and narrative situa-
tion located outside the narrated world, and the term intradiegetic, or diegetic,
to designate the narrator and narrative situation located within the narrated
Introduction 17

world.44 To designate the second- and third-degree narratives located below

the first intradiegetic level, I use the terms “second intradiegetic level,” “third
intradiegetic level,” and so forth, thus replacing, for clarity, both Genette’s
“metadiegetic” and the later “hypodiegetic” alike.45
Discussing the narrator’s relation to the narrated story, I use the term hetero-
diegetic to designate a narrator who is not a character in the story, and the term
homodiegetic to designate a narrator who is one of the story’s characters.46
Finally, I use the terms autodiegetic narrator or alternatively narrator-hero to
designate such a homodiegetic narrator who is at the same time the story’s
main character, and the term narrator-observer to designate such a homodi-
egetic narrator who, although present in the story, is not its main character.47
At the same time, when applied to Plato’s dialogues, some narratological
concepts prove more applicable than others, and not a few require further ad-
justment. The following points seem to be of special relevance to the present
discussion.

5.1 Narrative Levels and Homodiegetic Narrator

The narratee, namely, the reader or listener to whom the narrative is addressed,
is by definition placed outside the world of the narrated story; hence the lev-
el at which the narrative act is positioned will always lie beyond the story’s
boundaries. As John Pier put it, “every narrative, beginning with the first-level
narrative, is produced by an act of narration which is of necessity external to
that level.”48
In the case of the primary narrator, the level at which both the narrator and
the narratee are positioned will be external to the narrative as a whole. In other
words, the primary narrator and the narrative act in which he or she is involved
are extradiegetic by definition. In Genette’s words,

44 Genette 1980: 228–29.

45 Genette 1980: 228–29; Bal 1981: 41–43; cf. Baldick 2001: s.v. “diegesis.”
46 Genette: 1980: 244–45. The alternative terms “external narrator” and “internal, or character-
bound, narrator” are also being used in scholarly literature; see Bal 1997 [1985]: 21–23; de
Jong 2014a: 19. In the amendment of Genette’s taxonomy proposed by Wolf Schmid, Gen-
ette’s “extradiegetic-heterodeigetic narrator” would correspond to “primary non-diegetic
narrator,” the “extradiegetic-homodiegetic narrator” to “primary diegetic narrator,” the
“intradiegetic-heterodiegetic narrator” to “secondary non-diegetic narrator,” and so on.
See Schmid 2008: 89; cf. Pier 2014: 3.1.1.
47 Genette 1980: 245, 248.
48 Pier 2014: 2.
18 Chapter 1

M. de Renoncourt and Crusoe are authors-narrators, and as such they are

at the same narrative level as their public – that is, as you and me.49

Furthermore, while the identity of the primary narrator may or may not be
made known to the reader, the identity of his or her narratee and of the narra-
tive situation in which the primary narrator and narratee are involved is nor-
mally not made manifest. To quote Genette again,

The extradiegetic narrator, on the other hand, can aim only at an extradi-
egetic narratee, who merges with the implied reader and with whom each
real reader can identify. This implied reader is in principle undefined.50

However, in multi-level narratives we also encounter narrators who tell their

stories from within the narrative: while they too are positioned at a different
level than that of the stories they deliver, such narrators will be fully enclosed
by the fictional universe (intradiegetic narrator). Accordingly, their intradi-
egetic narratees will not merge with the implied reader or listener:

To an intradiegetic narrator corresponds an intradiegetic narratee…

We, the readers, cannot identify ourselves with those fictive narratees
anymore than those intradiegetic narrators can address themselves to us,
or even assume our existence.51

Now, since the only model of narrator that Plato employs is that of the first-
person narrator who is also one of the characters in the story (homodiegetic
narrator), Plato’s narrator, both the extradiegetic and the intradiegetic, finds
himself in an ambivalent position. Since the extradiegetic first-person narrator
addresses the dialogue’s implied reader, he is positioned outside the narrated
world; on the other hand, since the first-person narrator – or, more precisely, his
earlier self – is also one of the characters in the story, he is positioned within the
narrated world:

M. de Renoncourt as the “author” of the Mémoirs is extradiegetic: a­ lthough

fictive, he addresses the actual public, just like Rousseau or Michelet; the
same Marquis as hero of the same Mémoirs is diegetic, or intradiegetic …52

49 Genette 1980: 229; see also Schmid 2008: 86–95. Marquis de Renoncourt is the fictional nar-
rator of the seven-volume Mémoires et aventures d’un homme de qualité by Abbé Prévost, the
final volume of which contains the story of the Chevalier des Grieux and Manon Lescaut.
50 Genette 1980: 260.
51 Genette 1980: 259–60.
52 Genette 1980: 229.
Introduction 19

By the same token, the intradiegetic first-person narrator will belong to two
intradiegetic levels: that of the narrative act in which he is involved and that
of the narrative which he delivers and which features his former self as one of
the characters. This double position, aptly conveyed through Wolf Schmid’s
distinction between “the narrating self” and “the narrated self,”53 accounts for
every single narrative situation in which Plato’s narrator is involved.

5.2 The First-Person Narrator and Focalization

As I argue in this book, owing to his ambivalent position as outlined above, the
first-person narrator of Plato’s dialogues adopts the point of view of his former
self as a character in the story: that is, he both narrates and focalizes the story.54
Such merging of the homodiegetic narrator and the source of focalization
would be at variance with the rigorous separation between narration and fo-
calization pursued in classical narratology.55 However, in that it eliminates the
difference between the omniscient third-person narrator and his first-person
counterpart,56 the separation in question proves difficult in more than one re-
spect. Since knowledge of the first-person narrator is necessarily limited by the
personal experience acquired by his former self, his relation to focalization can
hardly be identical to that of the omniscient narrator.
At the same time, the merging of first-person narration and focalization
squares very well indeed with the re-assessment of the focalizing role of the
first-person narrator that took place in post-Genettian theory. To quote Wil-
liam F. Edmiston,

The fpn [first-person narrator] may efface all temporal distance and
adopt the intradiegetic vision of the hero, who presents his own mental
activity and his view of others at the moment of event.57

53 Schmid 2008: 86–87: “erzählendes Ich” and “erzähltes Ich.”

54 “Focalization, a term introduced by Genette (1972), may be defined as a selection or re-
striction of narrative information in relation to the experience and knowledge of the nar-
rator, the characters or other, more hypothetical entities in the storyworld” (Niederhoff
2011a: 1.2). The introduction of the term “focalization” was purported to replace such rela-
tively vague categories as “perspective” and “point of view.”
55 As epitomized, e.g., in Baldick 2001: s.v. “focalization”: “The nature of a given narrative’s
focalization is to be distinguished from the narrative ‘voice,’ as seeing is from speaking.”
56 As made explicit, e.g., in Genette 1980 [1972]: 194; Bal 1997 [1985]: 158; Rimmon-Kenan
2002 [1983]: 75. For a criticism see Shen 2003.
57 Edmiston 1989: 742. Cf. ibid. 742–43: “When the fpn speaks about past events and char-
acters from his present vantage point, he is temporarily and spatially external to his story.
He knows more now than he did then, and he is less involved now than he was then. He is
perceptually limited nonetheless because he continues to be a part of the same world in
which he lived as the hero.”
20 Chapter 1

Ansgar Nünning elaborates on the relation between “the narrating I” and “the
narrated I” in first-person narratives:

First-person narratives usually involve two main perspectives: the

narrator-perspective represents the private domain of the homodiegetic
narrator, whose worldview is conditioned by the knowledge, psychologi-
cal dispositions, attitudes, and values of the narrating “I.” Telling the story
in retrospect and knowing its outcome, the narrating I has a wider per-
spective than the narrated or experiencing I, who previously inhabited
the time and space of the story level. Since the first-person narrator re-
members what the narrated I knew, experienced, thought, and felt, the
character-perspective of the experiencing I is embedded in the cognitive
domain of the narrating I.58

As we shall see, this assessment will adequately account for the position of
narrator in Plato’s dialogues.

5.3 Zooming-in
One of Plato’s principal strategies of focalization is to open a dialogue with
the narrator’s gradually focusing his attention on a certain, not yet sufficiently
identified, object, person or theme, and acquiring fuller knowledge of it at the
end of the process. Such gradual focalization is nothing other than “zooming-
in,” a narrative technique attested as early as Homer. This is how Irene de Jong
describes its being applied at Odyssey 9.166–223, Odysseus and his men focal-
izing the cave of the Cyclops:

…we gradually get closer to the Cyclops’ cave, following the pace of the
Greeks’ focalization. First, standing on Goat-Island, they see smoke and
hear sheep and goat bleating (166–7). Then, from the coast where they
have landed, they see the outside of the cave and the yard in front of
it (182–92. …). Finally, they arrive at the cave and admire its interior
(216–23).59

58 Nünning 2001: 218. Cf. Niederhoff 2011b: 3.2 [28]: “After all, it [Genette’s separation of
narrators and focalizations] makes sense only if narrators and perspectives are distinct
categories, in other words if the choice of a particular kind of narrator does not entail a
particular perspective.” See also Phelan 2001; Fludernik 2001a, 2001b.
59 De Jong 2001: 235. Cf. de Jong 2001: 139, on Odysseus approaching the Land of the Phaea-
cians. For an overview see de Jong 2001: 70.
Introduction 21

The zooming-in that de Jong describes is instrumental in the building of sus-

pense toward the meeting with the Cyclops, which culminates in the latter’s
arrival (233–51).
As far as I can see, the closest Platonic parallel to Homer’s episode would
be the Euthydemus description of Crito’s two-stage focalization, first from afar
and then through a first-hand report, of Socrates’ encounter with the soph-
ists Euthydemus and Dionysodorus (below, p. 85). The dialogue, however, is
a literary genre committed to the dynamics of conversation, so that in Plato’s
dialogues zooming-in normally proceeds as a series of questions posed by the
narrator, thus taking the form of gradual absorption of information by both
the narrator and the reader. Such gradual absorption of information may be
further compared to what Genette refers to as “initial ‘ignorance,’” which “has
become a topos of novellistic beginning.”60

5.4 Narrative Frame and Frame Narrator

The Platonic corpus has six multi-level dialogues, each containing at least one
frame story and therefore featuring more than one narrator.61 Thus, the primary
narrator of the Parmenides, Cephalus, delivers a story from which he himself is
absent; only when we arrive at the third intradiegetic level does ­Pythodorus, a
narrator who is also a character in the story, emerge (Chapter 2). At the same
time, the primary narrator delivers a story of his own, which is the dialogue’s
frame story. Genette’s taxonomy treats the two kinds of narrator as equal in sta-
tus, so that the same criteria are applied to both. But it would be counter-intui-
tive to give equal weight to the narrative of Cephalus, who relates how he met the
prospective narrator of the story about Socrates’ encounter with Parmenides,
and the narrative of Pythodorus, who tells the story itself. Although formally
equivalent in that each represents a separate narrative level, the significance of
the frame is not equivalent to the significance of the narrative it encloses.
As Monika Fludernik has pointed out, classical narratology makes no provi-
sion for narrative situations that involve frame narratives:

Unlike the story-within-the-story, the specific structure of frame narra-

tive, on the other hand, is never properly caught in these mathematical
outlines. Although, technically speaking, a similar relation of embedding
obtains, the constitution of stereotypical frame narrative needs to be

60 Genette 1980: 191 n. 53. Genette brings the openings of Flaubert’s L’Éducation sentimental
and Zola’s Germinal as representative examples. He claims that the rule no longer oper-
ates in Henry James, but see the first chapter of The Portrait of a Lady, where the identity
of the characters participating in the opening scene is only gradually disclosed.
61 The Protagoras, the Euthydemus, the Phaedo, the Symposium, the Parmenides, the
Theaetetus.
22 Chapter 1

connected with a proportional factor relating to frame’s length and sig-

nificance. However structurally it is the case that both in a frame narra-
tive and in an embedded narrative a character tells a story, usually in the
setting of a frame narrative the framing primary story is quite marginal in
relation to the embedded story, which takes up most of the text.62

To distinguish these two categories of narrator, I shall refer to the narrator who
appears in the dialogue’s frame story as “frame narrator.”

5.5 Metalepsis
Metalepsis, or frame-breaking, is a narratological concept purporting to ac-
count for the literary strategy of transgressing the boundaries between differ-
ent narrative levels. It was first identified by Genette, who defined it as “any
intrusion by the extradiegetic narrator or narratee into the diegetic universe
(or by diegetic characters into a metadiegetic universe, etc.), or the inverse.”63
Owing to the pioneering studies of Irene de Jong and Tim Whitmarsh, the
use of metalepsis in Greco-Roman literature has recently drawn considerable
scholarly attention.64 Surprisingly therefore the application of this narrative
strategy in Plato’s dialogues has largely passed unnoticed. This seems to be
mainly due to the way metalepsis is employed by Plato.
First, the form of metalepsis most widespread in contemporary fiction,
hence most frequently discussed, concerns the breaking of artistic illusion is-
suing from intrusion of the real world into the fictional one or vice versa; in
narratological terms, this would amount to breaking the boundary between
the extradiegetic and the intradiegetic level of narrative. In equal measure,
however, metalepsis can occur between different intradiegetic levels, that is,
between the narrative frame and the story it encloses. As it happens, this is
precisely the type of metalepsis that Plato favored.65

62 Fludernik 1996: 257. Fludernik’s conclusion is that “the situation is methodologically in-
tractable, at least within a structuralist methodology” (ibid.).
63 Genette 1980: 234–35; on “metadiegetic” see above, with n. 45. See also Genette 2004;
Cohn 2005; Ryan 2006: 204–11; Pier 2013; Hanebeck 2017. For a discussion of various types
of metalepsis see Fludernik 2003: 382–89; for examples of metalepsis in Homer and other
early poetry see de Jong 2009 and 2014a: 41–42; for metalepsis in Plato see Collins 2012:
163–73; 2015: 124–33; Finkelberg, forthcoming, and below, Chapter 5.
64 De Jong 2009; Whitmarsh 2013. A conference on Metalepsis in Classical Literature took
place in Oxford in September 2015; as far as I know, metalepsis in Plato was not among the
subjects discussed.
65 Metalepsis that involves the extradiegetic level occurs only in two dialogues – the
Charmides and the Republic, both of them having only one intradiegetic level (below,
pp. 34, 37).
Introduction 23

Second, in most cases metalepsis in Plato does not affect the plot level: what
we have instead is so-called rhetorical metalepsis.66 From apostrophic ad-
dresses to the audience in the Charmides, the Protagoras and the Phaedo (be-
low, pp. 33–34, 82–83, 89n33) to short asides in the Republic, the Symposium,
and the Protagoras and the Phaedo again (37, 95, 82–83, 89–90), rather more
often than not metalepsis in Plato (to borrow Marie-Laure Ryan’s characteriza-
tion of rhetorical metalepsis)

opens a small window that allows a quick glance across levels, but the
window closes after a few sentences, and the operation ends up reassert-
ing the existence of boundaries.67

Third, in the multi-level Symposium and Parmenides Plato’s use of a two-tier

system of reported speech results in the frame narrator permanently intruding
into the narrated world (below, pp. 43, 96). Although no frame-breaking in the
strict sense of the word is involved, the ensuing intermingling of different nar-
rative levels is not unlike that of “metaleptic contamination” characteristic of
ontological metalepsis (metalepsis at the plot level) (cf. Chapter 4, with n. 24).
Unattainable by modern languages and lost in translation, this metaleptic strat-
egy is without parallel in modern fiction and has passed unnoticed as a result.68
It is only in the Euthydemus and the Phaedo that metalepsis acts as an insep-
arable part of the plot. As far as I can see, both cases closely parallel metaleptic
practices of modern fiction (cf. below, pp. 114–15); still, J.J. Collins (n. 63 above)
seems to have been the only one to acknowledge the presence of metalepsis in
these two dialogues. The role of metalepsis in the Euthydemus and the Phaedo
is discussed in detail in Chapter 4.

5.6 Emplotment
As John Pier has pointed out, the absence of the Aristotelian principle of plot
is a distinctive feature of Genette’s narratalogical model, which is focused on
modality and voice:

As his [Genette’s] definitions of “story” and “narrative” show, the assimi-

lation of narrative content and narrative text into concepts of structural

66 To be distingished from ontological metalepsis, see further Cohn 2005; Ryan 2006: 204–11;
Pier 2013: 2.1 [7].
67 Ryan 2006: 207.
68 This, of course, does not mean that the effect as such has not been described or com-
mented on; see, e.g., the comments of Halperin and Hunter adduced in Chapter 4 (with
n. 50).
24 Chapter 1

linguistics shifts the focus away from the notion of plot. … Nowhere in
Narrative Discourse, however, is there a mention of “plot,” or mûthos.69

We shall see, however, that the question of who, the narrator or the author,
is envisaged as responsible for the dialogues’ plot is essential in assessing the
narrator’s role in Plato’s dialogues. That is why in Chapter 6, which deals with
the material manifestations of narrative voice and their plot-building func-
tion, I turn to the strategies of emplotment as evinced in the work of Peter
Brooks and Paul Ricouer. The assumption from which I proceed is that, by fol-
lowing what Ricoeur defined as “mise en intrigue” (“emplotment”) and Brooks
as “plotting,” or “the moments where we seize the active work of structuring
revealed or dramatized in the text,”70 one can obtain a deeper insight into the
dialogues’ design and the response they were intended to produce. The over-
arching approach, however, remains the same, and it can be formulated as fol-
lows: whatever methodology is used, it is above all the very fact of inquiry into
the literary strategies by means of which Plato presented his arguments that ef-
fectively bridges what is still often perceived as a gap between philosophy and
literature in Plato’s oeuvre and, as a result, allows us to approach his dialogues
as integral wholes.

69 Pier 2008: 119. Cf. also ibid.: 116–17: “In this narratology geared to ‘discourse’ rather than to
the ‘story’ (cf. Bremond’s récit racontant as opposed to récit raconté) or, in Genette’s terms,
a ‘modal narratology’ (‘analysis of narrative as a mode of representation of “stories,” ruling
out drama, for example’), narrative is considered the ‘expansion of a verb.’” On the vexed
question as to whether drama should be regarded as a subdivision of narrative see Jahn
2001; Fludernik 2008; Nünning 2008; Hühn and Sommer 2013: 3.2; Klauk and Köppe 2014.
70 Ricouer 1984 [1983]: 31–51; Brooks 1992 [1984]: 35. The two terms were introduced almost
simultaneously to give expression to the dynamic aspect of plot, cf. Brooks 1992 [1984]: 14:
“Ricoeur’s emphasis on the constructive role of plot, its active, shaping function, offers a
useful corrective to the structural narratologists’ neglect of the dynamics of narrative and
points us toward the reader’s vital role in the understanding of plot.” On emplotment in
contemporary narratology see Abbot 2011; Scheffel 2013; Kukkonen 2014.
 Part 1
The Dialogues

∵
 Chapter 2

The Explicit Narrator: Narrated Dialogues

1 Introducing the Narrated Dialogues

The Platonic corpus contains four narrated dialogues: the Charmides, the Lysis,
the Republic, the Parmenides. They form a fairly homogeneous group. Their
narrator is explicitly presented as the first-person narrator, who may or may
not be Socrates,1 and each opens with an introductory episode of two clearly
articulated stages.
– In the Charmides, Socrates narrates how, (i) having returned from the army
at Potidaea, he entered the palaestra of Taureas where he met Chaerephon
and Critias, and how (ii) they introduced to him the beautiful Charmides
on the latter’s arrival; this is followed by Socrates’ report of the conversa-
tion that developed as a result of the encounter.
– In the Lysis, Socrates narrates how (i) on his way from the Academy to the
Lyceum he encountered Ctesippus and Hippothales, and how (ii) they led
him to a newly established palaestra, where the boys Lysis and Menexenus
were introduced to him; this is followed by Socrates’ report of the conver-
sation that developed as a result of the encounter.
– In the Republic, Socrates narrates how, (i) having arrived in the Piraeus in
order to attend the festival of the goddess Bendidea, he and Glaucon ran
into Polemarchus and a group of friends and were invited to join them in
Polemarchus’ house, and how (ii) arriving there, they met other people, in-
cluding the rhetorician Thrasymachus; this is followed by Socrates’ report
of the conversation that developed as a result of the encounter.
– In the Parmenides, Cephalus of Clazomenae narrates how (i) on arriving in
Athens he and his compatriots met Glaucon and Adeimantus in the Agora,
and how (ii) these two took them to the estate of their half-brother Anti-
phon; this is followed by Antiphon’s rehearsal of Pythodorus’ story about
Socrates’ encounter with the philosophers Parmenides and Zeno.

1 Socrates also appears as the explicit narrator in the narrated parts of the mixed Protagoras
and Euthydemus; in the spurious Rival Lovers, Axiochus and Eryxias, as well as in the frag-
mentary Alcibiades and Miltiades of Aeschines. The Parmenides and the narrated parts of
the Phaedo and the Symposium are delivered by other first-person narrators. In Xenophon’s
dialogues the explicit first-person narrator is always anonymous.

© koninklijke brill nv, leiden, ���9 | doi:10.1163/9789004390027_003

28 Chapter 2

Note that the transition from one stage to the other involves not only a spa-
tial change (the Charmides being the only exception) and modification in the
cast of characters, but also results in the introduction of the principal inter-
locutor or, in the case of the Parmenides, the narrator of the main story. Char-
mides, Lysis, Thrasymachus, Antiphon – all characters whose presence makes
the ensuing conversation possible – only appear at the second stage. As we
shall see immediately, such gradual unveiling of information (“zooming-in;”
see above, pp. 20–21) is inextricably connected with the way the issue of narra-
tive voice is approached in Plato’s dialogues.
As for the closures, the Lysis and the Charmides bring the story back to the
original act of narration, more precisely, to its second stage: the tutors of Lysis
and Menexenus arrive and interrupt their conversation with Socrates; the dis-
cussion between Socrates and Charmides comes to an end. The Republic and
the Parmenides end abruptly, although at the end of its penultimate book the
Republic supplies what looks like a formal closure.2
Plato was most probably not the first to compose narrated dialogues.3 Con-
sider, for example, the following fragment from the Alcibiades, a lost dialogue
by the prominent Socratic writer Aeschines of Sphettus:

We were seated on the benches of the Lyceum, where the judges organize
the games,

or a more substantial one from the Miltiades by the same author:

It happened to be a Great Panathenaic procession, and we were seated

in the porch of Zeus the Liberator – myself, Hagnon the father of Thera-
menus, and the poet Euripides – and … Miltiades approached us, seem-
ingly on purpose.4

In both cases Socrates is the speaker, and the setting is highly reminiscent of the
openings of Plato’s Charmides, the Lysis, and the Republic. The story’s presen-
tation however is not nearly as elaborate as what we find in Plato: neither the
two-stage exposition nor the gradual intensifying of suspense, both e­ ventually

2 Cf. Annas 1981: 335: “Book 9 ends the main argument of the Republic, and ends it on a rhetori-
cal and apparently decisive note;” similarly Halliwell 1997: 325. Contra Burnyeat 1997: 288–92;
Schur 2014: 73.
3 Thesleff 1982: 199–210. Cf. Tarrant 2000: 217 n. 5: “other Socratics are likely to have anticipated
Plato in the use of ‘narrated’ dialogue.” On the argument of Aeschines’ priority see now Pen-
tassuglio 2017. On Plato and the “invention” of dialogue see also Chapter 3, Section 1.
4 ssr vi A 43 = Demetr. Eloc. 205; tr. C.H. Kahn; ssr vi A 76 = P. Oxy. 2889; my translation.
The Explicit Narrator: Narrated Dialogues 29

focusing on Socrates’ principal interlocutor (the zooming-in technique), are to

be found here. Likewise the opening of the pseudo-Platonic Eryxias:

I happened to be strolling about the Stoa of Zeus the Liberator with Eryx-
ias, from the deme Stiria, when Critias and Erasistratus … came up to us.
Erasistratus, it turned out, was just recently back from Sicily…5

A degree of sophistication more closely resembling Plato’s narratives can

be found in the opening of Xenophon’s Symposium, a dialogue that, like
all Socratic writings of Xenophon, was probably influenced by Plato and other
­Socratics.6 It is indeed difficult to avoid the impression that the way Xenophon
introduces the setting in this dialogue is modeled on the opening of Plato’s
Republic:

It was horse-race day at the Great Panathenaea. Callias son of Hipponi-

cus happened to be courting Autolycus, a boy who had just won the pan-
cration, and Callias had brought him to see the race. When it was over, he
set off for his house in the Piraeus taking both Autolycus and the boy’s
father with him, and Niceratus went with him as well. Then Callias no-
ticed Socrates, Critobulus, Hermogenes, Antisthenes and Charmides in a
group together. He arranged for someone to take the Autolycus party on
and then went across to the Socrates group himself…7

Just like Polemarchus in the Republic, Callias meets Socrates in the street and
manages to bring him and his companions to his house in the Piraeus, where
the conversation takes place. But, as distinct from Plato, whose Polemarchus
sends a slave to catch Socrates and make him wait for him, Xenophon, in what
is probably a deliberate criticism of Plato, makes his Callias approach Socrates
in person.8

5 Eryx. 392a; tr. Mark Joyal; the dialogue was considered inauthentic already in antiquity, see
Diog. Laert. 3.62. On the dubious and spurious dialogues and their position in the Platonic
corpus see Joyal 2014: 73–81.
6 Aeschines, Plato and Xenophon were approximately of the same age, yet given the circum-
stances of Xenophon’s life it is reasonable to suppose that he started writing about Socrates
considerably later than the other Socratics. cf. Kahn 1996: 29–30, Huss 1999: 15–18, but see
already Wilamowitz-Moellendorff 1920: 23. For a more nuanced approach see Danzig 2018.
7 Xen. Symp. 2–3; tr. A.J. Bowen. The opening of the Republic is quoted below in this chapter,
see pp. 34–35.
8 Similarly Huss 1999: 78; cf. Danzig 2017: 133. Callias son of Hipponicus was one of the
wealthiest people in fifth-century Athens. Friend and follower of the Sophists, he appears
as their host in Plato’s Protagoras (below, Chapter 4); he also emerges in Plato’s Apology,
30 Chapter 2

As in the Republic again, the opening consists of two stages, and the tran-
sition from one to the other involves a spatial change. However, contrary to
Plato, Xenophon does not exploit the narrative potential of the patterns he
employs. Neither the build-up of suspense nor the gradual concentration on
the main interlocutor will be found here. The same persons are present at each
stage of the exposition: as a result, the brisk narrative progression, a hallmark
of Plato’s openings, is irrevocably lost.
The two-stage openings of the pseudo-Platonic Rival Lovers and Axiochus,
though not as static as what we find in Xenophon, also fall short of the openings
of the Charmides, the Lysis, or the Republic. The Rival Lovers handles rather inef-
fectively the Charmides-inspired description of Socrates’ enthusiasm at the sight
of the beauty of the two boys he met in the class of the grammarian Dionysius
(133a). First, as distinct from Charmides of Plato’s Charmides or from Lysis and
Menexenus of the Lysis (below, Section 3), the boys of the Rival Lovers are pres-
ent on the stage from the very beginning, so that the zooming-in effect that could
be produced by their entrance is lost. Second and more important, not the two
boys but their two admirers are the focus of the dialogue, which makes Socrates’
enthusiasm somewhat misplaced: instead of drawing the reader’s attention to
the dialogue’s subject, it draws it away from it. In the Axiochus, whose opening,
as in Xenophon, involves a spatial change, no effort is made to build up narrative
suspense around Cleinias breaking the news that his father Axiochus lies on his
deathbed (364a-b) or around Socrates’ meeting with the dying man (365a-b).
We can conclude, therefore, that even if Plato was neither the first nor the
only one to write narrated dialogues, he took this genre in directions not en-
visaged by his predecessors or easy to imitate. This was recognized already in
antiquity. Thus, after mentioning other candidates for the title of inventor of
dialogue, Diogenes Laertius wrote:

In my opinion Plato, since he brought this form to the point of perfection,

should deservedly take first prize for both the beauty and the invention.9

The impression that Plato’s dialogues have made on ancient and modern read-
ers is to a considerable extent due to his skillful use of the narrative voice.

in Aeschines’ Callias and Aspasia, in Xenophon’s Hellenica, in Andocides, Lysias, Antiphon,

in Aristophanes’ Frogs and other Attic comedy, see Nails 2002: 68–74. On Polemarchus see
below, n. 21; on Hermogenes see 58n26.
9 Diog. Laert. 3.48; my translation. On Plato’s reputation as a writer and the extent of his liter-
ary influence in antiquity see Hunter 2012.
The Explicit Narrator: Narrated Dialogues 31

While the Charmides, the Lysis and the Republic, delivered by a single ­narrator,
possess only one intradiegetic level, the Parmenides is a multi-level dialogue
introducing three successive narrators. In other words, we have here two differ-
ent types of narrative, which should be approached separately.

2 A Single Narrator (the Charmides, Lysis, Republic)

2.1 The Charmides and the Lysis

The narrative and the narrator of these two dialogues are negotiated in a close-
ly similar way.10

2.1.1 Narrator-Hero
Everything we encounter in the narrated world of the Charmides and the Lysis
is seen through Socrates’ eyes. No alternative source of perception is available.
This means that the Charmides and the Lysis are thoroughly focalized through
the first-person narrator who is also a character in the story (the ­homodiegetic
narrator; cf. above, p. 17); in the two dialogues under consideration, the narrator
is the story’s leading character (the autodiegetic narrator, or the narrator-hero).
In the dialogues’ two-stage openings, Socrates’ perception is mobilized to
focus the reader’s attention on the conversation’s setting and participants. At
the first stage the setting is introduced:

Yesterday evening, I returned from the army at Potidaea… I went into the
palaestra of Taureas, which is over against the temple adjoining the porch
of King Archon, and there I found a number of persons, most of whom I
knew, but not all.11
I was going from the Academy straight to the Lyceum, intending to take
the outer road, which is close under the wall. When I came to the postern
gate of the city, which is by the fountain of Panops, I fell in with Hippo-
thales, the son of Hieronymus, and Ctesippus the Paeanian, and a com-
pany of young men who were standing with them.12

10 For a detailed analysis of the plot structure of the Lysis see Chapter 6.
11 Chrm. 153a1–6.
12 Ly. 203a1–5. Ctesippus is also a character in the Euthydemus; he was present at the death
of Socrates; see Phd. 59b9. Very little is known of Hippothales. See further Nails 2002:
119–20, 174.
32 Chapter 2

After the setting has been established, Socrates the narrator gradually focal-
izes his prospective interlocutor. This is where the two-stage arrangement of
the openings is at its most effective. As mentioned above, the transition from
one stage to another involves a change of scene and modification in the cast
of characters. Both Charmides and Lysis appear at the second stage, and their
identity is disclosed step by step:

I, in my turn, began to make enquiries about matters at home – about the

present state of philosophy, and about the youth. I asked whether any of
them were remarkable for wisdom or beauty, or both. Critias, glancing at
the door, noticed some youths who were coming in, and talking noisily
to one another, followed by a crowd. “Of the beauties, Socrates,” he said,
“I fancy that you will soon be able to form a judgment. For those who are
just entering are the advanced guard and admirers of the great beauty, as
he is thought to be, of the day, and he is likely to be not far off himself.”
“Who is he?” I said. “And who is his father?” “I rather think that you know
him too, although he was not grown up at the time of your departure,” he
replied. “Charmides is his name; he is my cousin and the son of my uncle
Glaucon.” “Certainly, I know him,” I said, “for he was remarkable even
then when he was still a child, and I should imagine that by this time he
must be almost a young man.” “You will see,” he said, “in a moment what
progress he has made and what he is like.” He had scarcely said the word,
when Charmides entered.13

“This Lysis must be a youngster,” I said, “for the name does not recall any-
one to me.” “Why,” he said, “his father being a very well-known man, he is
still commonly called by his father’s name. But I am sure that you must
know his face, for that is quite enough to distinguish him.” “But tell me
whose son he is,” I said. “He is the eldest son of Democrates, of the deme
of Aexone.” “Ah, Hippothales, I said; what a noble and really perfect love
you have found!”14

13 Chrm. 153d2-154b7. Plato’s relative (his mother’s cousin) Critias (460-403 bce) was the
leader of the oligarchy of the Thirty (“The Thirty Tyrants,” 404 bce); poet and philosopher,
one of the circle of Socrates. He appears as a speaker in the Protagoras, and possibly also
in the Timaeus and the Critias (although his identity here is disputed); frequently men-
tioned in Xenophon’s Memorabilia and Hellenica. Critias’ younger cousin and ward Char-
mides was also involved with the Thirty; he appears in Plato’s Protagoras and is one of the
characters in Xenophon’s Symposium. See further Nails 2002: 90–94, 106–11. On Critias and
Charmides in Plato and Xenophon see also Danzig 2014.
14 Ly. 204e1–10. The gravestone of Lysis and his son was unearthed in the Piraeus in 1974. See
further Nails 2002: 195–97.
The Explicit Narrator: Narrated Dialogues 33

Such gradual focalization through a series of questions and answers is no

other than Plato’s favorite strategy of zooming-in (see above, pp. 20–21). It is
important to keep in mind that here and elsewhere the narrator is not he who
supplies the missing information but he who absorbs it, thereby making the
reader absorb this information as well. Put as simply as possible, the narrator
is the one who asks questions.
But Socrates’ role as the focus of perception is not restricted to the dia-
logues’ openings, nor is zooming-in the only means of focalization that Plato
employs. As the dialogues proceed, it is Socrates again who communicates to
us the characters’ emotional state as he perceives it: the laughter of Charmi-
des or of Lysis and Menexenus,15 the blushing of Charmides and Lysis (Chrm.
158c, Ly. 204bc), the distress and embarrassment of Critias,16 the blushing and
delight of Hippothales (Ly. 213d, 222b), the deep concentration of Lysis and
even his silence (Ly. 213d, 222a, cf. 222b). The same holds true of Socrates’ own
emotional responses,17 for example, his reaction to Charmides’ beauty:

Oh my noble friend! I caught the sight of the inwards of his garment, and
took the flame, and no longer belonged to myself.18

Rather more often than not, such descriptions of a character’s emotional state
signal turning points in the dialogue’s plot. In the Charmides, Critias’ distress
and lack of restraint at Socrates’ demolishing his argument that temperance is
one’s doing one’s own business (161b6, put in the mouth of Charmides) marks
the transition to the next stage of the discussion (162cd), whereas Socrates’
diagnosis that Critias is ashamed “to admit before the company that he could
not answer my challenge” (169c7-d1) leads us to the sixth and final definition

15 Chrm. 156a4, 162b11; Ly. 207c6, 208d7. Lysis’ companion Menexenus also appears as
Socrates’ interlocutor in the eponymous dialogue (below, Chapter 3). See further Nails
2002: 202–203.
16 Chrm. 162cd, 169cd. On Socrates “remarkably thick description of the motivations of Cri-
tias” at Chrm. 162cd see Morgan 2004: 363–64. Still, I do not share Morgan’s contention
that in this and similar passages Socrates the narrator “displays a degree of knowledge
that borders on omniscience”: as Morgan herself points out further on, Socrates’ conclu-
sions about what the other characters think and feel “are based on his interpretations
of the conversational dynamics;” see Morgan 2004: 363, 364. On the discussion of what
ostensibly looks like Socrates’ momentarily omniscience in the opening scene of the Re­
public see Ferrari 2010: 23; on Plato’s preventing in the Phaedo the effect of narrator’s om-
niscience by means of metalepsis see below, Chapter 4, with n. 37.
17 On the role of Socrates the narrator in providing the background information and de-
scribing his own and the characters’ emotional responses see also Murley 1955: 283–85;
Bowery 2007: 85, 92–96.
18 Chrm. 155d3-4; my emphasis (see below, on breaking the boundaries between narrative
levels that the passage involves).
34 Chapter 2

of temperance as the knowledge of good and evil. The narrator’s text at the
reversal of the argument in the Lysis is no less effective (below, pp. 135–38).

2.1.2 Metalepsis
On three occasions in the Charmides, all of them exclamations expressing
Socrates’ excitement at the sight of Charmides’ beauty, Socrates the narra-
tor directly addresses his unidentified narratee (“O my friend!” or “O my no­
ble friend!”).19 This is equivalent to the practice, widespread in the 18th- and
19th-century novel, of the author/narrator addressing the implied reader
directly – as, for example, in Charlotte Brontë’s Jane Eyre: “Gentle reader, may
you never feel what I then felt!” (Ch. 27; my emphasis). In causing the extradi-
egetic narrative act to intrude into the intradiegetic domain, such an apos-
trophic address would qualify as a mild case of rhetorical metalepsis, namely
metalepsis that does not affect the plot level (see above, p. 23).20

2.2 The Republic

As in many other respects, so in its treatment of narrative voice Book 1 of the
Republic stands apart from the rest of the dialogue. Some of the differences are
due to its position as the introductory book – a “prologue” (prooimion), as Plato
himself put it (Resp. 357a2), others much less so. However that may be, Socrates
the narrator of Republic 1 is undeniably much closer to Socrates of the Charmi­
des and the Lysis than to Socrates the narrator of Republic 2–10 (more below).
This however does not affect his playing the part of narrator-hero in both.

2.2.1 Narrator-Hero
The narrative of the Republic opens with Socrates’ description of the dialogue’s
setting:

I went down yesterday to the Piraeus with Glaucon the son of Ariston,
that I might offer up my prayers to the goddess… When we had finished

19 Chrm. 154b8, 155c5, 155d3 (quoted above, with n. 18).

20 On a metaleptic relation with the reader by means of the second-person pronoun see
McHale 1987: 223–25; on the metaleptic potential of apostrophic addresses to the read-
er see Pier 2013: 3.2.26. Since rhetoric, to which the term “apostrophe” (“a turning away
from”) originally belongs (see de Jong 2009: 93 n. 16), does not create a narrated world
different from the world of the audience, the orator’s addresses to the audience do not
involve metaleptic transgression. In fiction the situation is different. Here, the addresses
to the reader satisfy the sufficient conditions for metalepsis as recently formulated by Ju-
lian Hanebeck, in that the narrator and his addressee belong “to distinct spatiotemporal
frames of reference, which results in a fiction-internal paradoxical transgression defying
representational logic” (Hanebeck 2017: 25).
The Explicit Narrator: Narrated Dialogues 35

our prayers and viewed the spectacle, we turned in the direction of the
city; and at that instant Polemarchus the son of Cephalus chanced to
catch sight of us from the distance … and told his servant to run and bid
us wait for him. … and in a few moments Polemarchus appeared, and
with him Adeimantus, Glaucon’s brother, Niceratus the son of Nicias, and
several others who had been at the procession.21

As distinct from the Charmides and the Lysis, Socrates makes no new acquain-
tances in this dialogue; still, the narrator’s focusing his attention on the main
interlocutor proceeds in stages here as well. As with Charmides and Lysis,
Socrates meets Thrasymachus,22 his main interlocutor in Republic 1, only in the
second stage of the exposition, and it is not until he has first conversed with
Cephalus and then with Polemarchus does their confrontation begin. Cepha-
lus, Socrates’ first interlocutor, is the first to be zoomed-in:

Accordingly we went with Polemarchus to his house; and there we found

his brothers Lysias and Euthydemus, and with them Thrasymachus the
Chalcedonian, Charmantides the Paeanian, and Cleitophon the son of
Aristonymus. There too was Cephalus the father of Polemarchus, whom I
had not seen for a long time, and I thought him very much aged. (328b4-c1)

As in the Charmides and the Lysis, the first-person narrator of the Republic fo-
calizes the narrative not only at the beginning but also in the body of the dia-
logue. It is indeed Socrates the narrator who makes us perceive the laughter of
Cephalus (331d9), Thrasymachus’ anger and sarcastic cackle (336b-d, 337a3),
his eagerness to interfere in the discussion (338a5-6), his frustrated intention
to leave (344d1-5), as well as his sweating and blushing (350d1-3).23 And it is
Socrates again who describes his own emotional responses to the events of the
dialogue: “I was panic-stricken at his words, and could not look at him without
fear” (336d5-6). As in the Charmides and the Lysis again, no focus of perception
other than the narrator is available.

21 Resp. 327a1-c3. Polemarchus, son of Cephalus of Syracuse and brother of the orator Lysias,
was one of the wealthiest people in Athens; he was executed by the Thirty in 404 bce; see
further Nails 2002: 251. Apart from the Republic, Plato’s brother Glaucon appears in the
frame stories of the Parmenides and the Symposium; see n. 33 below and n. 46 to Chapter 4.
22 A prominent rhetorician, Thrasymachus is also referred to in the Phaedrus and the
pseudo-Platonic Cleitophon, as well as in Aristophanes and Lysias; in 407 bce he appar-
ently spent time in Athens on a diplomatic mission on behalf of his native Chalcedon. See
further Nails 2002: 288–90.
23 On the descriptions of emotional states of the characters in Republic 1 and the other nar-
rated dialogues see Murley 1955; Bowery 2007.
36 Chapter 2

In Books 2–10 of the Republic Thrasymachus withdraws into the back-

ground, and the atmosphere of the dialogue changes. I can hardly improve
on Ruby Blondell’s assessment of the stylistic transformation that takes place
here: “Book 2 presents us with an unequivocal shift in philosophical method,
accompanied by an equally radical change in dramatic style.”24 This would be
true of the dialogue’s narrative voice as well. To be sure, Socrates the narrator is
still capable of producing such vivid descriptions as, for example, the following
episode at the beginning of Book 5:

I was proceeding to tell the order in which the four evil forms appeared
to me to succeed one another, when Polemarchus, who was sitting a little
beyond Adeimantus, stretched out his hand, took hold of the upper part
of his coat by the shoulder, and drew him towards himself, leaning for-
ward so as to be quite close and saying something in his ear, of which we
only caught the words, “Shall we let it go, or what shall we do?” (449a7-b6)

Significantly, this intervention emerges at a turning point of the dialogue. It

introduces the great digression consisting of “three waves” of argument (equal-
ity of women, wives and children in common, philosopher kings), which em-
braces Books 5–7 of the Republic and forms its philosophical core.25 As we have
noted, placing the narrator’s text at structurally important junctions is charac-
teristic of the other narrated dialogues as well.
On the whole, however, Socrates the character is much more prominent
in Books 2–10 than Socrates the narrator. He not only leads the conversation
but also punctuates its course by recapitulating and commenting on the ar-
gument.26 Socrates’ principal interlocutors are now Plato’s brothers Glaucon
and Adeimantus. While they certainly take part in navigating the dialogue’s
plot (the beginning of Book 2 immediately comes to mind here),27 whatever
their emotions they mostly keep them to themselves. Through 264 pages of
the Stephanus text laughter is registered only twice.28 The contrast with Plato’s
other narratives, including the narrated parts of the mixed dialogues, could

24 Blondell 2002: 190. See also Ferrari 2010: 22–28.

25 See further below, Chapter 6, p. 128.
26 See e.g. Resp. 357a, 358c, 368d-369a, 376d, 392c, 398bc, 403c, 420b, 424b, 432c, 434de,
435cd, 443b, 445c, 450a, 451bc, 472b, 484a, 490cd, 502c, 502e, 503a, 504b, 528a, 528e, 530e,
531d, 532de, 541b, 543a-c, 545ab, 548cd, 572b, 588b, 596a, 603de. On the role of metanarra-
tive comments in Plato’s dialogues see Chapter 6.
27 Discussed in more detail in Chapter 6, p. 128.
28 Resp. 398c7, 451b2. Glaucon is the one who laughs on both occasions. At 608d5 he is de-
scribed as looking at Socrates “in amazement” (thaumasas).
The Explicit Narrator: Narrated Dialogues 37

not be stronger: in the Euthydemus, whose narrated part is equivalent to 34

pages of the Stephanus text, laughter is registered six times, in the Protagoras
(51 Stephanus pages) four times, and in the Phaedo (59 Stephanus pages) no
fewer than eight times.29

2.2.2 Metalepsis
A remarkable case of metalepsis, which, moreover, is much more sophisticated
than the apostrophic addresses we encountered in the Charmides, emerges in
Republic 1. On one occasion Socrates the narrator explicitly distances himself
from Socrates the character:

Thrasymachus agreed on all these points, not as smoothly as I tell now,

but reluctantly and with great effort, sweating abundantly – for it was
summer into the bargain. And then I saw what I had not yet seen before:
Thrasymachus blushing.30

The words “not as smoothly (rhaidiōs) as I tell now,” relating to Thrasymachus’

reluctant yielding to Socrates’ argument, belong to the extradiegetic level,
that is, to the “now” of the act of narration in which Socrates the narrator is
involved. For a moment, Socrates’ remark breaks the boundary between the
extradiegetic and the intradiegetic level of the narrative without, at the same
time, substantially affecting the latter, which qualifies it as rhetorical metalep-
sis (above, p. 23).
The metaleptic passage in Republic 1 draws a clear-cut dividing line between
the first-person narrator and his earlier self as a character in the story. It un-
ambiguously demonstrates that Plato was fully aware of the distance between
“the narrating self” and “the narrated self” – a kind of awareness, it should be
added, which is attested in Greek literary tradition as early as the Homeric Od­
yssey, in the remarks of Odysseus the narrator on the reckless behavior of his
former self in the Cyclops’ cave.31
But the passage is also significant in other respects. Socrates’ words “not as
smoothly as I tell now” make the reader realize that the dialogue presents an
edited version, as it were, of the original conversation, adjusted as Socrates the
narrator saw fit. In other words, Socrates is envisaged not merely as the one

29 On the Euthydemus, Protagoras, and Phaedo see below, Chapter 4.

30 Resp. 350c12-d3; my translation and emphasis.
31 Od. 9.228–29: “But I would not agree [with the plea of the companions to leave the place]
– better, much better, if I had! But no, I was eager to see the master”… (tr. W. Shewring,
slightly adapted).
38 Chapter 2

who reports the conversation in which he himself has taken part but also as
the one who, to borrow G.R.F. Ferrari’s expression, “edits it for smoothness.”32
Additional instances of narrator’s selective presentation of what is supposed
to be a report of real events will be found in the Symposium and the Theaetetus
(Chapter 4).

...
To sum up, in the Charmides, the Lysis, and the Republic the narrator and the
focus of perception concur. Socrates the narrator is consistently presented as
the person through whom the story is focalized. As Republic 1 makes clear, Pla-
to is fully aware of the distance between Socrates the narrator and Socrates the
character; nevertheless, his dialogues’ first-person narrator consistently adopts
the focalizing role of his former self (cf. above, pp. 19–20). This strongly sug-
gests that the markers of focalization appearing in the text, such as Socrates’
description of the dialogue’s setting, his gradual focusing on the main interloc-
utor (the zooming-in technique) and his perception of the characters’ physical
appearance and emotional responses, should be regarded as signals of narrato-
rial authority. The point is important, and I will return to it in the concluding
section of this chapter.

3 Multiple Narrators (the Parmenides)

In the Parmenides the narrator’s role is negotiated differently from the other
narrated dialogues. Cephalus of Clazomenae narrates how, on arriving in Ath-
ens, he and his compatriots met Glaucon and Adeimantus and asked them to
be introduced to their half-brother Antiphon. As a youth Antiphon had often
heard from Zeno’s friend Pythodorus about the young Socrates’ meeting with
the philosophers Parmenides and Zeno, held in Pythodorus’ house.33 The com-
pany arrives in the house of Antiphon and finds him absorbed in supervising
the fitting of a bridle for one of his horses. That task accomplished, Antiphon

32 Ferrari 2010: 23. Cf. Henry James, The Portrait of a Lady, Ch. 19: “‘Tell me what they do
in America,’ pursued Madame Merle, who, it must be observed parenthetically, did not
deliver herself at once of these reflexions, which are presented in a cluster for the conve-
nience of the reader.”
33 Cephalus of Clazomenae appears only in this dialogue. Plato’s brothers Adeimantus and
Glaucon are Socrates’ main interlocutors in the Republic (see n. 21 above). Antiphon the
son of Pyrilampes was Plato’s maternal half-brother. Pythodorus was an Athenian gen-
eral in the Peloponnesian War; he is frequently mentioned by Thucydides; see Nails 2002:
259–60.
The Explicit Narrator: Narrated Dialogues 39

reluctantly accedes to Cephalus’ request and starts recounting what Pythodor-

us told him about Socrates’ meeting with Parmenides and Zeno. Cephalus’
­retelling the narrative of Antiphon, who was in turn retelling Pythodorus’ nar-
rative, eventually switches into dramatic form (see Chapter 1).
As distinct from the other narrated dialogues, the Parmenides contains more
than one intradiegetic level. The narrative act of Pythodorus, who was present
at the original conversation, is intradiegetic by virtue of its being mediated
by the narrative act of his narratee Antiphon. The narrative act of Antiphon
is in turn intradiegetic by virtue of its being mediated by the narrative act of
his narratee Cephalus. Cephalus is the dialogue’s extradiegetic narrator, who
addresses its unidentified narratee and, simultaneously, the implied reader. At
the same time, Cephalus is also a character in the frame story he delivers, so
that his narration forms the dialogue’s first intradiegetic level (see above, pp.
19–20, on the ambivalent position of the first-person narrator). The transition
to the second intradiegetic level is accompanied by a spatial change: Cepha-
lus and his company move to the house of Antiphon, another frame narrator
of the dialogue. The transition to the third intradiegetic level is accompanied
by a spatiotemporal change: we are transferred to the past, to the house of
Pythodorus in the Ceramicus and to the original narrator of the story. Each
transition is accompanied by a shift of narrative voice: the narratee on one
intradiegetic level becomes the narrator on another.
As a result, the dialogue features two frame narrators, Cephalus and Anti­
phon,34 who are absent from the main story (heterodiegetic narrators; see
above, p. 17), and the narrator properly speaking, Pythodorus, who is present in
the main story but plays no active part in it (homodiegetic narrator-observer;
see above, p. 17):

1st level Cephalus FRAME

2nd level Antiphon FRAME
3rd level Pythodorus STORY

3.1 Frame Narrator i: Cephalus

Cephalus, the primary narrator of the Parmenides, is absent from the main
story. Accordingly, as distinct from Socrates of the Republic, the Lysis, and the
Charmides, Cephalus is in no position to focalize the dialogue in its entirety. His
narratorial authority spreads only onto the frame narrative, namely his meet-
ing with Glaucon and Adeimantus, their joint visit to the house of Antiphon,

34 On frame narrators see above, pp. 21–22.

40 Chapter 2

and the description of Antiphon’s daily routine and his grudging consent to
tell the story (the first intradiegetic level). Within this limited sphere, however,
Cephalus’ role as the focus of perception is exhibited in full. This is how, for
example, he approaches Adeimantus:

“I want you to tell me the name of your half-brother, which I have forgot-
ten. He was a mere child when I last came here from Clazomenae, but
that was long time ago; his father’s name, if I remember rightly, was Pyri­
lampes?” “Yes,” he said. “And his?” “Antiphon.” (126b1-7)

This is a standard case of zooming-in, typical of Plato’s descriptions of the ini-

tial focalization by the narrator (above, pp. 20–21). Even Cephalus’ beginning
the inquiry by naming the father of the person on whom his narrative is focal-
ized parallels similar inquiries in other dialogues (see above on the Lysis, and
below on the Laches and the Theaetetus).

3.2 Frame Narrator ii: Antiphon

Similar limitations apply to Antiphon (the second intradiegetic level). Anti-
phon is of course one of the characters in Cephalus’ story, but his own story
and the circumstances of his becoming exposed to Pythodorus’ report are
barely mentioned: we are only told that Cephalus and his countrymen have
heard that Antiphon, who spent considerable time in the company of Py-
thodorus, memorized a report of the conversation among Parmenides, Zeno,
and Socrates, which he heard from Pythodorus more than once (126b8-c3).
In fact, Antiphon’s narratorial authority is made manifest only twice – at the
beginning of his narrative, when he mediates Pythodorus’ description of his
guests’ appearance (below), and in the episode signaling the transition to Par-
menides’ demonstration of the method of deduction, that is, to the dramatic
part of the dialogue:

When Zeno had thus spoken, Pythodorus, according to Antiphon’s ac­

count, said that he himself and Aristoteles and the whole company en-
treated Parmenides to demonstrate what he is speaking of. “I cannot
refuse,” said Parmenides, “and yet I feel rather like the horse of Ibycus,
for, when in his old age, against his will, Ibycus fell in love, he likened
himself to an old racehorse, who was about to run in a chariot race,
shaking with fear at the course he knew so well. And I also experience a
trembling when I remember through what an ocean of words I have to
wade at my time of life. But I must indulge you for, as Zeno says, we are
The Explicit Narrator: Narrated Dialogues 41

here alone. Where shall I begin, and what shall be our first hypothesis if
I am to play this laborous game (paidian)?”35

The passage is heavily marked (see especially the digression on Ibycus), and
justifiably so. This is the turning point of the dialogue, and it is tempting to sug-
gest that by referring to Antiphon’s narratorial authority at this specific point
Plato meant to create the impression that, as with Euclides of the Theaetetus
(Chapter 1), it is Antiphon who is envisaged as responsible for the dramatic
form assumed in the rest of the dialogue.

3.3 Narrator-Observer: Pythodorus

The frame narrators’ narratorial authority does not exceed the frame story, but
this does not mean that the main narrative of the Parmenides remains unfocal-
ized. The focalization, filtered though it is through the frame narrators’ voices
(more below), is effected by Pythodorus, the ultimate source of the narrative
(the third intradiegetic level). Pythodorus is he to whom the narratorial au-
thority of the Parmenides’ main story belongs.
At the same time, as distinct from Socrates the narrator of the Charmides,
the Lysis, and the Republic, Pythodorus the narrator of the Parmenides does
not play the leading part in the story he delivers. His role is that of observer. As
Genette put it,

Absence is absolute, but presence has degrees. So we will have to differ-

entiate within the homodiegetic type at least two varieties: one where
the narrator is the hero of his narrative (Gil Blas) and one where he plays
only a secondary role, which almost always turns out to be a role as ob-
server and witness: Lockwood, the anonymous narrator of Louis Lam­
bert, Ishmael in Moby Dick, Marlowe in Lord Jim, Carraway in The Great
Gatsby, Zeitblom in Doctor Faustus – not to mention the most illustrious
and most representative one of all, the transparent (but inquisitive) Dr.
Watson of Conan Doyle.36

This is precisely where Pythodorus is positioned. Judging by Xenophon’s

Symposium and Plato’s own Symposium and Phaedo, all employing narrators-
observers, such positioning of the narrator would not have come as a surprise
to Plato’s readers.

35 Prm. 136e5-137b1; my emphasis. On Plato’s concept of paidia see Chapter 7, pp. 157–59.
36 Genette 1980: 245. On homodiegetic narrator see above, p. 17.
42 Chapter 2

Plato never forgets to emphasize that Pythodorus, a witness to the original

conversation, is the bearer of narratorial authority:

And Antiphon told us that Pythodorus said that Parmenides and Zeno
came to Athens at the Great Panathenaea; Parmenides was well advanced
in years, with white hair, of distinguished appearance, about sixty-five
years old; Zeno was then nearly forty years of age, tall and fair to look
upon. (127a7-b5; my emphasis)

Moreover, only from the moment Pythodorus appears on the stage are we able
to follow the action:

These [his writings] Zeno himself read to them in the absence of Par-
menides, and had very nearly finished when, according to Pythodorus’
account, he himself entered, and with him Parmenides and Aristoteles
who was afterwards one of the Thirty, and heard the little that remai­
ned. Pythodorus, however, had already heard Zeno earlier. (127c5-d5; my
emphasis)

And again, through Pythodorus we perceive Parmenides’ and Zeno’s reaction

to the exposition of the Theory of Forms by the youthful Socrates:

Pythodorus told that, while Socrates was speaking, he himself supposed that
Parmenides and Zeno would be upset at each one of his arguments, but
they listened attentively and often looked at one another, and smiled as if
in admiration of Socrates. (130a3-7; my emphasis)

Note that what Pythodorus describes here is his perception of Socrates being
focalized by characters in his story. Such focalization by characters, occurring
fairly often in Plato, should be distinguished from the macro-focalization ef-
fected by the narrator. To quote Monika Fludernik:

Focalization on the story level (i.e. one character observing another) does
not properly belong to macro-focalization, i.e. the focalization of an en-
tire text, but it is a small-scale management of the plot function. The only
really important issue is that of the consistent or inconsistent rendering
of the entire story by means of a particular slant or filter.37

37 Fludernik 1996: 258.

The Explicit Narrator: Narrated Dialogues 43

But the episode also outlines the limits of focalization as implemented by the
first-person narrator. Note indeed that Pythodorus’ perception fails him the
very moment we move from the characters’ outward expression of emotions
(in this case their smiles and glances) to their inner emotional life: before his
perceiving the philosophers’ reaction, Pythodorus could only speculate on
what their attitude to Socrates’ argument may be like. The episode underscores
the fact that as regards Plato’s narrators we will always be dealing with the
so-called external focalization: the idea of omniscient narrator is absent from
Plato’s dialogues.38
The Parmenides is the only dialogue in the Platonic corpus whose opening
is a frame narrative in the strict sense of the word: all the other frame stories
in Plato’s dialogues are thoroughly dramatic. Accordingly, the Parmenides is
also the only Platonic dialogue that explicitly introduces multiple intradi-
egetic levels as represented by the successive narrations of Cephalus, Anti-
phon, and Pythodorus. Plato compensates for this cumbersome arrangement,
first by almost completely suppressing the narrative level that Antiphon’s
narration belongs to, and second by eventually abandoning the narrated
form altogether. Small wonder that in his other structurally complex narra-
tives a more economic model of multi-level arrangement has been adopted
(see Chapter 4).

3.4 Metalepsis
No frame-breaking in the strict sense of the world occurs in the Parmenides.
Nevertheless, thanks to Plato’s use of reported speech in the first part of the
dialogue, the presence of the frame narrators, first and foremost Antiphon,
is felt throughout Pythodorus’ narrative (see the quotations above). The Par­
menides is thus consistently presented as a narrative of a narrative of a narra-
tive. Repeatedly reminding the readers of this fact, Plato blurs the boundaries
between different narrative levels, thereby leaving us no opportunity to im-
merse ourselves in the world of the narrated story. This mode of presentation,
resulting in all narrative levels being simultaneously present throughout the
narrative, is especially characteristic of the Symposium (see above, p. 23 and
below, Chapter 4).

3.5 Juxtaposition of the Frame and the Narrative

The framing story and the framed narrative of the Parmenides are both f­ irmly
focalized on Socrates. Thus Pythodorus not only lavishly depicts the ­youthful

38 On external focalization see Genette 1980: 189: “the narrator says less than the character
knows.” On Plato’s narrator’s lack of omniscience see above, with n. 16.
44 Chapter 2

Socrates expounding the Theory of Forms and coping with Parmenides’

criticism of it, but also strives to convey Parmenides’ and Zeno’ admiration of
the gifted and intellectually aspiring youth (above, p. 42). This fully corresponds
to the initial request of the primary narrator Cephalus to learn about “a conver-
sation that took place once between Socrates, Zeno and Parmenides” (126c1-2),
a phrase that is clearly meant to advertise the scope of the subsequent narra-
tive.39 Alongside the Symposium and the Theaetetus, which will be discussed
in Chapter 4, the Parmenides is a dialogue whose multi-level structure is mobi-
lized for creating an in-depth biographical sketch of the protagonist.
An additional effect of the Parmenides’ juxtaposing the frame story and
the main narrative is its establishing an intertextual connection with the
Republic: Plato’s brothers Glaucon and Adeimantus, Socrates’ main interlocu-
tors in the Republic, are presented in the Parmenides as listening to Anti-
phon’s retelling Pythodorus’ account of the meeting between Socrates and
Parmenides.40

4 Conclusions

As the first-person narrative adopted in Plato’s narrated dialogues has nu-

merous counterparts in modern fiction, these dialogues readily lend them-
selves to being approached in terms of narrative theory. The first-person
narrator of the narrated dialogues can be placed either within the narrated
world of the main story (the homodiegetic narrator) or, as in the case of the
multi-level Parmenides, within that of the narrative frame (the frame narra-
tor, who is heterodiegetic as regards the main story). The narrator of Plato’s
narrated dialogues is either the story’s main character (the autodiegetic nar-
rator, or the narrator-hero) or an eyewitness of the events described (the
narrator-observer).
In the Charmides, the Lysis, and the Republic, which feature a single narra-
tor, the narratorial authority, issuing from the narrator’s first-hand experience
of the narrated world as acquired by his former self, finds its expression in

39 Cf. Zuckert 1998: 877: “At whatever stage in his career Plato wrote the Parmenides, he
clearly meant his readers to take it as their first view of a young Socrates.” A similar adver-
tising strategy is applied in the frame story of the Symposium; cf. Symp. 172a6-b3, Glaucon
to Apollodorus, “I want to ask you about the gathering of Agathon, Socrates, Alcibiades,
and other men of those days (τῶν τότε) who attended the dinner. What were their speech-
es in praise of love?” On the expression “men of those days” see below, n. 44 to Chapter 4.
40 Cf. Scolnicov 2003: 44: ‘By them [Glaucon and Adeimantus], we are directed to the meta-
physical scene of the Republic.’
The Explicit Narrator: Narrated Dialogues 45

markers of focalization that regularly accompany the narrator throughout the

text of the dialogues. In the multi-level Parmenides, the primary narrator is
placed within the narrated world of the frame story and his narratorial author-
ity is manifested, again, in markers of focalization issuing from his first-hand
experience of this world. At the same time, the first- and second-level narrators
are both excluded from the narrated world of the main story. The narratorial
authority over the dialogue’s main story is delegated to the bottom-level
narrator-observer who belongs to the world of the main story and manifested,
again, in markers of focalization that regularly accompany the narrator
throughout the text of the dialogue.
As Republic 1 makes clear, Plato was fully aware of the distance between the
first-person narrator and his former self who is present as a character in the
narrated world of the dialogues (“the narrating self” and “the narrated self;”
see above, p. 37). This, however, did not prevent him from treating the markers
of focalization that accompany the narrator’s former self as a character in the
story as issued by the narrator himself (see further Chapter 1, pp. 19–20). As
a result, the first-person narrator of Plato’s narrated dialogues is consistently
presented as focalizing the narrative in its entirety. This strongly suggests that
the markers of focalization that appear in the narrator’s text should be treated
as signals of narratorial authority. Four such markers of focalization can be
distinguished:
(a) the narrator perceives and describes the dialogue’s setting;
(b) the narrator gradually absorbs information concerning the principal in-
terlocutor and/or the subject of the dialogue (the zooming-in technique);
(c) the narrator registers the characters’ physical appearance, movements,
emotional responses, and the like;
(d) the scope of the narrative and the amount of information it contains
are circumscribed by the narrator’s personal experience (external
focalization).
As the above list demonstrates, Plato’s narrator is wholly identified in terms
of perception and knowledge. The signals of narratorial authority are in fact
nothing but different aspects of one and the same phenomenon, namely the
positioning of the narrator as the dialogue’s focus of perception through whose
mediation the readers cognitively absorb everything that takes place in the
dialogue.
Could the same signals of narratorial authority be used in a text with no
explicitly manifested narrator? Testing the dramatic dialogues for the signals
of narratorial authority will be our primary task in the next chapter. An ad-
ditional function of Plato’s narrator to be further examined is the narrator’s
ability, highlighted in Republic 1, to edit the original conversation according to
46 Chapter 2

his idea of how this conversation should be presented to the audience. Finally,
it has been remarked more than once that Plato positions the narrator’s text
at structurally important junctions. If the practice is consistent, it may imply
that Plato negotiates the narrator as the one who not only focalizes the story
but also fashions it, thereby taking responsibility for the dialogue’s plot. At this
stage, however, our evidence is not solid enough (for example, so far we have
encountered only one case of narrator-observer) to draw an informed conclu-
sion. The discussion therefore should be postponed until the way the two other
groups of dialogues behave in this respect has been examined as well.
 Chapter 3

The Implicit Narrator: Dramatic Dialogues

1 Introducing the Dramatic Dialogues

Of the three groups of Plato’s dialogues, the dramatic dialogues form the larg-
est and the most heterogeneous one. It comprises the earliest and the latest
dialogues, the shortest and the longest pieces in the corpus, the most elaborate
and the least elaborate, and not a few dramatic dialogues are spurious. Small
wonder, then, that Plato’s dramatic dialogues exhibit a great variety of forms.
This becomes immediately obvious from their openings.
The short and simply structured “Socratic” dialogues normally open with a
concise and efficient “zooming-in” in the course of which Socrates or another
character establishes for the reader’s benefit his interlocutor’s identity and
­social and professional standing, as well as the immediate circumstances of
the encounter:
− In the Euthyphro, Euthyphro questions Socrates as to the reasons for his ar-
rival at the Porch of the King Archon.
− In the Crito, Socrates questions Crito as to the reasons for his arrival in the
prison early in the morning.
− In the Ion and Hippias Maior, Socrates questions the foreigners Ion and Hip-
pias, respectively, as to the reasons for their arrival in Athens.
− In the Hippias Minor, Eudicus questions Socrates as to the reasons for his
silence after Hippias’ display of his art.
− In the Menexenus, Socrates questions Menexenus as to the outcome of the
proceedings at the meeting of the Athenian Council.
Of the dialogues whose authenticity has been questioned, only the Hippias
Maior functions as a regular Platonic dialogue in this respect. Alcibiades ii
opens with a single question (quoted above, p. 10) rather than a series of ques-
tions necessary for producing the zooming-in effect; in the Theages, the ques-
tioning is preceded by a long speech by the as yet unidentified Demodocus
(121b-122b), which is sharply at variance with the usual procedure. There is no
zooming-in in the spurious Cleitophon, Hipparchus and Minos nor, s­ ignificantly,
in Alcibiades i.1

1 Cf. Friedländer 1964: 232: “While the dialogue corresponds, in this general situation, to others
like the Charmides and the Lysis, there are no secondary figures in the Alcibiades nor is there
any setting full of charm or symbolic meaning.”

© koninklijke brill nv, leiden, ���9 | doi:10.1163/9789004390027_004

48 Chapter 3

In the longer and more complex dialogues the introductory zooming-in is

enhanced into an elaborate two-stage opening closely resembling the open-
ings of the narrated dialogues. As in the latter, the transition from one stage to
another involves a spatial change and/or modification in the cast of characters.
− In the Laches, (i) the opening conversation between Lysimachus and Me-
lesias on the one hand and Nicias and Laches on the other does not include
Socrates, (ii) who joins the initial interlocutors only at a later stage (below,
pp. 64–65).
− In the Gorgias, (i) Callicles, Socrates, and Chaerephon meet in the street,
apparently before Callicles’ house, (ii) and only at a later stage do they enter
the house, where they meet Gorgias and Polus.2
− In the Phaedrus, (i) Socrates meets Phaedrus in the city, (ii) and they walk
together to a shady spot on the banks of the Ilissus, where the conversation
takes place; no change in the cast of characters is involved.
− In the Laws, (i) the interlocutors meet in Cnosus, and, (ii) on their way
to the cave and temple of Cretan Zeus, they arrive at a grove of cypresses,
where most of their conversation apparently takes place; no change in the
cast of characters is involved (625b8–c2, cf, 229a9–b2).
It is hard to avoid the impression that the transition from one stage to another,
as well as the changes in the setting and/or the cast of characters that accom-
pany the transition, are modeled on the narrated dialogues, where they form a
standard and well-developed pattern (above, pp. 27–28). This can be taken as
an additional indication of Plato treating the dramatic dialogues as a form of
narrative.3
The same pattern, albeit in reduced form, is present in the Sophist-Statesman
and Timaeus-Critias: in these dialogues, Plato economizes on the first stage
of the exposition by creating the impression that it had taken place just be-
fore the dialogue began. This is obviously because all the dialogues of this
group are set as sequels (above, p. 7). A similar strategy is applied in the

2 Cf. Dodds 1959: 188: “A similar change of scene from outdoors to indoors is explicitly indi-
cated at Lysis 206de and in the prelude to the Theaetetus, 143b”; in Dodds’ opinion, however,
the action takes place at the entrance to a public building. On the two-stage opening of the
Gorgias and its similarity to the settings of narrated dialogues see also Thesleff 2003: 552–53.
3 Cf. de Jong 2004: 7–8: “In the case of the dramatic dialogues of Plato and Lucian, that is to say,
those dialogues which lack a narrative frame and consist solely of speeches, it seems most
sensible – in view of their close similarity to the dialogues with a narrative frame – to see
them as narratives with a suppressed primary narrator and suppressed primary narratees”
(original emphasis); cf. also de Jong 2014a: 17 and above, Chapter 1 with nn. 16 and 20. On Lu-
cian see Whitmarsh 2004: 468; on parallels between the narrated and the dramatic dialogues
see also Tarrant 1996: 143–46.
The Implicit Narrator 49

Cratylus and the Philebus, which start in the middle of an ongoing discussion –
“with the second act,” to borrow Dorothea Frede’s apt characterization of the
Philebus.4 Two dramatic dialogues, the Meno and the Laws, combine an abrupt
start with a series of strategically placed remarks elucidating the d­ ialogue’s
setting.
The closures of the dramatic dialogues often bring the reader back to
the opening scene: this is done either by indicating the departure of one or
more participants (Euthyphro in the Euthyphro, Crito in the Crito, Cratylus
and ­Hermogenes in the Cratylus, Socrates in the Meno, the Philebus, and the
Theaetetus, Socrates and Phaedrus in the Phaedrus) or, as in the Menexenus
and the Laches, by bringing the opening scene to a conclusion. But just as fre-
quently the setting, once introduced, is ignored, and it is the conclusion of
the argument that supplies the closure: this is the case of the Ion, the Hippias
Maior, the Hippias Minor, the Gorgias, the Sophist, the Statesman, the Timaeus,
and the Laws.
It is not out of the question that Plato was the first to compose dramatic
dialogues. At least this is what seems to transpire from a papyrus fragment that
probably belongs to the first or the second century ce:

… imitating Sophron the mime writer too in respect of the dramatic ele-
ment (to dramatikon) of the dialogues; for Aristotle is not to be believed
when he says in his malice against Plato, in Book One of On Poetry, that
dramatic dialogues had been written even before Plato by Alexamenus
of Tenos.5

Aristotle’s naming of Alexamenus of Tenos, an unknown entity, as Plato’s

predecessor in the writing of dialogues is also mentioned in Athenaeus and
Diogenes Laertius;6 yet, the papyrus fragment is unique in that it limits Alexa-
menus’ supposed priority to the dramatic dialogues alone.
Michael Haslam has edited and thoroughly studied the papyrus. He con-
cludes that like the other sources at our disposal it refers to the entire genre
of the dialogue, therefore adds nothing substantial to the extant evidence.7

4 D. Frede 1996: 218.

5 P. Oxy. 3219. Tr. M. Haslam.
6 Ath. 11.505c; Diog. Laert. 3.48.
7 Haslam 1972: 17–24. The argument proceeds as follows. (1) Since the terms “dramatic” and
“mimetic” are often interchangeable, there is no substantial difference in meaning between
the two. (2) Therefore, the narrated dialogues, although not dramatic in the technical sense,
should also be qualified as mimetic. (3) Ergo, when using the term “dramatic” the author of
the papyrus meant the vividness characteristic of Plato’s dialogues rather than their mode
50 Chapter 3

However, as Haslam himself admits, this would be at variance with the stan-
dard classification of Platonic dialogues into dramatic, narrated (“diegetic” or
“diegematic”) and mixed, which was widely used in antiquity.8 We also might
consider that first, our sources repeatedly credit Plato with bringing Sophron’s
mimes to Athens,9 and second, that Plato was in all probability not the first to
compose narrated dialogues (above, pp. 28–29); this would lead to the con-
clusion that whatever role Alexamenus might have played, the argument for
Plato’s priority in introducing dramatic dialogues should be taken seriously.
Reasonably therefore, this was the argument that the author of the papyrus
text polemically emphasized.10
However that may be, we saw in Chapter 1 that Republic 3 leaves no room for
doubt that in Plato’s eyes the dramatic dialogues, as indeed any composition
cast in dramatic form, should be approached as a form of narrative. We also saw
that Theaetetus 143b5–c5 negotiates a dramatic dialogue as a narrated dialogue
whose narrator, although suppressed, is nevertheless implicitly present in the
text. In our analysis of the narrated dialogues in Chapter 2 we concluded that
Plato’s narrator is consistently defined in terms of perception, which allowed
us to isolate signals of narratorial authority, all based on markers of focaliza-
tion (pp. 44–45). In this chapter, the signals of narratorial authority identified
in Chapter 2 will serve as criteria on the basis of which the dramatic dialogues
will be examined.
I shall start with the so-called Athenian part of the Theaetetus (its frame
story takes place in Megara) for two reasons. First, although the Theaetetus
formally belongs to the mixed dialogues and will be treated as such in the
next chapter, its bottom-level narrative is, as we have seen, programmatically
­introduced as a regular dramatic dialogue. Secondly, the Theaetetus is the only
dialogue the results of whose analysis can be controlled by Plato’s explicit
identification of it as a dialogue with an implicit narrator. The results obtained

of presentation. We saw, however, that for Plato at least, “the dramatic” and “the mimetic”
are not fully interchangeable entities: a dialogue can be mimetic – for example, the narrated
dialogues – without at the same time being presented in dramatic form (above, pp. 1–5).
8 See e.g. Diog. Laert. 3.50: “I am not unaware that there are other ways in which certain
writers classify the dialogues. For some dialogues they call dramatic, others narrated,
and others again they call mixed. But the terms they employ in their classification of the
dialogues are better suited to the tragic stage than to philosophy.” Tr. R. D. Hicks, slightly
adapted.
9 On Sophron and the tradition of his influence on Plato see Haslam 1972: 18–19; Laborderie
1978: 18–25; Clay 1994: 33–37; differently Ford 2010.
10 Similarly Tarrant 2000: 6. As Wilamowitz-Moellendorff (1920: 28) pointed out, Aristotle
is reported to have referred to Alexamenus as having composed “dialogues,” not “Socratic
dialogues.”
The Implicit Narrator 51

for the Theaetetus will thus be of utmost importance in our approach to the
other dramatic dialogues in the corpus.

2 The Theaetetus as a Test Case

The Athenian part of the Theaetetus starts with Socrates asking Theodorus of
Cyrene,11 whose schoolroom he is visiting, to tell him about promising young
Athenians among his students: this is the context in which Theaetetus12 makes
his appearance.
The opening, namely, Theaetetus being gradually introduced to Socrates,
thereby to the reader, closely resembles the openings of the narrated Charmi-
des and Lysis.13 Acceding to Socrates’ request, Theodorus starts with a lavish
description of one of his Athenian students, highly praising the youth’s char-
acter and intellectual prowess but also emphasizing his lack of physical beauty
(143e-144b and below, p. 100). Socrates asks whose son the boy is, but Theodor-
us has forgotten the father’s name: he simply points at the boy himself, who
at that very moment enters the stage in the company of two other students. “I
know the youth,” Socrates says, “but I do not know his name.” Like the Socrates
of the Lysis, the Socrates of the Theaetetus is better informed about his new
acquaintance’s father than about the boy himself: “He is the son of Euphronius
the Sunian, who was himself an eminent man.” At this point, Theodorus at
last supplies the name of the person to whom the dialogue is dedicated: “The-
aetetus, Socrates, is his name” (144c5–d1). Plato uses the zooming-in technique,
that is, releasing the relevant information by gradually focusing the narrator’s
attention on the principal interlocutor, exactly as in the dialogues with an ex-
plicit narrator: Socrates focalizes Theaetetus by asking questions (above, p. 33).
As in the narrated dialogues again, Socrates’ role as the focus of perception
is not restricted to the exposition. As the dialogue proceeds, Socrates commu-
nicates to the reader his own feelings and those of the others as he perceives
them. Thus he speaks freely about “a terrible passion” (tis erōs deinos) for dia-
lectical exercise that has taken hold of him (169c1) and of his feeling of awe

11 Geometer, the tutor of Theaetetus and according to some sources of Plato as well; appears
also in the Sophist and the Statesman. See further Nails 2002: 282–83.
12 A brilliant mathematician on whose work Books 10 and 13 of Euclid’s Elements are appar-
ently based; the lunar crater Theaetetus is named after him. Appears also in the Sophist
and the Statesman. Cf. Nails 2002: 274–78.
13 Cf. Morgan 2003:104: “Socrates’ narration in the Charmides is precisely the narrative for-
mat we would have encountered in the Theaetetus if Euclides had not removed the nar-
ratorial comments.” On the Charmides and the Lysis see above, Chapter 2.
52 Chapter 3

(aischunomai) toward Parmenides (183e4–5); he confesses that he is angry

(duscherainō) with what he calls his own “stupidity and tiresome garrulity,”
which has reduced to aporia what formerly seemed a winning definition of
knowledge (195c1–7). He not only diagnoses Theaetetus with spiritual preg-
nancy but also discerns that the boy himself is aware of his condition (151b7–8),
and he registers Theodorus’ uncharacteristic eagerness to proceed with the
discussion of perception and flux, thereby drawing the reader’s attention to
the subject (181b8).
The two other characters also focalize on several occasions, Theaetetus hesi-
tantly (157c4–6 “I cannot make out, Socrates, whether you are giving your own
opinion or only wanting to draw me out”), Theodorus much less so (177c3–4
“I would rather have the digressions [sc. than the argument], for at my age I
find them easier to follow”). The macro-focalization, however, is consistently
implemented by Socrates,14 and it is Socrates again who, in a manner charac-
teristic of Plato’s narrators, accounts for his own movements in the dialogue’s
closure:

And now I have to go to the porch of the King, where I am to meet Me-
letus and his indictment. Tomorrow morning, Theodorus, I shall hope to
see you again at this place.15

To recapitulate, in the Theaetetus Plato does not just show us how he sup-
pressed the dialogue’s narrator and rendered him implicit as a result: he also
identifies the implicit narrator by the same means by which his explicit nar-
rators are identified. In spite of the fact that the Theaetetus’ narrator is sup-
pressed, the signals of narratorial authority appearing in the text allow us to
construct him as a fully operational factor. To put it as simply as possible, the
narrator is he who focalizes the dialogue in its entirety: no alternative focus
of perception is sustained throughout the dialogue. To find out whether this
model of an implicit narrator is generally applicable, let us cast a closer look at
the other dramatic dialogues, starting with the earlier ones.16

14 On macro-focalization see above, p. 42.

15 Tht. 210d1–4; cf. also the last sentences of the Protagoras and the Symposium (quoted
below, pp. 54, 79).
16 The order of the so-called earlier dialogues in this chapter does not assume the order of
their composition. On the relative chronology of the so-called middle and late dialogues
see Chapter 5.
The Implicit Narrator 53

3 Bifocality or a Single Focus of Perception?

(the Euthyphro, Crito, Menexenus vs. the Ion and Hippias Maior)

If Plato indeed conceives the dramatic dialogues as narratives with an implicit

narrator, each of them should reasonably exhibit a clearly discernible single
focus of perception. However, in plain contradiction to what we have observed
in the narrated dialogues and the Theaetetus, three dramatic dialogues do not
conform to this pattern.

3.1 The Euthyphro

Socrates and the diviner Euthyphro meet in the Porch of the King Archon.
“Why have you left the Lyceum, Socrates? And what are you doing in the Porch
of the King Archon? Surely you cannot be concerned in a suit before the King,
like myself”? (2a1–4). Answering Euthyphro’s questions one by one, Socrates
explains that he is the prosecuted, not the prosecutor. He names his prosecutor
(2b8–10 “His name is Meletus, and he is of the deme of Pitthis”) and describes
his appearance (2b10–11 “He has a beak, and long straight hair, and a beard
which is ill grown”). Euthyphro does not remember the man. He asks Socrates
what he is charged with, and Socrates supplies him and the reader the required
information.
At this stage, Euthyphro is clearly the one who focalizes the story. It is he
who identifies the dialogue’s setting for us (“what are you doing in the Porch
of the King Archon?”), and it is Euthyphro again who, by asking questions, is
gradually zooming in the background information on Socrates’ impending
trial. Euthyphro expresses unconditional solidarity with Socrates in his plight
(3a7–8 “My opinion is that in trying to do you wrong [adikein] he [Meletus] is
simply aiming a blow at the foundation of the state”) and fully identifies with
him (3c3–4 “But they are jealous of all people of our kind; but we must be brave
and go at them”).
But when Euthyphro expresses his confidence that the affair will come to
naught and Socrates will win his case just as he himself will win his own (10e4–6),
Socrates takes over. “What is your suit, Euthyphro? Are you the pursuer or
the defendant?” (10e7–8). The procedure is repeated, but now it is Socrates
who gradually focalizes Euthyphro’s story by asking questions. As a result,
the zooming-in procedure is applied in both directions: Euthyphro focalizes
Socrates and is in turn focalized by him.
In the concluding part of the dialogue, the focus is again transferred to Eu-
thyphro, for it is he who informs us about his leaving the stage: “I am in a hurry
and must go now” (15e3–4). The remark may look trivial, unless we take into
54 Chapter 3

account that announcing the characters’ movements is normally the ­narrator’s

prerogative – for example, in Socrates’ words at the end of the Protagoras:
“Having said and heard that much, we went our way.”17

3.2 The Crito

A similar reciprocal focalization takes place in the Crito. In the dialogue’s open-
ing, we perceive the setting through the eyes of Socrates who, as we will learn
in a moment, awakens from sleep: “Why have you come at this hour, ­Crito? It
must be quite early?” (43a1). Crito informs Socrates and the reader of the hour
of the day at which the conversation takes place (43a4 “The dawn is break-
ing”), and at this point we learn that the action of the dialogue takes place in
the Athenian prison: “I wonder that the keeper of the prison would let you in”
(43a5–6). Socrates learns that Crito has been sitting for hours at his bedside
watching him sleep (43a9, b4), and finally is informed that the ship from Delos,
whose arrival signals that the execution is approaching, is expected this very
day (43c5–8).
From this point onwards, it is Crito who, in turn, learns from Socrates the
particulars of the latter’s recent vision and its relevance in determining cor-
rectly the execution date (44ab). In the subsequent confrontation, either in-
terlocutor argues in favor of his position in the debate as to whether Socrates
should escape from prison (Crito) or stay and face the execution (Socrates). As
in the Euthyphro, the focalization is reciprocal, with no unambiguous signs of
narratorial authority being attached to either.

3.3 The Menexenus

Socrates meets Menexenus,18 who has just left the meeting of the Council that
dealt with the election of the speaker for the annual funeral oration. The dia-
logue starts with a standard zooming-in, conducted by Socrates.

“Whence come you, Menexenus? Are you from the Agora?” “Yes, Socrates;
I have been at the Council.” “And what might you be doing at the Council?
…” “I went to the council chamber because I heard that the Council was
about to choose someone who was to speak over the dead. For you know
that there is to be a public funeral?” “Yes, I know. And whom did they
choose?” “No one; they delayed the election until tomorrow.” (234a1–b8)

17 Prt. 362a4. Also in the Theaetetus, the Cratylus, the Symposium, the Euthydemus, and the
Phaedrus, cf. above, with n. 15.
18 Menexenus is also one of Socrates’ interlocutors in the Lysis, see above, n. 15 to Chapter 2.
The Implicit Narrator 55

Having learnt from Menexenus that the speaker has not yet been elected,
Socrates intimates that an alternative funeral oration, presumably composed
by Aspasia, is already written. Now it is Menexenus’ turn to focalize whatever
information Socrates is ready to share with him:

“Do you think that you could speak yourself if there should be a neces-
sity, and if the Council were to choose you?” “That I should be able to
speak is no great wonder, Menexenus, considering that I have an excel-
lent mistress in the art of rhetoric – she who has made so many good
speakers, and one who was the best among all the Hellenes – Pericles, the
son of Xanthippus.” “And who is she? I suppose that you mean Aspasia?”
(235e1–8)

A series of additional questions put by Menexenus leads to Socrates’ delivery

of the oration, but not before he clarifies to the reader that no other audience
is present (236d1–2 “since the two of us are alone”), thus signaling that it is he
who focalizes the dialogue at this stage.
Socrates and Menexenus constantly switch places, which strongly suggests
that neither is invested with narratorial authority. This is probably why the
question as to whether the funeral oration delivered by Socrates is a genuine
piece of patriotic rhetoric or a parody on just such a piece has been disputed
up to the present day.19 However that may be, the Menexenus stands apart from
the Euthyphro and the Crito in that its bifocality does not reflect two conflict-
ing positions, and the encounter as a whole is hardly more than an introduc-
tion to the speech that follows.

...
In their recent study of the prologue of the Theaetetus, Eleni Kaklamanou and
Maria Pavlou contrast the narrated and the dramatic dialogues by character-
izing the latter as follows:

When a story is rendered entirely in direct speech, it is not mediated by a

narrative voice but through the voices of the characters, who both focal-
ize and speak. … Moreover, the reader is rendered a witness to narrated
events and is free to see and interpret things for himself without being

19 Cf. Guthrie 1975: 315: “In fact to reconcile the introduction with the content of the speech
has proved the crux of interpretation, and led to endless controversy.” See also Friedlän-
der 1964: 216–17.
56 Chapter 3

guided to understand them in a particular way, influenced by evaluative

comments from the narrator.20

The above analysis of the Euthyphro, the Crito, and the Menexenus fully agrees
with this assessment. The three dialogues are undeniably bifocal, and they act
as purely mimetic pieces on a par with drama or mime. Compare, for exam-
ple, the third episode of Euripides’ Medea. Aegeus arrives in Corinth where he
meets Medea. First, Medea focalizes Aegeus through a series of questions, thus
allowing the audience to identify the visitor and to learn about the circum-
stances of his sudden appearance in Corinth (665–88). Then Medea is in turn
focalized by Aegeus, who has caught sight of her glum countenance; through a
series of questions, he is gradually exposed to the story of her current predica-
ment (689–708). The strictly bifocal exchange provides an almost exact paral-
lel to the encounter of Socrates and Euthyphro in the Euthyphro.
Still, that is not to say that multiple perspectives in drama or mime cannot
form a hierarchy favoring some character perspectives over others.21 In both
the Crito and the Euthyphro Socrates’ perspective is manifestly more privileged
than his interlocutors’. But a single overarching perspective, the prerogative of
the narrator, is absent from these dialogues. The markers of focalization are
evenly distributed among the dialogues’ characters, which is blatantly at vari-
ance with what we have observed for the narrated dialogues and the Theaete-
tus. In other words, there is no implicit narrator in the Euthyphro, the Crito, or
the Menexenus. The situation however changes as soon as we turn to the other
dramatic dialogues.

3.4 The Ion

The opening of the Ion, a short and simply structured dramatic dialogue, runs
as follows:

“Welcome, Ion. Where have you come from? From your home city of
Ephesus?” “Not at all, Socrates, but from Epidaurus, from the festival of
Asclepius.” “Do the Epidaurians have also a contest of rhapsodes in the
god’s honor?” “O yes; and of other sorts of musical performers.” “Well, and
did you take part in the competition? And how did it go?” “I obtained the
first prize, Socrates.” “Well done; and I hope that you will do the same for
us at the Panathenaea.” (530a1–b3)

20 Kaklamanou and Pavlou 2016: 129.

21 On the hierarchy of perspectives in drama see Pfister 1988: 57–68. Cf. also Cohn 2001:
495–96.
The Implicit Narrator 57

Similar to what we have observed in the case of the Euthyphro, the Crito, and
the Menexenus, by asking questions the speaker gradually focalizes the missing
information and hands it over to the reader: Ion’s city of origin, his profession,
his winning of the first prize at the Epidaurian festival of Asclepius, the reason
for his coming to Athens.22 The procedure is minimalist but effective – that is
all we need to know for the dialogue to proceed. However, in contrast to the
Euthyphro, the Crito, and the Menexenus, the focalization in the Ion is strictly
one-sided. We perceive Ion through Socrates’ eyes, but when it comes to Ion’s
perception of Socrates, we suddenly have a blind spot: there is no indication of
how Socrates is perceived by Ion. The rhapsode’s capacity for focalization has
been “switched off.”

3.5 The Hippias Maior

Authentic or not, the Hippias Maior displays a narrative strategy closely similar
to what we witnessed in the case of the Ion. The dialogue opens with Socrates
meeting the famous sophist Hippias: “It is Hippias, the beautiful and wise!
What a long while it is since you came to anchor at Athens!” (281a1–2). Hippias
explains that he was too busy on diplomatic missions on behalf of his native
Elis. A series of questions and answers follows, until Socrates, and with him
the reader, eventually concentrate on the dialogue’s main subject, the defini-
tion of the beautiful. In the course of the discussion we, again together with
Socrates, are further informed that, following the request of Eudicus the son
of Apemantus,23 Hippias promised to repeat in Athens (“in the schoolroom of
Pheidostratus”) the discourse he had recently delivered in Sparta (286b).
At one point Socrates issues a brief reference to Hippias’ sumptuous attire
(291a6–7 “you who are so beautifully clad, so beautifully shod”); on another,
he remarks that his interlocutor gets angry whenever their opinions diverge
(302a). Hippias’ behaviour throughout the dialogue amply confirms this as-
sessment; yet, just as with Thrasymachus in Republic 1, Socrates mediates and
eventually neutralizes everything Hippias says about Socrates and his argu-
ments (cf. below, pp. 107–108). Consider, for example, Socrates’ reaction to Hip-
pias’ sharp criticism of his method:

But your frequent admonitions are a help to us. This time, for example,
before these admonitions from you about the stupid way we operate,
shall I make a still greater display, and tell you what we had in mind about
them? Or not tell? (301c6–d1; tr. P. Woodruff)

22 Nothing is known of Ion beyond what we learn from this dialogue; see Nails 2002: 175.
23 On Eudicus see below, p. 63.
58 Chapter 3

As a result, Socrates emerges as the sole focus of perception, the one through
whom his opponent’s opinions are filtered.

...
As distinct from what we observed in the case of the Euthyphro, the Crito, and
the Menexenus, in the Ion and the Hippias Maior the story is not mediated
“through the voices of the characters, who both focalize and speak,” and the
reader is not given the option “to see and interpret things for himself without
being guided to understand them in a particular way” (Kaklamanou and Pav-
lou, with n. 20 above). All options for perceiving the conversations between
Socrates and Ion and between Socrates and Hippias differently from how
Socrates perceived them are blocked. As a corollary, the reader is also deprived
of the opportunity to empathize with Socrates’ opponents. Indeed, even if they
often express themselves in an awkward and self-serving way,24 both Euthy-
phro and Crito genuinely care for Socrates, a feature that humanizes them and
renders the treatment they receive at Socrates’ hands much less devastating
than what we observe in the case of Ion and Hippias.25
The comparison of the Euthyphro, the Crito, and the Menexenus on the one
hand with the Ion and the Hippias Maior on the other demonstrates that the
Theaetetus is neither the first nor the only dramatic dialogue in which the nar-
ratorial authority has been delegated to the bearer of a single focus of percep-
tion to produce the effect of implicit narrator. But this comparison also shows
that Plato’s propensity for unilateral focalization should not be taken for grant-
ed. This must have been a deliberately chosen and, as we shall see immediately,
a consistently pursued narrative strategy that the Theaetetus’ thematization of
an implicit first-person narrator justified retroactively. This gives us sufficient
reason to proceed with examination of the other dramatic dialogues along the
same lines.

4 An Implicit Narrator-Hero (the Cratylus, Meno, Phaedrus, Laws)

In the Ion and the Hippias Maior, as well as in the Athenian part of the Theaete-
tus Socrates the implicit narrator is also the main character. In narratological

24 See esp. Cri. 45d8–46a4, with Beversluis 2000: 59; Danzig 2010: 80.
25 See Beversluis 2000: 163, quoting Jowett: “whatever his intellectual and personal short-
comings, Euthyphro ‘is not a bad man.’” On a defense of Crito and the common-sense
validity of his arguments see Beversluis 2000: 59–74.
The Implicit Narrator 59

terms, he is an autodiegetic narrator, or narrator-hero (above, p. 17). As we shall

see in this section, this is also the role he plays in the Cratylus, the Meno, and
the Phaedrus, while the Laws introduces an implicit autodiegetic narrator who
is not Socrates. Let us examine these dialogues one by one.

4.1 The Cratylus

Hermogenes invites Socrates to join in a discussion on the correctness of
names that Hermogenes and Cratylus led before his arrival.26 The dialogue be-
gins thus:

“Suppose that we make Socrates a party to the argument?” “If you please.”
“I should explain to you, Socrates, that Cratylus here has been arguing
that there is a correct name (onomatos orthotēta), supplied by nature, for
each one of the existing things.” (383a1–5)

As the conversation proceeds, Socrates the newcomer assumes the role of the
leading interlocutor and the moderator of the discussion.27
No zooming-in occurs in the opening of the Cratylus. As already mentioned,
the dialogue starts “with the second act” (p. 49), that is, in the middle of an on-
going discussion. Until he hears Hermogenes’ explanation, Socrates is unaware
of the subject of the discussion led before his arrival (cf. Pythodorus’ ignorance
in the Parmenides). Socrates’ lack of knowledge is shared by the reader, which
means that the scope of the dialogue is limited by Socrates’ personal experi-
ence (cf. above, p. 43, on external focalization).
The only reference to the setting, which emerges at the very end of the
­dialogue, is also by Socrates:

Then, another day, my friend, when you come back, you shall give me a
lesson; but at present, go into the country, as you are intending, and Her-
mogenes will set you on your way … (440e3–5)

As we saw above, such registering of the characters’ intentions and movements

is one of the narrator’s prerogatives. Since Socrates is also the one whose ex-
perience circumscribes the scope of the narrative, there seems to be sufficient

26 Hermogenes of Alopece, a member of the Socratic circle, was among those present at
the death of Socrates (Phd. 59b7); Hermogenes was Xenophon’s main source for the last
years of Socrates’ life. Cratylus son of Smicrion, a follower of Heraclitus, was one of the
philosophical influences on the young Plato (Aristot. Meta. 987a32–b1). See further Nails
2002: 162–64, 105–106.
27 On Socrates’ metanarrative comments in this dialogue see below, n. 27 to Chapter 6.
60 Chapter 3

reason to conclude that Socrates is envisaged as the dialogue’s focus of percep-

tion and accordingly its implicit narrator.

4.2 The Meno

In the Meno, the delayed disclosure of the fact that Anytus, in whose house
the encounter between Meno and Socrates takes place, is also present at
the conversation,28 serves as a litmus test for who should be considered the
dialogue’s implicit narrator. Precisely in mid-dialogue Socrates suddenly
says: “Here we fortunately have sitting by us Anytus, whom we should make
a ­participant in our enquiry” (89e9–10).29 Socrates proceeds with a detailed
account of Anytus’ father, education, and social standing (90ab). Anytus’ house
guest Meno is hardly in need of this kind of information: Socrates is introduc-
ing Anytus to the readers of the dialogue.
An animated discussion follows, and again it is Socrates who registers Any-
tus’ emotional state as he perceives it: “O Meno, it seems to me that Anytus is in
a rage.”30 This matches the way Socrates registers his interlocutors’ emotional
responses in the dialogues where his narratorial authority is explicit. Compare
the following passages from Plato’s narrated texts: “It seemed to me that he
[Critias] became angry with him” or “It seemed to me that Protagoras was get-
ting ruffled and excited.”31 Both are unambiguous examples of the narrator’s
text that validate Socrates’ narratorial authority in the dialogues in which they
appear.
Like the Cratylus, the Meno begins in medias res, with no description of
the setting or zooming-in. Nevertheless, whatever background information
there is, it is invariably supplied by Socrates. This concerns not only Socrates’
introduction of Anytus in the middle of the dialogue but also his references
to Meno’s Thessalian background and his studies with Gorgias that emerge in
the dialogue’s opening (70a5-b5), and as the dialogue proceeds his acknowl-
edgement of Meno’s physical beauty (76b5). This also concerns Socrates’ an-
nouncement, typical of Plato’s narrators (above, with n. 15), of his leaving the

28 Meno the Thessalian, represented here as a youth, was one of the leaders of The Ten
Thousand, the army of Greek mercenaries at the service of Cyrus the Younger immortal-
ized by Xenophon in the Anabasis. Anytus was a prominent Athenian politician and one
of Socrates’ prosecutors; see further Nails 2002: 204–205, 37–38.
29 Ἄνυτος ὅδε (“Anytus here”) at 89e10, 90a1. Contrary to Nails 2002: 38, Anytus is presented
as attending the conversation to its end, see Ἄνυτος ὅδε and τόνδε Ἄνυτον at 99e2 and
100b8 respectively.
30 Meno 95a2 μοι δοκεῖ χαλεπαίνειν, echoed by Meno at the end of the dialogue, see 99e2
καίτοι ἴσως Ἄνυτος ὅδε σοι ἄχθεται λέγοντι. On focalization by characters see above,
p. 42.
31 Chrm. 162d2 ἀλλά μοι ἔδοξεν ὀργισθῆναι αὐτῷ; Prt. 333e2–3 μοι ἐδόκει ὁ Πρωταγόρας ἤδη
τετραχύνθαι τε καὶ ἀγωνιᾶν.
The Implicit Narrator 61

stage at the end of the dialogue (100b7). All foregoing indicate unequivocally
that Socrates is envisaged here as focalizing the dialogue in its entirety. This
puts him in the position of the bearer of the narratorial authority in the Meno.

4.3 The Phaedrus

Just as his acknowledgment of Anytus’ presence in the Meno points firmly at
Socrates as the dialogue’s implicit narrator, Socrates’ visualization of the pasto-
ral landscape by the Ilissus River unambiguously indicates who should be seen
as the implicit narrator of the Phaedrus.

By Hera, a fair resting place! Here is this lofty and spreading plane-tree,
and the agnus is high and clustering and, being in full blossom, it seems
to spread the greatest fragrance over the place; and the stream which
flows beneath the plane-tree is deliciously cold to the feet. Judging from
the figurines and images, this must be a spot sacred to Achelous and
some Nymphs. How delightful is the breeze – so very sweet; and there is
a sound in the air shrill and summerlike which makes answer to the cho-
rus of the cicadas. But the greatest charm of all is the grass, like a pillow
gently sloping to the head. (230b2–c4; cf. 258c6–259a2)

From the moment he – and we – first cast a look at the plane-tree under which
the conversation will take place,32 Socrates makes us perceive the dialogue’s
scenery exactly as he does – even more so as he rarely leaves the city and is
therefore, like us, new to the landscape he describes (230c6–d3). Small wonder
that his description involves all the senses – vision, hearing, smell, and touch.
But Socrates’ role as the dialogues’ focus of perception is not confined to
the descriptions of landscape. From the moment he focalizes Phaedrus in the
dialogue’s opening,33 Socrates continually visualizes not only the setting of the
Phaedrus but also the behavior of the characters and communicates this infor-
mation to the reader. He describes Phaedrus hiding under his cloak the manu-
script of Lysias’ speech (228d6–7 “but you must first of all show to me what you
have in your left hand under your cloak”); Phaedrus’ delight with it (234d3–5
“I observed you while reading to be in an ecstasy”); his own covering of his head
before his first speech (237a4 “I will veil my face”); his nearly crossing the Ilis-
sus in order to return to Athens after the speech has ended (242a1 “I will cross
the river and make my way home”) and stopping at the last moment (242b8–9

32 See Phdr. 229a9–b2, Phaedrus to Socrates, “Do you see the tallest plane-tree in the distance?
… There are shade and gentle breezes and grass on which we may either sit or lie down.”
33 Phdr. 227a1–3 “My dear Phaedrus, where you come from and where are you going?” “From
Lysias the son of Cephalus, and I am going to take a walk outside the wall.”
62 Chapter 3

“As I was about to cross the stream the usual sign appeared to me”); his de-
livering his second speech with his face uncovered (243b6–7 “And this I will
attempt, not as before, veiled and ashamed, but with my head bare”). These
and other descriptions reach the reader through the mediation of Socrates, the
only character in the dialogue who focalizes it from beginning to end.34

4.4 The Laws

In the Laws, a dialogue from which Socrates is absent, the narrative is focalized
through the Athenian Stranger (Plato in disguise?), who is also the dialogue’s
leading interlocutor. He begins with a series of questions that allow the reader
to absorb the information concerning the identities of his two interlocutors,
Cleinias the Cretan and Megillus the Lacedaemonian.35 Further on, he makes
it clear that the conversation takes place in Crete and that its participants are
old men:

On our way we can pass the time pleasantly in talking about constitu-
tion and laws, for I am told that the distance from Cnosus to the cave
and temple of Zeus is considerable; and doubtless along the way there
are shady places under the lofty trees, which will protect us from this
­scorching sun. Being no longer young, we may often stop to rest beneath
them, and get over the whole journey without difficulty, beguiling the
time by conversation. (625a6–b7)

In an unmistakably Phaedrus-oriented passage, Cleinias completes the picture

by supplying the guests from abroad with additional information:

Yes, Stranger, and if we proceed onward we shall come to groves of cy-

presses, which are of rare height and beauty, and there are green mead-
ows, in which we may repose and converse. (625b8–c2, cf. 229a9–b2)36

Similar to the Phaedrus again, on two occasions the Athenian Stranger makes
the reader aware of the time of the day.37 Sparse as they are, the markers of

34 Phaedrus’ contribution is his revealing that not only Socrates but also he himself is bare-
foot (229a4–5) and his referring to the time of the day at 242a3–5 and 279b5–6 (at 258e6–
259a7, Socrates does this). See above, p. 42, on focalization by characters.
35 Both are probably fictional. Nails 2002: 197–98, is ambivalent as regards the historicity of
the latter.
36 Cf. Phdr. 229a9–b2, quoted in n. 32 above.
37 Laws 722c7–9 “All this time, from early dawn until noon, have we been talking about laws
in this charming retreat” and 811c7–8 “the words which we have spoken from early dawn
until now.”
The Implicit Narrator 63

focalization appearing in the Laws leave no room for doubt as to the identity of
the bearer of narratorial authority in this dialogue.
To recapitulate, the Cratylus, the Meno, the Phaedrus, and the Laws, along
with the Ion, the Hippias Maior, and the Athenian part of the Theaetetus dis-
cussed in the previous sections, are dramatic dialogues whose implicit narra-
tor is conceived of in the same way as the explicit narrator of the Charmides,
the Lysis, and the Republic, namely as a character who focalizes the dialogue
in its entirety. As in the Charmides, the Lysis, and the Republic again, these
dialogues’ implicit narrator is also their leading character (the autodiegetic
narrator, or narrator-hero). In other dramatic dialogues, a different model of
narrator is adopted.

5 An Implicit Narrator-Observer
(the Hippias Minor, Laches, Gorgias, Philebus)

Although Socrates appears in all Plato’s dialogues except the Laws, this does
not mean that he should a priori be considered the implicit narrator of every
single dramatic dialogue at our disposal. Just as in the narrated Parmenides, in
the dramatic Hippias Minor, Laches, Gorgias, and Philebus Socrates acts as a
character in a narrative focalized by someone else.

5.1 The Hippias Minor

As in the Hippias Maior, this dialogue describes an encounter between Socrates
and the renowned sophist Hippias. Nevertheless, it is the otherwise unknown
Eudicus son of Apemantus38 who opens the conversation:

Why are you silent, Socrates, after the magnificent display which Hippias
has been making? Why do you not either refute his words, if he seems
to you to have been wrong in any point, or join with us in commending
him? There is the more reason why you should speak, for we, who may
fairly claim to take part in a philosophical discussion (tēs en philosophiai
diatribēs), are now alone. (363a1–5)

Eudicus initiates the encounter and negotiates its terms (363a-c, cf. 364
b, 373a). He also describes the conversation’s setting (363a3–4 “we are now

38 Eudicus (whose father Apemantus is presented at 363b1–5 as an interpreter of Homer) is

referred to in Hp. mai. 286c as the man who invited Hippias to perform in Athens. Eudicus
is attested in these two dialogues alone; see Nails 2002: 146.
64 Chapter 3

alone,” etc.).39 Finally, he mediates between the antagonists at the critical stage
of the discussion (373ac). On the other hand, in contrast to his behavior in the
other dramatic dialogues treated so far, the Socrates of the Hippias Minor is not
presented as focalizing the conversation. The only occasion when he comes
fairly close to that is his remark “but now that there are not so many of us, and
my friend Eudicus bids me ask, I wish you would tell me …” (364b8–9): this not
only echoes Eudicus’ words to the same effect (above) but also points again to
Eudicus as a mediator. On one occasion, Socrates is focalized by Hippias, but
even this is done through Eudicus’ mediation: “But then, Eudicus, Socrates is
always troublesome in an argument and he looks as if he is meaning some
mischief” (373b4–5). With all this taken into account, it seems reasonable to
conclude that the dialogue’s focus of perception is Eudicus and that the dia-
logue is delivered to the reader through his mediation.
The Hippias Minor is a short dialogue (slightly more than 12 Stephanus
pages). The presence of the third speaker is unusual for the dialogues of this
category. In view of this it seems likely that the main reason for introducing Eu-
dicus was to prevent Socrates from being too closely associated with the para-
doxical conclusion he reaches at the end of the dialogue, namely that the good
man is the one who voluntarily does wrong (376b4–6). Socrates’ conclusion
is a reductio ad absurdum of the sophistic thesis that virtue is a skill (technē)
like any other.40 But Eudicus misses the point, and as later attempts to present
the dialogue as spurious owing to its content show, successfully communicates
his perplexity to the readers. This is largely due to his impartiality, by virtue
of which the opinions of the two opponents are presented as being of equal
weight. We are very close here to the bifocality of the Euthyphro, the Crito, and
the Menexenus (see further below, p. 109).
As we have seen, Eudicus is a mediating figure. Mediating characters in-
troduced in dramatic texts to create an overarching perspective have been
thoroughly discussed by Manfres Pfister in his authoritative analysis of drama.
Pfister’s example of such a character is the Singer in Brecht’s The Caucasian
Chalk Circle, about whom he comments as follows: “This figure is placed above
the action and the figures that he reports on, and the parable is presented as if
it had been produced by him.”41 This comes very close indeed to the position
of Eudicus in the Hippias Minor.

39 Echoed by Socrates at 364b8–9 (below). Cf. Menex. 236d1–2, “since the two of us are
alone,” and above, p. 55.
40 Similarly Kahn 1996: 118.
41 Pfister 1988: 59–60.
The Implicit Narrator 65

5.2 The Laches

Two prominent Athenian citizens, Lysimachus and Melesias, have invited two
distinguished generals, Nicias and Laches, to attend a public demonstration of
fighting in full armor. The purpose of the invitation was to seek the generals’
advice regarding the appropriate education of the sons of Lysimachus and Me-
lesias, more particularly to help them decide whether training in this military
art may contribute to the purpose. The conversation among the four begins
immediately on the show’s ending, with Lysimachus doingt most of the speak-
ing (178a-180b). “But why, instead of consulting us, do you not consult Socrates
here (Sōkratē de tonde) about the education of the youths?” Laches asks after
a while (180b7–c1). Although Socrates has not yet joined in the conversation,
Laches’ remark makes it clear that he is supposed to be present among the
people who have just watched the performance.
Lysimachus42 is an old man; he was a friend of Socrates’ father Sophronis-
cus, but he knows little about Socrates himself: this “initial ignorance” (above,
p. 21) closely parallels Socrates’ position vis-à-vis his younger interlocutors
both in the dramatic Theaetetus and the narrated Charmides and Lysis. Nicias
assures him that Socrates proved very helpful in recommending Damon as a
music teacher for his sons (180c8–d3). At this point, Lysimachus directly ad-
dresses Socrates. He praises Socrates’ father; he also recalls that his own son
and the son of Melesias often talk about Socrates, praising him most highly. Ly-
simachus’ son confirms that this is the same Socrates who was the object of the
youths’ admiration (180d4–181a3). Laches interferes again, to add that Socrates
was his companion in the retreat from Delium, where he distinguished himself
highly (181a7–b4). Lysimachus is impressed, and he asks Socrates to give his
opinion as to the educational value of fighting in full armor (181b5–c1). This is
where Socrates speaks for the first time (181d1–7).
As Charles Kahn has pointed out, “The Laches is unique among Socratic dia-
logues in that Socrates is not a participant in the opening conversation, and
several pages of text must be perused before he intervenes.”43 We saw indeed
that, although Socrates is apparently envisaged as being present among the
post-show crowd, the Laches starts as a private conversation between two pairs
of interlocutors, and Socrates joins it only after having been summoned – so
much so that Lysimachus finds it necessary to recapitulate for him the gist of
the previous discussion (181c7–9). This obviously precludes Socrates from be-
ing treated as the dialogue’s focus of perception.

42 Son of the great Aristides, disadvantageously compared with his father here (179c) and at
Meno 94a. See further Nails 2002: 194.
43 Kahn 1996: 151.
66 Chapter 3

Everything points in the direction that rather than through Socrates, the
Laches is focalized on him. Socrates is introduced to Lysimachus several times
– by Nicias, by Laches, by Lysimachus’ son, and it is through Lysimachus’ per-
ception that Socrates’ accomplishments are revealed. “That is very high praise
which is accorded to you, Socrates,” Lysimachus says, “by faithful witnesses and
for actions like those which they praise. Let me tell you the pleasure which I
feel in hearing of your fame; and I hope that you will regard me as one of your
warmest friends” (181b5–c1).44
Just like Socrates in the narrated dialogues, Lysimachus is the one who in
the dialogue’s opening makes the reader aware of the setting of the conver-
sation (178a1–2 “You have seen the exhibition of the man fighting in armour,
Nicias and Laches”). It is true of course that he withdraws to the background
at a relatively early stage: “I will therefore beg of you to carry on the proposed
discussion by yourselves; and I will listen, and Melesias and I will act upon your
conclusions” (189d1–3). Yet he re-emerges at the end of the dialogue, announc-
ing his decision to invite Socrates to his house to further consult him on the
education of the two youths (201b6–c3).
Lysimachus is thus the character through whom the dialogue is focalized in
its entirety. As we have seen, such macro-focalization is the prerogative of the
narrator (above, p. 42). It should be concluded, therefore, that Lysimachus is
the implicit narrator-observer of the Laches. At the same time, he is far from
being an impartial narrator, on a par with the narrator-observer we encoun-
tered in the Hippias Minor. Even though Lysimachus, like Eudicus of the Hip-
pias Minor, plays the part of arbiter, the dialogue’s conclusion leaves no room
for doubt about the identity of the speaker he sides with. This attitude will also
be found in the other dialogues featuring narrator-observer.

5.3 The Gorgias

Socrates45 and Chaerephon meet Callicles46 at the entrance to the latter’s house,
only to discover that they have just missed the display of rhetorical art by Cal-
licles’ guest, the great sophist Gorgias. This is the fault of Chaerephon, who de-
tained Socrates in the Agora. “’Tis no matter, Socrates,” Chaerephon interferes,

44 Cf. Kahn, 1996: 151: “By selecting an initial speaker, Lysimachus, to whom Socrates need to
be introduced, Plato has created an occasion for the two generals, Laches and Nicias, to
sing his praises as an outstanding citizen-soldier with a special interest in the education
of the young.”
45 The plot structure of the dialogue is analyzed in detail in Chapter 6.
46 An ambitious young politician whose historicity has been questioned. Socrates’ friend
Chaerephon also appears in the Charmides and the pseudo-Platonic Halcyon; he is men-
tioned in Plato’s Apology (20e8–21a8), in Xenophon and in Attic comedy; see further Nails
2002: 75–77, 86–87.
The Implicit Narrator 67

“for I can supply the remedy too. Gorgias is a friend of mine, and will treat us to
another display, now, if you want, or if not, later” (447b1–3).47
Callicles assures Socrates that Gorgias will answer his questions, and the
three enter the house. There they meet Gorgias and his disciple Polus, sur-
rounded, as we shall learn later, by a large audience. Socrates urges Chaere-
phon to ask Gorgias whether he is willing to take questions, but Gorgias is
tired, and Polus suggests that he answer Chaerephon’s questions himself. The
exchange between Chaerephon and Polus (447a-449c) serves as a prelude to
the dialogue, giving the first glimpse of the subsequent discussion.48 The mo-
ment the discussion between Socrates and Gorgias starts, Chaerephon with-
draws into the background.
Yet Chaerephon of the Gorgias is far from being merely an episodic char-
acter. Note that the entire dialogue is explicitly set as Chaerephon’s remedy
(pharmakon) for the presentation he and Socrates have missed. Chaerephon’s
special role twice comes to the fore again. First, at one of the structural breaks,
Gorgias, wishing to end the discussion, expresses his concern that the audi-
ence may lose patience if he and Socrates proceed. Chaerephon is he who ac-
knowledges the presence of the audience and registers their reaction:

You yourselves hear the audience cheering (thorubou … akouete), Gor-

gias and Socrates, which shows their desire to listen to you; as for myself,
Heaven forbid that I should have any business on hand which would take
me away from a discussion (logōn) so interesting and so ably maintained.
(458c3–7)

Chaerephon’s description of the audience’s reaction49 uses the same language

as the analogous descriptions by Plato’s explicit narrators. Thus, in the nar-
rated part of the Euthydemus, whose narrator is Socrates, we find: “At these
words those followers of Euthydemus and Dionysodorus, like a chorus at the
bidding of their director, laughed and cheered (anethorubēsan)” (276b6–c1);
“Then once more the admirers of the two men, in an ecstasy at their wisdom,
gave vent to another peal of laughter and cheers (ethorubēsan)” (276d1–2);

47 Tr. W. D. Woodhead. As Dodds 1959: 189–90, has pointed out, Chaerephon’s words prob-
ably allude to Euripides’ Telephus, whose protagonist was destined to find a remedy for his
wound in the very weapon that had inflicted it. See also above, n. 2, on a slightly different
construction of the setting.
48 Cf. Dodds 1959: 188: “a first lesson in dialectical method.” See also below, p. 138.
49 At Grg. 473e2–3 Socrates registers Polus’ laughter as a reaction to his argument, which
causes Polus to appeal to the audience (473e5 “ask the company”), thereby acknowledg-
ing its presence for the second time; this is reiterated by Gorgias at 506a8–b3; see further
below, p. 106. On focalization by characters see above, p. 42.
68 Chapter 3

“Up to now only Euthydemus’ partisans had cheered (ethorubēsan) at each

successive hit, but now even the columns of the Lyceum echoed their uproar
(ethorubēsan) and delight” (303b3–7). The Protagoras and the Symposium ex-
hibit a similar pattern: “When he had given this answer, the company cheered
him (anethorubēsan)”; “When Agathon had done speaking, Aristodemus said
that there was a cheer of those present” (anathorubēsai).50
The second time Chaerephon appears is again at one of the dialogue’s turn-
ing points. The discussion between Socrates and Polus has ended with Socrates’
counter-intuitive claim that doing wrong and not being punished is the worst
fate that can befall a person. “Tell me, Chaerephon, is Socrates in earnest or is
he joking?” Callicles bursts in, thus emphasizing again Chaerephon’s role as
the mediator. Note that Chaerephon’s position is authoritative enough to allow
him to confirm that this is indeed what Socrates meant to say: “I should say, Cal-
licles, that he is in most profound earnest; but you may well ask him” (481b6–9).
This is the point where the discussion between Socrates and Callicles, the last
and the most significant among the dialogue’s three verbal duels, begins.
Chaerephon is present on the stage all the time. He both inaugurates the
discussion and mediates between the speakers, and it is through him that we
first learn about the audience’s reaction to the discussion that takes place in
the dialogue. All this affords sufficient reason to conclude that the Gorgias is
focalized through Chaerephon and that it is Chaerephon who is envisaged as
its implicit narrator. To use Genette’s example, he is the one who should be
regarded as the dialogue’s “transparent (but inquisitive) Dr. Watson” (Chapter
2, p. 41).
As in the Laches, the person on whom the dialogue is focalized is Socrates.
Note indeed that the Gorgias culminates in Socrates’ programmatic defense
of his choice of a life of philosophical contemplation and in the expression of
his awareness that this choice may well cost him his life.51 This strongly sug-
gests that Chaerephon of the Gorgias belongs with those narrators-observers
– for example, Aristodemus of the Symposium or Phaedo of the Phaedo – who
are Socrates’ close associates and whose accounts are purported, among other
things, vividly to recreate his image and to justify the choices he made (see also
below, p. 166).

50 Prt. 334c7–8; Symp. 198a1–2. On the Euthydemus, the Protagoras, and the Symposium see
below, Chapter 4.
51 Grg. 521c-522e. Cf. Dodds 1959: 368: “In choosing his manner of life Socrates chose also
his manner of death, and chose it with his eyes open. It is not an argument calculated to
convince Callicles, and it does not convince him.”
The Implicit Narrator 69

5.4 The Philebus

The Philebus stands apart from the other dramatic dialogues in that it is fo-
calized neither through its protagonist Socrates nor through a bystander, but
through Socrates’ main interlocutor. Consider the opening words of the dia-
logue: “Look (horā), Protarchus, what kind of argument you are about to take
from Philebus at this moment, and what the other argument is which I main-
tain” (11a1–2). The phrase unambiguously points to Socrates’ addressee as the
dialogue’s focus of perception: Protarchus is the one who “looks.”52
Protarchus, a student of the hedonist Philebus,53 is supposed to defend his
teacher’s position that pleasure is the good as against Socrates’ initial thesis
that privileges knowledge alone. Protarchus’ gradual understanding of and
identification with the nuanced solution Socrates proposes provides most of
the action.54 Expressions of cognition and perception abound, all relating to
Protarchus.55
Protarchus is also the one who regularly alerts the reader to Philebus’ pres-
ence and his role in the discussion;56 makes us aware of the group of young
people present on the occasion and of their responses;57 does not let Socrates

52 He is also the character who, his youth notwithstanding, moderates the discussion, see,
e.g., 18a, 19a, 20a, 28b, 28d, 31b, 32d, 34c, 46b, 50e.
53 “Philebus’ boy,” see Phlb. 16b4–5, 36d6–7. He is also a student of Gorgias; see 58a7–8.
Nothing certain can be said of Philebus or Protarchus beyond what is found in Plato’s dia-
logue (Nails 2002: 238, 257), but it is hard to avoid the impression that, by making Socrates
address Protarchus as “son of Callias” (19b5, cf. 36d6–7), Plato had in mind the famous
admirer of the Sophists who acts as their generous host in the Protagoras (see n. 8 to
Chapter 2). Cf. also Meinwald 2008: 485–86.
54 See the comprehensive analysis by Dorothea Frede in D. Frede 1996; on the “double elen-
chus” of the Philebus see Davidson 2013 [1990]: 9–11.
55 Phlb. 45d6–7 “I understood (ἔμαθον) what you say, and see (ὁρῶ) that there is a great dif-
ference between them [moderate and immoderate pleasures]”; 49c6–7 “the mixture
of pleasures and pains is not as yet manifest (καταφανής) to me”; 51d4–5 “I am trying,
Socrates; do try in your turn to speak even more clearly (σαφέστερον)”; 51e6 “I perceive”
(κατανοῶ); 53e2 “Yet a third time I must say: speak more clearly (σαφέστερον), Socrates”;
53e8 “Your many repetitions make me slow to understand” (μόγις ἔμαθον); finally, 57c6–7
“We have got far enough, Socrates, to discern an astonishingly big difference between
kinds of knowledge in respect of precision (εἰς θαυμαστὸν διαφορᾶς μέγεθος εἰς σαφήνειαν
προεληλύθαμεν ἐπιστημῶν)” (tr. R. Hackforth, slightly adapted).
56 Phlb. 11c7–8 “Our excellent Philebus has refused to take part”; 12a9 “You, Philebus, have
handed over the argument to me”; 12b4–6 “And now, Socrates, either with or without
Philebus’ blessing, we will proceed with the argument”; 15c8–9 “Philebus, fortunately for
us, is not disposed to move, and we had better not stir him with questions.”
57 Phlb. 12b3–4 “We will be the witnesses (συμμάρτυρες) of your words”; 15c7 “all of us”;
16a4–5 “Considering, Socrates, how many we are, and that all of us are young men”; 20a4–
5 “If we are unable to do that, you must do it, as you have promised”; 29b1–2 “For truly
the storm gathers over us, and we are at our wits’ end (ὑπ’ ἀπορίας)”; 67b8–9 “And now,
Socrates, all of us agree that what you have been saying are the most truthful things of all.”
70 Chapter 3

depart before the argument has been completed (19de, 23b, 50cd, 67b). The last
words of the dialogue, registering Socrates’ leaving the stage, also are his. Pro-
tarchus is the dialogue’s focus of perception, the one through whose mediation
the reader copes with the argument’s difficulties and arrives at the solution.
This is emphasized by the fact that throughout the dialogue Protarchus acts as
a spokesman for a group of young people who attentively follow the discussion
(cf. n. 57 above).
The overall effect produced by this change of focus is that, rather than be-
ing dominated by a senior and more authoritative interlocutor – a situation
we are all too familiar with from other Platonic dialogues – Protarchus acts as
Socrates’ full-scale partner. As Dorothea Frede put it,

Although Socrates firmly directs the discussion, the investigation re-

mains a co-operative investigation throughout, and it is Protarchus who
confirms the final selection of the pleasures and the kinds of knowledge
appropriate for the good life.58

One can only wonder how Republic 2–10 would have looked like if there too the
focus of perception had moved in the direction of Glaucon or Adeimantus.59
It indeed seems reasonable to suggest that, like the other late dialogues, the
Philebus, with its peculiar characteristics unparalleled in the rest of the corpus,
was largely an experimental dialogue.60

6 Socrates as an Implicit Narrator-Observer

(the Sophist, Statesman, Timaeus, Critias)

Four late dialogues form a homogeneous group, in that they negotiate their
implicit narrator along the same lines.
− In the Sophist, Socrates meets Theodorus and Theaetetus on the day follow-
ing the conversation described in the Theaetetus, presumably at the same
place.61 Theodorus and Theaetetus bring with them a stranger from Elea,

58 D. Frede 1996: 223.

59 On Glaucon and Adeimantus as Socrates’ interlocutors see Blondell 2002: 199–214, and
below, Chapter 5.
60 Cf. Rowe 1996: 173, on Plato’s late dialogues: “It will be safer to explain variations between
one dialogue and another by reference to Plato’s continuing to experiment, or simply
making different choices of strategy, within his chosen form.” Cf. also Rutherford 1995:
276–81 and below, pp. 118–20.
61 Tht. 210d4; Soph. 216a1–2.
The Implicit Narrator 71

a disciple of Parmenides and Zeno, and Socrates asks him to examine and
adequately to define the sophist, the statesman, and the philosopher. The
Stranger agrees, and Socrates suggests he take Theaetetus as his interlocu-
tor. The rest of the dialogue is an exchange between the Eleatic Stranger and
Theaetetus. The dialogue is presented as the first installment in a trilogy.
− In the Statesman, the conversation started in the Sophist continues on the
same day.62 The cast of characters is the same as in the Sophist, but this time
the interlocutor of the Eleatic Stranger is Socrates’ namesake, the Younger
Socrates,63 who was a silent bystander in the Theaetetus and the Sophist.
− In the Timaeus, Socrates meets Critias and his guests Timaeus and
Hermocrates64 the day after he has delivered a discourse on what is sup-
posed to represent a version of the Republic. After Socrates’ recapitulation
of the main points of the yesterday’s discussion, the company decides to
continue it in a series of presentations. The first speaker is Timaeus the as-
tronomer, whose subject is the generation of the world.
− In the unfinished Critias, the conversation that started in the Timaeus con-
tinues. The argument is handed over to Critias.65
Who is the implicit narrator of the Sophist, the Statesman, the Timaeus, and
the Critias? As we have seen, the Sophist and the Statesman are intertextually
linked with the Theaetetus in that they are presented as its sequels.66 Since
Plato unambiguously indicates that the implicit narrator of the Theaetetus is
Socrates, this strongly suggests that Socrates is envisaged as an implicit nar-
rator in these two dialogues as well (cf. above, p. 7). In a similar manner, the
Timaeus and the Critias are presented as sequels to a version of the Republic, a
dialogue in which, again, Socrates is the first-person narrator.67

62 Plt. 258a3–4.
63 A member of the Academy, referred to in Aristotle’s Metaphysics (1036b 25) and in Plato’s
(spurious) Letter 11, see Nails 2002: 269.
64 Ti. 17b-19a, 26de. Timaeus of Locri in South Italy is known only from Plato’s Timaeus and
Critias. Hermocrates of Syracusae was a prominent Sicilian leader and the architect of the
Athenian defeat in Sicily in 415–413 bce. See further Nails 2002: 293, 161–62. The Critias of
the Timaeus and the Critias is probably not the same person as the Critias of the Charmi-
des and the Protagoras; on the latter see above, n. 13 to Chapter 2.
65 Cri. 106b6–7.
66 This does not mean that the Sophist and the Statesman were written immediately after
the Theaetetus. Cf. Kahn 2013: 94: “A significant change in style suggests that a consider-
able lapse of time may have occurred between the composition of these two dialogues
[the Theaetetus and the Sophist]. Nevertheless, the reappearance of Theaetetus as inter-
locutor in the Sophist is a clear reminder of continuity in this project.” See also the discus-
sion in M. Frede 1996: 145–48.
67 This would be equally true of the version of the Republic recapitulated at the beginning of
the Timaeus; see Ti. 20b1–2: “And therefore yesterday when I saw that you wanted me to
72 Chapter 3

As distinct from the Theaetetus and the Republic, however, Socrates plays no
significant part in the dialogues that are set as their continuations. Among Plato’s
late dialogues, only the Philebus features Socrates as the leading interlocutor.68
The other dialogues display a steady tendency to transfer this role to another
character – the Eleatic Stranger in the Sophist and the Statesman, Timaeus
and Critias in the eponymous dialogues.69 The culmination of this tendency is
found in the Laws, from which Socrates is absent.
In the Sophist, the Statesman, the Timaeus, and the Critias Socrates is pushed
into the background. After setting the stage for the forthcoming discussion, he
turns into a silent listener whose participation is no longer necessary for the
argument’s progress. This seems to imply that it was important for Plato to en-
sure Socrates’ presence even in the dialogues in which he plays no significant
part. The explanation that suggests itself is that this allowed him to position
Socrates as an eyewitness to the discussion and, consequently, as the dialogue’s
focus of perception. If correct, this would mean that Socrates’ position in these
dialogues is that of an implicit narrator-observer on whose personal experi-
ence the story delivered in the dialogue is based (compare Pythodorus’ posi-
tion in the Parmenides and Lysimachus’ position in the Laches).
In each of the four dialogues under review, signals of narratorial authority
abound. All are concentrated in the dialogues’ openings, and all unambigu-
ously point to Socrates. Thus, the Sophist begins with Socrates perceiving the
arrival of the three other interlocutors at the meeting place. Theodorus intro-
duces the Eleatic Stranger:

Here we are, Socrates, true to our agreement of yesterday; and we bring

with us a stranger from Elea, who is a disciple of Parmenides and Zeno,
and a true philosopher. (216a1–4)

Socrates’ reaction to the newcomer’s appearance further enhances the latter’s

status: “Is he not rather a god, Theodorus, whom, as in Homer, you bring to
us in the disguise of a stranger?” (216a5–6).70 The gradual introduction of the
Eleatic Stranger is of course yet another example of the zooming-in procedure,
widely attested in both narrated and dramatic dialogues.

describe the constitution of the State (τὰ περὶ τῆς πολιτείας διελθεῖν), I readily assented.”
For a concise but informative discussion of the relation of this version to Plato’s Republic
see Cornford 1937: 4–5. Cf. also Guthrie 1978: 245 with n. 1; Thesleff 1982: 251–52. On the
intertextual aspects involved see McCabe 2008: 109–1; Erler 2015: 95–100.
68 But not as the dialogue’s implicit narrator; see the previous section.
69 On the self-referential aspect of this strategy see Sedley 1995: 5–6.
70 For a discussion of the Stranger’s anonymity and lack of physicality see Blondell 2002:
318–26, esp. 323–24.
The Implicit Narrator 73

Socrates is also the one who establishes the terms of the discussion (216d2–
217a3) and proposes that the Stranger take Theaetetus as his interlocutor
(217d5–7, 218a1–5). As with Pythodorus of the Parmenides and with Socrates
of the Cratylus, here too the reader’s knowledge is limited to Socrates’ per-
sonal experience: before their arrival at the meeting, the Stranger, Theodorus
and Theaetetus had conversed on issues similar to those to be discussed, but
Socrates, and with him the reader, remain ignorant of the contents of their
conversation (217b4–6; cf. also above, p. 59). Having made all necessary prepa-
rations, Socrates withdraws into the background as a silent bystander (cf. 216d3
“I would listen to you with pleasure”).
Socrates’ role in the Statesman is equally restricted. At the beginning of
the dialogue he thanks Theodorus for introducing him to Theaetetus and the
Stranger, thereby making the intertextual connection with the Theaetetus and
the Sophist. Later he elaborates the idea that each of the two young interlocu-
tors of the Eleatic Stranger, namely, Theaetetus in the Sophist and the Younger
Socrates in this dialogue, is somehow related to him (257d1–258a6). He del-
egates his new-fangled kinsmen – his doubles, as it were71 – to represent him
in the discussions with the Eleatic Stranger, who is about to expound a new
brand of dialectic. However, Socrates will watch and listen, for both the Sophist
and the Statesman are explicitly staged as pieces of intellectual entertainment
Socrates has arranged for himself.
In the Timaeas and the Critias the idea of intellectual entertainment with
Socrates as the audience is even more pronounced. The opening of the Timae-
us is closely similar to that of the Sophist. Socrates greets his interlocutors –
Timaeus, Critias, and Hermocrates – who have promised to reciprocate his ear-
lier discourse on the State with discourses of their own. He eagerly anticipates
receiving “the guest-gift of arguments” (20c1 ta tōn logōn xenia) promised to
him: “And I, in return for my yesterday’s discourse, will now rest and be a listen-
er (hēsuchian agonta antakouein)” (26e7–8). At the moment of his withdrawal,
“the guest-gift of arguments” becomes “the feast of arguments”: “I see that I
shall receive in my turn a perfect and splendid feast of arguments. And now,
Timaeus, you, I suppose, should speak next, after duly calling upon the gods”
(27b7–9). At this point, the dialogue becomes a monologue, more precisely, a
lecture delivered by Timaeus.
In the unfinished Critias Socrates speaks only once, but this suffices to make
it clear that his part in the dialogue was supposed to be identical to the one he
had played in the Timaeus. After confirming the order of the speakers (Timae-
us, Critias, Hermocrates), he emphasizes his position of spectator with more
force and imagination than in any other dialogue:

71 On Theaetetus as Socrates’ double see Blondell 2002: 260–70.

74 Chapter 3

And now, friend Critias, I will announce to you the judgment of the the-
ater. They are of opinion that the last poet was wonderfully successful,
and that you will need a great deal of indulgence before you will be able
to take his place. (108b2–7)

The unfinished trilogy Timaeus – Critias – Hermocrates is treated here along the
lines of tragic competition whose only audience, or “theater,” is Socrates. The
theater metaphor not only encapsulates the role of implicit ­narrator-­observer
that Socrates plays in the four late dialogues but also presents the implicit nar-
rator in a new role of spectator. As we shall see in Chapter 7, this is far from
being merely a figure of speech.
As I have already pointed out, the Socrates of the Sophist, the Statesman, the
Timaeus, and the Critias differs from the other narrators-observers in that he
appears only in the opening parts of the dialogues. That is, although it is ex-
plicitly declared that he will be present throughout the conversation, Socrates’
presence is not punctuated, as it is in the other dialogues, by markers of focal-
ization running through the entire text. Implications of this new arrangement
will be discussed in Chapter 5.

7 Conclusions

Apart from the Euthyphro, the Crito, and the Menexenus, all Plato’s dramatic
dialogues are negotiated as narratives with an implicit narrator. The latter is
identified by the same features that characterize the explicit narrator of the
narrated dialogues. Plato’s implicit narrator, both narrator-hero and narrator-
observer, is (a) he who perceives and describes the dialogue’s setting: Socrates
in the Phaedrus, the Theaetetus, the Sophist, the Statesman, the Timaeus, the
Critias; Eudicus in the Hippias Minor; Lysimachus in the Laches; Chaerephon
in the Gorgias; Protarchus in the Philebus; the Athenian Stranger in the Laws;
(b) he who gradually focuses his attention on his principal interlocutor and
the subject of the conversation (the zooming-in technique): Socrates in the
Ion, the Hippias Maior, the Phaedrus, the Theaetetus, the Sophist; Lysimachus
in the Laches; (c) he who registers the characters’ physical appearance, move-
ments, and emotional responses: Socrates in the Hippias Maior, the Meno, the
Cratylus, the Phaedrus, the Theaetetus, the Sophist, the Statesman; Protarchus
in the Philebus; the Athenian Stranger in the Laws; (d) he whose personal ex-
perience circumscribes their scope of the narrative (explicitly demonstrated in
the Cratylus, the Laches, the Sophist; taken for granted in the other dialogues).
This is not to say that the signals of narratorial authority are evenly ­distributed.
They are especially dense in such dialogues as the Phaedrus and the Theaetetus
The Implicit Narrator 75

and rather thin in the Cratylus and the Statesman. But the same is true of the
dialogues with an explicit narrator. For example, it is not explicitly articulated
in the Charmides, the Lysis, and the Republic that their scope is delimited by
Socrates’ personal experience, whereas in the Parmenides Pythodorus’ account
has no description of the setting and does not employ the zooming-in: this,
however, does not affect the positions of Socrates and Pythodorus respectively
as the narrators of the dialogues in question. The evidence is cumulative, and
it attests unequivocally that the overwhelming majority of Plato’s dramatic
dialogues exhibit a single focus of perception coinciding with the position of
the explicit narrator in the narrated dialogues. This single focus of perception,
arising from the sum of the markers of focalization attributed to one character
to the exclusion of the others, allows us to isolate the template of the implicit
narrator in Plato’s dramatic dialogues.72
Like the explicit narrator of the narrated dialogues, the implicit narrator
of the dramatic dialogues is conceived as the first-person narrator who may
or may not be the dialogue’s leading character. Socrates the implicit narrator of
the Sophist, the Statesman, the Timaeus, and the Critias is merely an observer,
therefore of a different status from Socrates the implicit narrator-hero of the
Ion, the Hippias Maior, the Cratylus, the Meno, the Phaedrus and the Theaete-
tus, whereas the implicit narrators-observers of the Hippias Minor, the Laches,
the Gorgias, and the Philebus are persons other than Socrates.
Finally, again just as in the narrated dialogues, in the dramatic dialogues
the narrator’s text is often placed at the turning points of the discussion
(above, pp. 45–46). The prominent examples are Eudicus’ intervention in the
Hippias Minor, Socrates’ acknowledgment of Anytus’ presence in the Meno,
Chaerephon’s interventions in the Gorgias, and the landscape descriptions
in the Phaedrus (on the latter see below, pp. 126–27). On the other hand, the
Ion, the Laches, the Cratylus, and the Philebus are much less receptive to this
kind of test. Moreover, such late dialogues as the Sophist, the Statesman, the
Timaeus, and the Critias reduce the narrator’s text to the opening part of
the dialogue and proceed on their own, as it were. So the results are mixed,
but it is obvious by now that, although the narrator’s text is indeed frequently
placed at strategic positions in the dialogue, the practice is not universally ap-
plied. Accordingly, the location of the narrator’s text cannot be regarded as a
signal of narratorial authority. Implications of this conclusion for determining
the position of narrator in Plato’s dialogues will be discussed in Chapter 6.

72 Cf. Schmid 2008: 74, on implicit narrator in the third-person narrative: “Der implizit dar-
gestellte Erzähler ist ein Konstrukt, das aus den Symptomen des Erzähltextes gebildet
wird. Strenggenommen, ist er nicht anders als der Träger der indizierter Eigenschaften.”
 Chapter 4

The Explicit and the Implicit Narrator Combined:

Mixed Dialogues

1 Introducing the Mixed Dialogues

The so-called mixed dialogues differ from the two other categories in that here
Plato makes the primary narrative act manifest by dramatizing it. At the be-
ginning of each dialogue of this group the prospective narrator and narratee
are represented as engaged in what looks like an extradiegetic “real-life” en-
counter. As far as we can judge, this specific subgenre of Socratic dialogue was
Plato’s own invention, and Plato was the only Socratic writer who practiced it.
Generally speaking, presenting the frame story as standing for “the real” is,
as Peter Brooks has pointed out, a characteristic feature of frame narratives:

That is, in the frame-tale structure, the outer frame comes to represent
“the real” and movement from inner to outer tales suggests the move-
ment of reference, making real.1

It goes without saying that the “reality” introduced by the frame story is itself
fictional.
Plato’s frame stories are peculiar in that they assume dramatic rather than
narrated form. This is probably why their openings closely resemble the open-
ings of short dramatic dialogues.
− In the dramatic frame of the Protagoras, an unnamed acquaintance asks
Socrates to tell him and his friends about his recent meeting with Protagoras.
− In the dramatic frame of the Euthydemus, Crito asks Socrates to tell him
about his recent meeting with Euthydemus and Dionysodorus.
− In the dramatic frame of the Phaedo, Echecrates asks Phaedo to tell him
about Socrates’ final hours.
− In the dramatic frame of the Symposium, Socrates’ associate Apollodorus
complies with the unnamed acquaintance’s request to tell him about the
famous party in Agathon’s house.

1 Brooks 1992: 220. Cf. also Brooks 1992: 225–26, 235.

© koninklijke brill nv, leiden, ���9 | doi:10.1163/9789004390027_005

The Explicit and the Implicit Narrator Combined 77

− In the dramatic frame of the Theaetetus, Terpsion asks Socrates’ friend Eu-
clides to tell him about a conversation between Socrates and Theaetetus
years earlier.
Their dramatic form notwithstanding, the openings of the mixed dialogues are
not to be approached as non-narrative. As we have seen, Plato leaves us no
such classificatory option. Insofar indeed as he classifies any dramatic compo-
sition as “diegesis through mimesis” (Chapter 1), dramatic frames may not be
treated differently from his other dramatic texts. Namely, the figure of implicit
narrator should be considered here as well.
Let us return for a moment to the sequence of narrative levels observed in
the Parmenides, a multi-level narrated dialogue discussed in Chapter 2. The
narrative act of Pythodorus, an eyewitness to the conversation that is the dia-
logue’s subject, is intradiegetic by virtue of its being mediated by the narrative
act of his narratee Antiphon, and the narrative act of Antiphon is in turn in-
tradiegetic by virtue of its being mediated by the narrative act of his narratee
Cephalus. The latter is he who communicates the story to the extradiegetic
narratee, which renders his narrative act extradiegetic (above, p. 39). Thus
three intradiegetic levels are found in the Parmenides: Cephalus’ narrative act
produces the first intradiegetic level, Antiphon’s produces the second, and Py-
thodorus’ produces the third. The transition from one level to another is ac-
companied by a shift of narrative voice, so that one level’s narrator becomes
the next level’s narratee.
Let us now assume that the dialogue’s first intradiegetic level, Cephalus en-
counter with Glaucon and Adeimantus, is cast in direct rather than reported
speech. This would make the first intradiegetic level appear extradiegetic, os-
tensibly leaving us only with the narratives of Antiphon and Pythodorus, that
is, with a two-level narrative structure. This is precisely what happens in the
mixed dialogues.
The point can be further illustrated by comparing the narrated Parmenides
with the mixed Symposium. In the dramatic frame of the latter, Apollodorus
agrees to tell an unnamed friend what he heard from Aristodemus about Ag-
athon’s dinner party; this dramatic encounter is followed by Apollodorus’ re-
hearsal of Aristodemus’ narrative. The bottom-level narrator of the Symposium,
Aristodemus, is found in the same position as Pythodorus, the bottom-level nar-
rator of the Parmenides. Aristodemus’ narratee Apollodorus, who narrates the
story to an unnamed friend, is found in the same position as Pythodorus’ narra-
tee Antiphon, who narrates the story to Cephalus and his companions. Finally,
Apollodorus’ anonymous narratee is found in the same position as Antiphon’s
narratee Cephalus, the primary narrator of the Parmenides. Yet, the moment
the narrative frame becomes a dramatic one, its narrator becomes implicit:
78 Chapter 4

the Parmenides the Symposium

1st level Cephalus [Apollodorus’ friend]

2nd level Antiphon Apollodorus
3rd level Pythodorus Aristodemus

By dramatizing what is effectively a frame narrative, Plato – just as in the pro-

grammatic case of the Theaetetus discussed in Chapter 1 – economizes on the
narrative level that the introduction of such a full-scale frame narrative would
involve. That is, just as the embedded story of the Theaetetus is disguised as if
belonging to the first intradiegetic level (above, pp. 5–8), so too the dramatic
frame of this and the other mixed dialogues is disguised as if belonging to the
extradiegetic level, being in fact fully enclosed within the intradiegetic do-
main: it is, as simply as possible, pseudo-extradiegetic.2 This becomes immedi-
ately obvious if we take into account that the narratees of the dramatic frames
are invariably personalized as characters. To repeat Genette’s assessment of
such fictional narratees,

We, the readers, cannot identify ourselves with those fictive narratees
anymore than those intradiegetic narrators can address themselves to us,
or even assume our existence.3

But if they have no extradiegetic narrator, by whose agency will the dramatic
frames reach their implied readers? This, if anything, is why the figure of im-
plicit narrator should be introduced here as well.
Beyond the dramatic frame, the mixed dialogues for the most part proceed
as regular narrated dialogues. Their openings follow the same two-stage pat-
tern, successively introducing (i) the initial spatiotemporal situation and cast
of characters and (ii) the eventual spatiotemporal situation and cast of char-
acters (above, pp. 27–28).
− In the framed narrative of the Protagoras, Socrates starts narrating how (i)
the young Hippocrates came to his house early in the morning and how (ii)

2 See also below, p. 101. In his recent discussion of the Euthydemus and Protagoras, Collins 2015:
45–169, overlooks this point, referring to the dramatic frame in both dialogues as extradieget-
ic. The same is true of Halperin 1992: 98, who draws no distinction between the personalized
narratee of the Symposium and the dialogue’s implied reader.
3 Genette 1980: 260; cf. above, p. 18.
The Explicit and the Implicit Narrator Combined 79

they arrived in the house of Callias, where they found Protagoras and other
sophists; this is followed by Socrates’ report of the conversation that devel-
oped as a result of the encounter.
− In the framed narrative of the Euthydemus, Socrates narrates how (i) he
was sitting alone in the Lyceum dressing room and how (ii) Euthydemus,
Dionysodorus, and other interlocutors arrived and joined him; this is fol-
lowed by Socrates’ report of the conversation that developed as a result of
the encounter.
− In the framed narrative of the Phaedo, Phaedo narrates how (i) he and other
visitors assembled in the Athenian prison, and how (ii) Socrates was even-
tually left in the company of his close friends; this is followed by Phaedo’s
report of Socrates’ last conversation and his death.
− In the framed narrative of the Symposium, Apollodorus narrates how (i)
Aristodemus met Socrates who was on his way to Agathon’s house, and how,
(ii) on arriving there, they joined the party, where they met more people;
this is followed by Aristodemus’ report of a series of speeches delivered on
the occasion.
− In the framed pseudo-diegetic narrative of the Theaetetus, Euclides’ slave
reads his master’s account of how (i) Socrates met Theodorus the geometer,
and how (ii) the latter introduced to him his student Theaetetus on the lat-
ter’s arrival; this is followed by Socrates’ report of the conversation that de-
veloped as a result of the encounter.
More often than not, the dramatic frames of Plato’s mixed dialogues have no
closure.4 Only the Euthydemus and the Phaedo end precisely where they be-
gan, namely, with the frame conversations between Socrates and Crito and
between Phaedo and Echecrates, respectively. In the three other dialogues,
the closure corresponds to the opening of the framed story rather than to that
of the dramatic frame: Socrates and Hippocrates leave Callias’ house whither
they arrived at the beginning of the framed story (the Protagoras); Socrates
and Aristodemus leave the house of Agathon whither they arrived at the begin-
ning of the framed story (the Symposium); Socrates says goodbye to Theodorus
and Theaetetus whom he met at the beginning of the framed story and departs
for the Porch of King Archon (the Theaetetus).

4 Such incomplete framing has numerous counterparts in world literature. Wolf (2006: 187–88)
suggests that the concluding framing is frequently omitted in order to avoid anticlimactic
effects; he adduces Henry James’ The Turn of the Screw as an example. Cf. also Brooks’ (1992,
256–57 and 351 n. 8) discussion of the absence of a satisfactory closure in Joseph Conrad’s
Heart of Darkness.
80 Chapter 4

While the similarity between the narrated parts of the mixed dialogues
and the thoroughly narrated dialogues is undeniable, it would be incorrect to
approach the mixed dialogues as a more or less mechanical combination of
narrated dialogue and dramatic frame.5 In every dialogue of this group, the
presence of the dramatic frame is felt all along. In the Euthydemus and the
Phaedo the frame intrudes into the main narrative to produce metalepsis; in
the Symposium the frame narrator’s voice is heard throughout the narrated
story; and even in the relatively straightforward narrative of the Protagoras the
boundaries between the frame story and the narrative it encloses are signifi-
cantly transgressed (more below).6 This unambiguously demonstrates that the
mixed dialogues were conceived and executed as multi-level narratives.

2 The Protagoras

An unidentified acquaintance meets Socrates in the street. Learning that

Socrates has just left a meeting at which Protagoras and other illustrious soph-
ists were present, he is eager to know more about the encounter; Socrates com-
plies with the request and begins his account of the conversation that took
place in Callias’ house.

2.1 Frame Narrator (Implicit): An Unnamed Athenian

The dramatic frame7 of the Protagoras is unusually short (approximately one
Stephanus page), so that the regular signals of narratorial authority are quite
few here. Still, it is quite obvious that the dramatic frame of the Protagoras is
focalized through Socrates’ prospective narratee.
The dialogue begins: “Where do you come from, Socrates? And yet I need
hardly ask the question, for I know that you have been in chase of the beautiful
Alcibiades.” Socrates admits that he indeed arrives from a meeting at which
Alcibiades was present, and that the latter even helped him there, speaking in
his defense. “But shall I tell you a strange thing? I paid no attention to him, and

5 So Thesleff 1982: 207: “Moreover, pieces of dramatic dialogue seem in many cases to have
been secondarily attached to bodies of reported dialogue or speeches … Most instances of
dramatic ‘frame’ dialogues … can be explained as such secondary additions.”
6 The Theaetetus stands apart, in that in this dialogue the dramatic form replaces the narrative
at each successive level.
7 Plato’s narrated dialogues have no unnamed narrators. But Xenophon, for one, never names
his first-person narrators, who presumably belong with Socrates’ interlocutors or silent by-
standers. On Xenophon’s anonymous narrators see Gray 2004: 377–80, Danzig 2017: 132 n. 1;
on frame narrators see Chapter 1.
The Explicit and the Implicit Narrator Combined 81

several times I quite forgot that he was present.” The interlocutor is appalled.
Is it possible that Socrates has met in the city of Athens someone more beauti-
ful than the son of Cleinias? Socrates replies that he has indeed met someone
much more beautiful than Alcibiades, for the wiser is always the more beau-
tiful, and Protagoras, whom he has just met, is the wisest of all living men.
“What! Is Protagoras in Athens?” the other man exclaims, and the “zooming-in”
is complete (309a1–d3). Through the perception of the unnamed interlocutor
we come to learn the identity of the man whom Socrates has recently met.
Later on, it is again the unnamed interlocutor who sets the scene for
Socrates’ narration: “Why shouldn’t you sit down here and tell us what passed?
My slave will give up his place to you” (310a2–4).8 Socrates’ prospective nar-
ratee thus acts in the Protagoras’ dramatic frame as the one who introduces
the setting of the frame and of the following narration – so much so that he
and the narrative situation in which he is involved will re-emerge several times
in the course of the framed story (see below, on metalepsis). These signals of
narratorial authority substantiate the assumption, made in Section 1, that by
virtue of his position as Socrates’ narratee the unnamed interlocutor should be
identified as the implicit extradiegetic narrator of the dramatic frame (the first
intradiegetic level) and, accordingly, of the dialogue as a whole.

2.2 Narrator-Hero (Explicit): Socrates

In the narrated part of the Protagoras (the second intradiegetic level), Socrates
the narrator little by little introduces us to the story’s setting and characters.
Every detail is vividly vizualized:

When we entered, we found Protagoras taking a walk in the covered

court; and next to him, on one side, were walking Callias, the son of Hip-
ponicus, and Paralus, the son of Pericles … and Charmides, the son of
Glaucon.9 On the other side of him were Xanthippus, the other son of
Pericles, Philippides, the son of Philomelus; also Antimoerus of Mende …
A train of listeners followed him … Nothing delighted me more than the
precision of their movements: they never got into his way at all; but when
he and those who were with him turned back, then the band of listeners
parted regularly on either side; he was always in front, and they wheeled
round and took their places behind him in perfect order. (314e3–315b8)

8 The use of the pronomen “us” indicates that, just like Apollodorus’ unnamed interlocutor in
the Symposium (below), the unnamed interlocutor of the Protagoras represents a larger audi-
ence. Cf. Thesleff 1967: 40.
9 On Callias and Charmides see nn. 8 and 13 to Chapter 2.
82 Chapter 4

“If the Protagoras conveyed no philosophical lesson,” W.K.C. Guthrie wrote,

“it would remain a superb work of literature. It must be read, not just read
about.”10 A brilliant travesty of Homer’s account of Odysseus’ visit to the Un-
derworld follows, introducing two more celebrities, the sophists Hippias and
Prodicus, each surrounded by his own circle of admirers (315b9–316a2). The
introduction concludes with the entrance of Alcibiades and Critias, which em-
phatically brings back the dialogue’s dramatic frame:

No sooner had we entered than there followed us Alcibiades the beauti-

ful, as you say, and I agree with you; and Critias the son of Callaeschrus. On
entering we stopped a little, in order to look about us, and then walked up
to Protagoras, and I said: “Protagoras, my friend Hippocrates and I have
come to see you.”11

In the body of the dialogue, Socrates the narrator describes Hippocrates’ laugh-
ter and blushes (310d, 312a), the laughter and cheers of Protagoras’ supporters
(334c, 339d-e), the audience’s approval (320c, 337c, 338b, 340d), Protagoras’
pretentiousness, anger, dissatisfaction, and embarassment (333d, 333e, 335ab,
338e, 348c), Prodicus’ laughter (358b), as well as his own emotional responses
(see, e.g., 339d-e, quoted below).12 As with the other first-person narrators, we
depend entirely on Socrates for the amount of information we are exposed
to – see, for example, 336d7: “When Alcibiades had done speaking, someone –
Critias, I believe – went on to say …” As we have seen, all these are signals of
narratorial authority, applied in the narrated and the dramatic dialogues alike.

2.3 Metalepsis
On three occasions, the dramatic frame intrudes into the narrated world to
produce frame-breaking, or metalepsis. In the passage just cited, Socrates the
narrator directly addresses his narratee, quoting the remark by the latter about
Alcibiades made at a different narrative level (above, p. 80). Later on, when de-
scribing his distress at the moment he felt defeated by Protagoras’ eloquence,
Socrates addresses the narratee again. He does so in a way that closely resem-
bles what we observed in the Charmides (above, p. 34):

Many of the audience cheered and applauded this. And I felt at first giddy
and faint, as if I had received a blow from the hand of an expert boxer,

10 Guthrie 1975: 215.

11 Prt. 316a3–b2; my emphasis. On frame-breaking that the passage involves see the next
section; on Critias see above, n. 13 to Chapter 2.
12 Cf. Bowery 2007: 90.
The Explicit and the Implicit Narrator Combined 83

when I heard his words and the sound of the cheering. To tell you the
truth, I wanted to get time to think what the meaning of the poet really is.
So I turned to Prodicus …13

The third case is probably the subtlest of the three. It occurs, again, at one of
the structural breaks:

Thus I spoke and was rising from my seat, when Callias seized me by the
right hand, and in his left hand caught hold of this old cloak of mine. He
said: “We cannot let you go, Socrates, for if you depart our discussion (hoi
dialogoi) will not be the same.”14

Socrates’ emphatic pointing at his cloak not only brings us back to the dra-
matic frame, that is, to the narrative level on which the fictional act of narra-
tion takes place, but also vividly materializes the interpenetration of narrative
levels produced thereby. For a moment, the frame and the framed interlock,
but narrative flow resumes almost immediately. As we have seen, such quick
re-establishment of the boundaries between the narrative levels is a distinctive
characteristic of rhetorical metalepsis.15 Only in the Euthydemus and the Pha-
edo do the dramatic frame and the framed narrative interpenetrate so deeply
that the plot level as a whole becomes affected (below).

2.4 Juxtaposition of the Frame and the Narrative

A considerable gap separates the messages delivered in the dramatic frame
from those in the body of the Protagoras. In the latter, the evaluation of Pro-
tagoras is mixed. On the one hand the dialogue gives notable credit to some of
Protagoras’ arguments and the man himself.16 This especially concerns the ar-
gument of the teachability of virtue – so much so that the dialogue concludes
with Socrates teasingly admitting that the discussion has made him change
his initial position on the issue. Furthermore, he also admits that Protagoras’
myth of Prometheus, introduced to support this argument, has been of great
value to him:

13 Prt. 339d10–e5; my emphasis.

14 Prt. 335c7–d3; my emphasis.
15 Above, p. 23. On the distinction between rhetorical and ontological metalepsis see also
below, p. 86.
16 See, e.g., Gagarin 1969: 163: “we must conclude, it seems to me, that Plato is trying to pres-
ent a sympathetic portrait of Protagoras in this dialogue, and that he shows great respect,
not only for the man, but also for his thought.” Cf. Guthrie 1975: 233: “Protagoras was an
opponent, but a worthy one, and to devote a dialogue to showing the strength rather than
the weakness of the adversary was not wasted labour.”
84 Chapter 4

I liked Prometheus in the myth better than Epimetheus: so I make use of

him and spend my time on all these matters as a means for taking fore-
thought (promēthoumenos) for my whole life. (361d2–5)17

Significantly, throughout the dialogue Plato dodges such aspects of Protagoras’

doctrine that Socrates would find wholly unacceptable – first and foremost,
the “man-as-measure” proposition. Protagoras’ relativism will be given its due
in the Theaetetus, not in the dialogue that bears the great sophist’s name.
On the other hand, Socrates the character leaves no room for doubt as to his
unambiguously negative attitude to the kind of education that the sophists,
Protagoras included, claim to provide. In the opening of the framed part of the
dialogue, just before taking the young Hippocrates to the meeting with Pro-
tagoras, whose student the youth aspires to become, Socrates converses with
him at length about the dangers of sophistic education (311a-314c). No less sig-
nificant is that, as Michael Erler pointed out recently,18 at the end of the dia-
logue Socrates and Hippocrates depart together (362a4 “Having said and heard
that much, we went our way”), leaving no room for doubt as to the choice that
Socrates’ young friend eventually made.
As distinct from this, the perspective adopted in the dramatic frame is un-
ambiguously positive. Socrates’ praise of Protagoras’ wisdom, expressed in
terms of the Symposium scale of beauty,19 strongly suggests that, whatever hap-
pens in the body of the dialogue, Plato meant to focus the readers’ attention
on the elements in Protagoras’ doctrine (presumably, the Prometheus myth)
that he saw as worthy of appreciation. In other words, the perspective on Pro-
tagoras offered in the dramatic frame does not fully overlap the one given in
the narrated part of the dialogue; the reader is invited to form his or her own
opinion of the famous sophist. As we shall see immediately, a similar gap be-
tween the perspectives adopted in the dramatic frame and in the body of the
dialogue is also characteristic of the Euthydemus.

3 The Euthydemus

Crito questions Socrates as to the identity of the stranger with whom he saw
him converse at the Lyceum the day before, and Socrates tells him of his

17 Tr. W.K.C. Guthrie, slightly modified. On Plato’s treatment of the Prometheus myth in the
Protagoras see Morgan 2000: 134–54.
18 Erler, forthcoming.
19 Quoted above, p. 81. On the role of the theme of love in the dialogue’s dramatic frame see
Friedländer 1964: 5–6.
The Explicit and the Implicit Narrator Combined 85

­meeting with the brothers Euthydemus and Dionysodorus, two famous practi-
tioners of the art of eristics.

3.1 Frame Narrator (Implicit): Crito

Let us look closer at the way the dialogue begins.

“Who was the person, Socrates, with whom you were speaking yesterday
at the Lycaeum? There was such a crowd around you that I could not get
within hearing, but I caught a sight of him over their heads, and I made
out, as I thought, that he was a stranger with whom you were talking:
who was he?” “There were two, Crito, which of them do you mean?” “The
one whom I mean was seated second from you on the right-hand side. In
the middle was Cleinias the young son of Axiochus, who has wonderfully
grown … the other is thin and looks younger than he is.” “He whom you
mean, Crito, is Euthydemus; and on my left hand there was his brother
Dionysodorus, who also took part in the conversation.” (271a1–b8)

We perceive the setting of the dialogue, the participants’ physical appearance

and their respective positions through Crito’s eyes. We also experience the
gradual unveiling of the participants’ identities through Crito’s mediation –
first by means of an incomplete picture of the yesterday’s gathering he has
formed in his mind, and then by his completing it by eliciting additional infor-
mation from Socrates.20 As we have seen, in both the narrated and the dramat-
ic dialogues Plato repeatedly employs such markers of focalization to signal
the source of narratorial authority. This fully agrees with our preliminary as-
sumption that the narratees of the mixed dialogues should be regarded as im-
plicit narrators of the dialogues’ dramatic frames (the first intradiegetic level).

3.2 Narrator-Hero (Explicit): Socrates

Beyond the dramatic frame, however, Socrates’ narratorial authority is absolute
(the second intradiegetic level). First he describes himself sitting alone in the
Lyceum dressing room; this is followed by a description of Euthydemus, Diony-
sodorus and their followers coming in and starting to walk about in the covered
court; and finally, he registers the arrival of the beautiful Cleinias ­accompanied

20 See also Euthyd. 271c2–4, and above, p. 21. Cf. Sermamoglou-Soulmaidi 2014: 126: “So the
reader, with him [Crito], first visualizes the scene as an image, before the actual discus-
sion (i.e. what we would hear) is reported. The reference to the crowd surrounding the
interlocutors adds the chorus to the dramatis personae, and the stage is set” (original
emphasis). Cf. also Collins 2015: 72–74.
86 Chapter 4

by Ctesippus and other admirers (272e1–273b1).21 Everything happening in the

story – Euthydemus’ and Dionysodorus’ laughter (273d); Cleinias’ blushing
(275d); the cheers and roaring laughter of Euthydemus’ followers (276b, 276d,
303b); the audience’s silence and amazement (276d); Ctesippus’ anger, lack of
restraint, and silence (283e, 285a, 288b; 286b, 294d); Euthydemus’ anger and
Dionysodorus’ blushing (295d, 297a); Ctesippus’ and Cleinias’ laughter (298e,
300d) – is communicated to us through Socrates’ perception.

3.3 Metalepsis
The action of the Euthydemus proceeds simultaneously on two narrative lev-
els. Socrates the narrator never loses sight of his narratee, repeatedly address-
ing Crito throughout the dialogue.22 The result is a unique mixture of dramatic
and narrated form,23 which goes far beyond an occasional apostrophe we en-
countered in the Charmides and the Protagoras. Insofar indeed as throughout
the Euthydemus metalepsis is operated by means of the second-person pro-
noun, the dialogue comes close to what contemporary literary theory identi-
fies as the second-person narrative. The frame and the framed intermingle,
exhibiting what Marie-Laure Ryan calls “metaleptic contamination,” a distinc-
tive feature of ontological metalepsis.24
A more narrowly focused case of frame-breaking occurs at Euthydemus
290e. Socrates’ young interlocutor Cleinias arrives at the conclusion that just
as geometricians, astronomers and mathematicians, having made their discov-
eries, hand them over to the dialecticians, so too generalship cannot be consid-
ered the supreme art because generals hand over the fruits of their victories to

21 Cleinias was a relative of Alcibiades; he also appears in the pseudo-Platonic Axiochus and
in the Symposium of Xenophon. See further Nails 2002: 100–101. On Ctesippus see n. 12 to
Chapter 2.
22 Euthyd. 275c5–6 “What followed, Crito, how can I rightly narrate?”; 283a1–2 “Thus I spoke,
Crito, and was all attention to what was coming”; 291d5–6 “You shall judge, Crito, if you are
willing to hear what followed”; 291e2 “And would not you, Crito, say the same?”; 292d5–6
“Shall we say, Crito, that that is the knowledge by which we are to make other men good?”;
292e8–293a1 “Thereupon, Crito, seeing that I found myself in such perplexity, I lifted up
my voice”; 294d7–8 “At last, Crito, I too was carried away by my incredulity”; 303a4–5 “At
this I was quite struck dumb, Crito, and lay prostrate”; 303b1–2 “Then, my dear Crito, there
was universal applause of the speakers and their words”; 304b6–7 “Such was the discus-
sion, Crito, and after a few more words had passed between us we went away.”
23 Cf. Guthrie 1975: 267: “a skilful blending of direct and reported forms.” See also Morgan
2004: 368; Collins 2015: 58.
24 Ryan 2006: 207: “Whereas rhetorical metalepsis maintains the levels of the stack distinct
from each other, ontological metalepsis opens a passage between levels that results in
their interpenetration, or mutual contamination.” On the metaleptic character of the
second-person narrative see McHale 1987: 222–27; cf. Fludernik 2003: 385–86, 393–94.
The Explicit and the Implicit Narrator Combined 87

the statesmen. The idea that mathematics (arithmetic, geometry, astronomy,

harmonics) is subordinated to dialectic is central to the program of higher edu-
cation of the Guardians as outlined in Book 7 of the Republic.25 So it is hardly
mere chance that at this very moment Crito the narratee interferes by asking:
“And do you mean, Socrates, that the youngster said all this?” (290e1–2; my em-
phasis). A brief exchange follows, in which Socrates ironically suggests that
it was probably “some superior entity (tis tōn kreittonōn)” who uttered these
words (291a4). The startling reference to the superiority of dialectic over math-
ematics has been effectively highlighted. As we shall see immediately, a similar
strategy of foregrounding by means of metalepsis is also characteristic of the
Phaedo.

3.4 Juxtaposition of the Frame and the Narrative

Everything indicates that the Euthydemus constant switching back and forth
between the story and the dramatic frame was meant to create a distance be-
tween what Socrates asserts in this dialogue and what Plato’s actual position is.26
The Socrates of the Euthydemus frequently expresses unalloyed admiration
for the virtuoso performance of the champions of eristics Euthydemus and
Dionysodorus. To alert the reader that this admiration should not be taken
at face value, Plato takes a series of steps. First, he introduces an additional
character, Ctesippus the admirer of Cleinias, who supplies common-sense re-
sponses, including laughter, to the brothers’ hair-splitting arguments. Secondly
he makes Crito echo these responses in the dramatic frame,27 primarily in his
favorable report of the opinion of an unnamed critic of eristics (most probably
Isocrates28) adduced at the end of the dialogue (304d-305a). Finally and most
importantly Plato lets Socrates express a nuanced opinion in the words that
supply the dialogue’s closure and are clearly addressed not to Crito alone:

25 Resp. 521d-541b. Cf. Friedländer 1964: 190, on the Euthydemus passage: “We sight the sys-
tem of sciences developed in the Republic.” Similarly Kahn 1996: 321: “The most advanced
reference to dialectic in Group i [sc. the dialogues preceding the Republic].” On the cen-
tral position of this section within the Euthydemus see Thesleff 1967: 40, 111; cf. Chance
1992: 126.
26 Cf. Kahn 1996: 325: “Plato makes use of the frame dialogue between Socrates and Crito,
both in the middle and at the end of the narrated dialogue, like the parabasis in an Aris-
tophanic comedy, to speak as it were directly to the readers.” Cf. also Collins 2015: 67: “the
running commentary which the framing encounter provides.” On the difference between
the parabasis and the form of metalepsis used by Plato see below, Chapter 5.
27 Cf. Ferrari 2010: 18: “By fully characterizing the narratee, Plato makes the situation clear”;
on Crito’s attitude see also Collins 2015: 62.
28 See the discussion in Sermamoglou-Soulmaidi 2014: 143–51.
88 Chapter 4

Do you then be reasonable, Crito, and do not mind whether the teachers
of philosophy are good or bad, but examine philosophy itself well and
carefully and, if it appears to you as unworthy, seek to turn away every-
one from it, and not your sons only; but if it appears as what I believe it
is, then cheer up and follow and serve it, you and yours, as the saying is.
(307b6–c4)

Despite Socrates’ statements in the body of the dialogue, the constant inter-
play between the frame and the framed results in that the perspective adopted
in the Euthydemus is unmistakably multifocal: not only Crito but also the read-
ers are invited to form an independent opinion by choosing between the op-
tions the dialogue displays.

4 The Phaedo

The dramatic frame29 of the dialogue takes us to Phlius in the northwestern Pelo-
ponnese. Socrates has been dead for some time now, but before Phaedo’s arrival30
there has been very little traffic between Athens and Phlius. Echecrates31 and
other Pythagoreans of Phlius have already been informed of the particulars of
Socrates’ trial (58a1–3); however, besides the bare fact that Socrates took poi-
son in prison, they are ignorant of the circumstances of his death.

4.1 Frame Narrator (Implicit): Echecrates

Like the other mixed dialogues, the Phaedo has two narrators, one implicit in
the dramatic frame, the other explicit in the body of the dialogue. By definition,
Phaedo’s narratee Echecrates should occupy the position of the implicit nar-
rator of the dramatic frame (the first intradiegetic level, see above, pp. 77–78).
And indeed, just like the frame narrators of the Protagoras and the Euthyde-
mus, Echecrates is the one who focalizes the information supplied by Phaedo
by becoming gradually exposed to it. Again, as in the two other dialogues, this
gradual focalization takes the form of a series of questions.
The dialogue opens with a question by Echecrates: “Where you yourself,
Phaedo, in the prison with Socrates on the day when he drank the poison?”
(57a1–2). Echecrates is eager to learn what Socrates said before his death,

29 The plot structure of the dialogue is discussed in detail in Chapter 6.

30 Phaedo of Elis, author of Socratic dialogues, was a student of Socrates; he founded a
school of philosophy in his native Elis. See further Sedley 1995: 8–9; Kahn 1996: 9–12; Nails
2002: 231.
31 Echecrates of Phlius was a prominent Pythagorean. See further Sedley 1995: 10–11; Nails
2002: 138.
The Explicit and the Implicit Narrator Combined 89

and how he died ((57a5–6). He also wonders why the execution was delayed
(58a3–5), and Phaedo provides him and the readers with a detailed account
of the Athenian custom of sending a sacred embassy to Delos, and how it
held up the execution (58a6–c5). Echecrates again asks about the manner
of Socrates’ death, about what was said and done on the occasion and who
among Socrates’ followers was present (58c6–9). Phaedo is delighted to supply
the required information. “You will have listeners who are of the same mind
with you,” Echecrates says before Phaedo starts his narration, “and I hope that
you will be as exact as you can” (58d7–9). The frame story is over, and the main
narrative begins.

4.2 Narrator-Observer (Explicit): Phaedo

The main narrator of the Phaedo (the second intradiegetic level) is positioned
differently from the narrators of the Protagoras and the Euthydemus. Phaedo
is not the dialogue’s leading character nor is he one of the main interlocutors.
Even though on one memorable occasion he is pushed to the foreground (more
below), Phaedo’s principal position is that of observer and witness.
Not that Phaedo’s role of narrator-observer affects his narratorial author-
ity: throughout the dialogue, we are exposed only to what Phaedo perceives,
remembers, or chooses to tell us.32 This concerns the dialogue’s setting,33 the
respective positions of those present,34 Socrates’ body language,35 as well as
the near clinical description of Socrates’ agony and death.36 When the audi-
ence’s emotional state is beyond the limits of the narrator’s perception, this
is acknowledged and accounted for by means of metalepsis: “when we had
heard them [Simmias and Cebes] state their objections, we all felt very much

32 Cf. Phd. 102a10–b3 “As far as I can see, after all this had been accepted … he said …” or
103a4–5 “Hereupon one of the company, though I do not exactly remember which of
them, said …”
33 Phd. 59e8–60a2 “On entering we found Socrates just released from chains, and Xantippe,
whom you know, sitting by him, and holding a child in her arms” (my emphasis; see further
below, on metalepsis in the Phaedo).
34 Phd. 89a9–b2 “I was close to him [Socrates] on his right hand, seated on a sort of stool,
and he on a couch which was a good deal higher.”
35 Phd. 60b1–3 “Socrates, sitting up on the couch, bent and rubbed his leg saying …”; 86d 5
“Socrates looked fixedly (διαβλέψας) at us as his manner was”; 103a11 “Socrates inclined
(παραβαλών) his head to the speaker and listened”; 117b5–6 “looking at the man from
under the brows (ὑποβλέψας) like a bull, as his manner was.”
36 Phd. 117e4–118a14. Cf. Murley 1955: 286: “The familiar details of the closing scene of the
Phaedo (117E) become impressive from their stark simplicity fraught with tragic meaning,
and would seem to have determined the treatment of Fantine’s death in Les Miserables as
Jean Valjean follows its progress.”
90 Chapter 4

depressed, as we told one another later.” Plato’s narrator is never configured as

omniscient.37
The atmosphere of the dialogue – sombre and relaxed at one and the same
time, wholly depends on Phaedo’s careful selection of details. No cheers and
roaring laughter here, only gentle laughing and smiling – by Socrates (84d8,
86d6, 115c5), by Cebes (62a8, 77e3, 101b3), by Simmias, who also comments:
“Though I am not in a laughing mood today, Socrates, you have made me
laugh” (64a10–b1). Understandably enough, weeping and lament loom large –
by Xantippe (60a1–b1), by the jailor (116d2), by Crito, by Phaedo himself, by
Apollodorus (the frame narrator of the Symposium), and by everyone present
except Socrates himself (cf. 59a8–9 “we were laughing and weeping by turns”
and 117c5–d6). On both the artistic and personal level (see especially Socrates’
playing with Phaedo’s hair at 89ab), the dialogue is essentially Phaedo’s story,
and one may wonder whether Plato’s account of Socrates’ last day would have
looked different had he chosen a different narrator – for example, Xenophon’s
source Hermogenes.38

4.3 Metalepsis
Like the Euthydemus, the Phaedo is a metaleptic dialogue whose action pro-
ceeds simultaneously on two narrative levels. In the framed narrative delivered
by Phaedo, Socrates’ opening arguments for the immortality of the soul – the
argument from alternation, the argument from recollection, the argument
from affinity – are followed by a structural break, which introduces the objec-
tions raised by Simmias and Cebes39 and serves as a transition to the culminat-
ing argument (below, pp. 144–45). The audience’s reaction to the objections is
worth being quoted in full:

Well, when we had heard them state their objections, we all felt very
much depressed, as we told one another later. We had been quite con-
vinced by the earlier part of the discussion, and now we felt that they had
upset our convictions and destroyed our confidence not only in what had

37 Phd. 88c1–2 πάντες οὖν ἀκούσαντες εἰπόντων αὐτῶν ἀηδῶς διετέθημεν, ὡς ὕστερον ἐλέγομεν
ἀλλήλους; tr. H. Tredennick (quoted in full below). On the lack of omniscience in Plato’s
narrators see also above, n. 16 to Chapter 2; on metalepsis in the Phaedo see the next
section.
38 Plato was not among those present at Socrates’ death (Phd. 59b10), and it is not out of
the question that it was indeed Phaedo who was his source. On Hermogenes (one of the
interlocutors in the Cratylus) see above, n. 26 to Chapter 3.
39 Pythagoreans from Thebes, friends of Socrates; also mentioned in the Crito (45b). See
further Sedley 1995; Nails 2002: 82–83, 260–61.
The Explicit and the Implicit Narrator Combined 91

been said already, but also in anything that was to follow later. (88c1–7;
my emphasis)

Suddenly, this reaction is echoed on the upper level by Phaedo’s narratee.

Echecrates interrupts Phaedo’s narrative, expressing the same sentiments as
he audience in the narrated story:

ECHECRATES: There I feel with you and the others, by heaven I do, Pha-
edo, and when I was listening to you I was asking myself the same question:
What argument can I ever trust again? For what could be more convincing
than the argument of Socrates, which has now fallen into discredit?40

This is followed by the memorable episode of Socrates’ playing with Phaedo’s

hair (89bc). For a while, Phaedo replaces Simmias and Cebes as Socrates’ in-
terlocutor, and Socrates’ diatribe against misologists, “haters of logoi,” is ad-
dressed to him (89d-91c; see also below, p. 145). Socrates’ response to the
objections comes next; it prepares the ground for the final argument for
the immortality of the soul and introduces the Theory of Forms (91c-101e). Now
Simmias and Cebes are fully satisfied with Socrates’ argument: “‘What you say
is most true,’ said Simmias and Cebes, both speaking at once” (102a2). In an-
other metaleptic leap, Echecrates echoes them again from the upper narrative
level:

ECHECRATES: Yes, Phaedo; and I do not wonder at their assenting. Any-

one who has the least sense will acknowledge the wonderful clearness of
Socrates’ reasoning. PHAEDO: Certainly, Echecrates; and such was the
feeling of the whole company at the time. ECHECRATES: Yes, and equally
of ourselves, who were absent, and are now listening. But what followed?
(102a3–9).

The double frame-breaking that encloses the argument of the Theory of

Forms not only highlights its central position in the Phaedo41 but also trans-
fers the dialogue into the realm of actuality, providing it with a broader tem-
poral perspective than that supplied by its immediate context. This is neatly

40 Phd. 88c8–d3. Cf. Taylor 1960: 194: “The purpose of all this by-play is to call attention to the
critical importance of the two problems which have just been raised. We are, in fact, at the
turning-point of the discussion.” Cf. also Thesleff 1967: 40–41.
41 On a similar function of metalepsis in the Euthydemus see above, pp. 86–87.
92 Chapter 4

conveyed by Echecrates’ words “and equally of ourselves, who were absent, and
are now listening,” encompassing all readers and listeners of the Phaedo.42
Echecrates’ interventions thus affect the entire dialogue, functioning as
an integral element of its plot (see above, p. 83, on ontological metalepsis).
It seems significant in this connection that the Euthydemus and the Phaedo
are the only dialogues in which metaleptic transgression is initiated not by
the narrator, as in the other dialogues and elsewhere in the Phaedo (see 60a2,
88c2, 117b3), but by the narratee. This results in the two dialogues functioning
not only as stories delivered by Socrates and Phaedo respectively, but also as
dramas of the reception of these stories by the audience.43

4.4 Juxtaposition of the Frame and the Narrative

While in the Protagoras and the Euthydemus the juxtaposition of the frame
and the framed was used to introduce several contrasting perspectives, in the
Phaedo it serves to multiply and enhance the perspective adopted in the main
story by deepening its temporal dimension. It is hardly by mere chance that
the chronological distance between Socrates’ death and Phaedo’s delivery of
his story is not specified. Admittedly, the time that has passed could not have
been very long, but Plato’s vagueness creates the impression of a long interval
suitable for a great historical event worthy of perpetual commemoration. This
is made especially clear from Phaedo’s anachronistic use of the expression “the
men of those days” in the dialogue’s closure:

Such was the end, Echecrates, of our friend; concerning whom I may truly
say, that of all the men of those days (τῶν τότε) whom I have known, he
was the wisest and justest and best (118a15–17).44

Such a long-term perspective rising from juxtaposition of the dramatic frame

and the main narrative also characterizes the Symposium and the Theaetetus.

42 Phd. 102a8 καὶ γὰρ ἡμῖν τοῖς ἀποῦσι, νῦν δὲ ἀκούουσιν. Cf. Morgan 2004: 367: “It is hard not
to think that his [Echecrates’] reactions are a guide to reader reception and response.” Cf.
also Rowe 1993: 211; Hösle 2012: 174–75; Collins 2012: 165–67.
43 Cf. Brooks 1992: 236: “Framed narratives and those, such as [Albert Camus’] La Chute,
that incorporate the listener in the discourse of the speaker incorporate most explicitly
a condition of all narrative: shape and meaning are the product of the listening as of the
telling.”
44 The expression is also used, with a greater justification, in the dramatic frame of the Sym-
posium, see Symp. 172b1, 173b4, and below in this chapter.
The Explicit and the Implicit Narrator Combined 93

5 The Symposium

Socrates’ follower Apollodorus fulfils the request of an unnamed friend to tell

him and his companions, all wealthy tradesmen, what he has heard from Aris-
todemus, another associate of Socrates, about the speeches in praise of love
delivered years before at a dinner party in the house of the poet Agathon.

5.1 Frame Narrator i (Implicit): Multiple Narrators

Rather than with a dramatic exchange characteristic of the other mixed dia-
logues, the Symposium starts with a monologue by the second-level narrator,
Apollodorus, addressed to his unindentified fictional narratee:

Concerning the things about which you ask to be informed I believe

that I am not ill-prepared with an answer. For the day before yesterday I
was coming from my own home at Phalerum to the city, and one of my
acquaintance, who had caught a sight of me from behind, calling out
playfully in the distance, said: “Apollodorus, O you Phalerian man, halt!”
(172a1–5)

It is not for the first time that Apollodorus tells this story.45 Just two days earlier
he had told it to Glaucon46 when the two were on their way from Phaleron to
Athens. In fact, we learn everything about the date and circumstances of Ag-
athon’s party and the man who was the source of the story from Apollodorus’
retelling of his earlier conversation with Glaucon. It is Glaucon as presented in
Apollodorus’ report who actually focalizes the background information that
was missing at the beginning of the dialogue.
Glaucon has already heard the story from another person, whose source was
Phoenix son of Philip; however he did not find Phoenix’s version satisfactory
(172b). Together with Glaucon, we learn about when and on what occasion Ag-
athon’s party took place: not recently, as Glaucon supposed, but when both he
and Apollodorus were boys, the day after Agathon won the prize with his first
tragedy (173a). Apollodorus also describes the man who told him the story:

45 Critias’ story of Atlantis as delivered in the Timaeus is presented in a similar way, see Ti.
20c6–d6, 26a7–b2.
46 Symp. 172c2. Cf. Nails 2002: 315: “The lack of any further specification of Glaucon by de-
motic or patronymic makes it almost certain that the reference is to Plato’s brother” – that
is, the same Glaucon who appears in the frame story of the Parmenides and the opening
of Republic 1, and who is one of the two main interlocutors of Socrates in Republic 2–10; cf.
n. 21 to Chapter 2.
94 Chapter 4

The same one who told Phoenix: Aristodemus, of the deme of Cydathen­
eum. He was a little fellow, who never wore any shoes. He had been at
Agathon’s feast, and I think that among men of those days (tōn tote) there
was no one who was a more devoted admirer of Socrates. (173b1–4)

Finally, Apollodorus informs Glaucon that he consulted Socrates about Aris-

todemus’ account, and Socrates confirmed its reliability (173b4–6). The frame
is restored to dramatic form by the interference of the man to whom Apol-
lodorus actually tells the story, and by Apollodorus’ admonition to his listen-
ers (only now do we learn that they are wealthy tradesmen) to follow his own
example by choosing the philosophical way of life (173d4–e6).
So in fact Glaucon, whose part is narrated by Apollodorus, assumes all prin-
cipal functions of the dialogue’s narratee, thus becoming one of the implicit
narrators in the dialogue’s frame (the first intradiegetic level). This unique
arrangement seems to have been aimed at presenting the story of Agathon’s
party as reverberating in multiple versions.47

5.2 Frame Narrator ii (Explicit): Apollodorus

As with Antiphon, the intermediary narrator of the Parmenides (Chapter 2),
Apollodorus’ part of the frame story (the second intradiegetic level) is mini-
malized: from him we learn only that he heard about Agathon’s party from
Aristodemus and that Socrates confirmed the latter’s account. Nevertheless,
Apollodorus’ presence is clearly discernible in the story he delivers. To begin
with, the account of Agathon’s party that we receive is based not only on what
Aristodemus the eyewitness remembered or chose to communicate but also
on what his narratee Apollodorus chose to retain and transmit further on:

Aristodemus did not recollect all that was said, nor do I recollect all that
he related to me; but I will tell you what I thought most worthy of remem-
brance and by the speakers worthy of remembrance as well. (178a1–5)

This is probably the reason why, as distinct from the Protagoras, the Euthyde-
mus, and the Phaedo, the full list of those present is not supplied in the dia-
logue’s opening. The ensuing story is thus presented as resulting from editing
and rearranging by both the third-level and the second-level narrator.48

47 On the Symposium as reflecting “a fundamentally oral culture of intellectual heroic tales”

see Tarrant 1996: 138. Cf. also the charts of transmission adduced in Halperin 1992: 98, and
Hunter 2004: 23.
48 Cf. above, pp. 37–38, Socrates the narrator of Republic I admitting his editing the original
conversation with Thrasymachus to render it smoother.
The Explicit and the Implicit Narrator Combined 95

On one occasion, Apollodorus explicitly takes responsibility for the form in

which the story is presented:

Pausanias came to a pause (Pausaniou de pausamenou) – this is the bal-

anced way (isa legein) in which I have been taught by the wise to speak; and
Aristodemus said that the turn of Arisrophanes was next … (185c4–6; my
emphasis)

Apollodorus’ remark comes very close to Socrates’ remark in similar circum-

stances in Republic i (p. 37); also as in Republic i, it produces metalepsis (more
below).

5.3 Narrator-Observer (Explicit): Aristodemus

Nevertheless, as regards the main story (the third intradiegetic level), Aristode-
mus’ narratorial authority is unquestioned. This transpires, for example, from
the following remark: “And some other speeches followed which Aristodemus
did not remember; the next which he repeated was that of Pausanias” (180c1–3).
Whatever selection of speeches Apollodorus might have configured, he had to
choose from the materials Aristodemus supplied.
No less importantly, the dialogue’s concluding part clearly indicates that the
narrative of the Symposium is envisaged as strictly circumscribed by what Aris-
todemus has witnessed himself:

He [Aristodemus] was awakened towards daybreak by a crowing of

cocks, and when he awoke, the others were either asleep, or had gone
away; there remained only Socrates, Aristophanes, and Agathon, who
were drinking out of a large goblet which they passed round, and
Socrates was discoursing to them. Aristodemus was only half awake,
and he did not hear the beginning of the discourse; the chief thing,
however, was … (223c2–d2)

As with Pythodorus’ ignorance in the Parmenides and Socrates’ ignorance

in the Cratylus, the Laches, and the Sophist (pp. 39, 65, 72–73), Aristode-
mus’ ignorance as to what happened while he was asleep foregrounds his
lack of omniscience and, consequently, our limited knowledge of the event
described.
As in the other dialogues, the characters’ appearance, movements, emo-
tional responses, and the like are presented to us in the way the narrator per-
ceives them. For example, this is how Aristodemus describes the audience’s
impression as to the nature of Alcibiades’ feelings towards Socrates: “He
[­Aristodemus] said that, when Alcibiades had finished, there was a laugh at
96 Chapter 4

his outspokenness; for he seemed to be still in love with Socrates” (222c1–3).49

Socrates’ unusually clean and tidy appearance (174a), Aristophanes’ hiccups
(185ce), Alcibiades’ intrusion and drunkedness (212c-214a), the laughter and
cheer of those present (189b, 198a) – all these are communicated to the reader
through the mediation of Aristodemus.

5.4 Metalepsis
The recurrent use of a two-tier system of reported speech, which is the signal
characteristic of the Symposium, constantly reintroduces the frame narrator
Apollodorus, thereby not allowing the reader fully to immerse into the world
of the narrated story. As David Halperin put it,

Plato constantly reminds the reader of the many narrators that intervene
between the reader and the transmitted story – he emphasizes the re-
ported character of the account by sprinkling throughout Apollodorus’
narrative such phrases as “he said that he said” (ephē phanai).50

Aristodemus’ story is thus thoroughly intermingled in Apollodorus’ presenta-

tion of it. As mentioned, this kind of “metaleptic contamination,” also charac-
teristic of the narrated part of the Parmenides, is unique to Plato (see above,
Chapter 1, p. 23).

5.5 Juxtaposition of the Frame and the Narrative

It is not hard to see that the main narrative of the Symposium is focalized on
Socrates. It is not just that the concluding episode of the dialogue is staged
as Alcibiades’ unadulterated praise of Socrates. The embedded stories as de-
livered by Socrates and Alcibiades highlight such pivotal events in Socrates’
life as his choice of the philosophical way of life, presented in the dialogue as
triggered by his encounter with Diotima, and his participation in the military
campaigns at Potidaea and Delium. As a result, like what we observed in the
Laches, Socrates is presented to the reader as an accomplished philosopher
and a model citizen-soldier.51 In the Symposium, however, Socrates is also seen

49 Cf. Prm. 130a3–7, Pythodorus’ assessment of the impression made by the youthful Socrates
on Parmenides and Zeno (above, p. 42).
50 Halperin 1992: 97 (original emphasis). Cf. also Hunter 2004: 23: “this is a mode of presenta-
tion which never lets us forget that we are not being offered unmediated access to a ‘true
account’ of ‘what happened’ in Agathon’s house.”
51 Characteristically, both dialogues place special emphasis on Socrates’ exemplary perfor-
mance during the retreat from Delium (424 bce; also mentioned at Ap. 28e3); see above,
p. 65, and Symp. 220d-221b.
The Explicit and the Implicit Narrator Combined 97

from the vantage point of the frame story, whose dramatic date is set shortly
before his death.52 By offering such a comprehensive retrospective of Socrates’
life, the Symposium produces an in-depth portrait of its hero, unattainable
in the single-level Laches. As we shall see immediately, a similar biography-
oriented approach is also adopted in the Theaetetus.

6 The Theaetetus

Terpsion asks Euclides, who has just seen the dying Theaetetus being carried
off home to Athens by way of their native Megara, to tell him about Socrates’
encounter with Theaetetus as a youth, which Euclides heard of once from
Socrates himself. Euclides agrees, and his slave reads aloud his master’s written
account of how Theaetetus’ tutor Theodorus introduced him to Socrates and
of the conversation that followed.53
As in the Parmenides and the Symposium, there are three narrators here –
Socrates, Socrates’ narratee Euclides, and Euclides’ narratee Terpsion. Also
as in the Parmenides and the Symposium, the two upper-level narrators are
absent from the body of the dialogue: they are homodiegetic as regards the
frame narrative they deliver but heterodiegetic as regards the main story. But
the Theaetetus differs from the other multi-level dialogues in that all its nar-
rators are implicit. The first-level narrator, Terpsion, is implicit because, as in
the other mixed dialogues, the Theaetetus’ frame story is cast in dramatic form.
The third-level narrator, Socrates, is implicit because, as has been emphasized
more than once, his narrative act has been deliberately suppressed. Finally, by
converting Socrates’ narrative into a dramatic dialogue, thereby suppressing
its narrator, Socrates’ narratee Euclides has transformed not only Socrates but
also himself into an implicit narrator. The following comparison of the The-
aetetus with the two other multi-level dialogues illustrates this (the implicit
narrators are bracketed):

52 The dramatic frame of the dialogue is placed shortly before Agathon’s death in 401 bce
(see Symp. 172c), that is to say, approximately two years before the death of Socrates. On
“biographical criticism” in the Symposium see Halperin 1992: 100; see also above, pp. 43–
44, on the Parmenides.
53 Euclides was the founder of the Megarian School of philosophy and author of Socratic
dialogues; after Socrates’ execution, Plato and other Athenian disciples of Socrates stayed
for a while in Megara with him. Both Euclides and Terpsion were present at Socrates’
death; see Phd. 59c2. See further Kahn 1996: 12–15; Nails 2002: 144–45, 274. On Theaetetus
and Theodorus see nn. 11 and 12 to Chapter 3.
98 Chapter 4

the Parmenides the Symposium the Theaetetus

1st level Cephalus [Apollodorus’ [Terpsion]

friend]
2nd level Antiphon Apollodorus [Euclides]
3rd level Pythodorus Aristodemus [Socrates]

Actually, the only thing preventing the Theaetetus from being treated as a fully
dramatic dialogue is Euclides’ account of how he suppressed the dialogue’s
narrator, referred to several times throughout this book (above, Chapter 1, and
passim).

6.1 Frame Narrator i (Implicit): Terpsion

Euclides arrives at his house in Megara where he finds Terpsion, who has been
looking for him. “But I was not in the city,” Euclides explains. “Where then?”
“As I was going down to the harbor, I met Theaetetus – he was being carried up
to Athens from the army at Corinth.” “Was he alive or dead?” Euclides explains
that Theaetetus has been badly wounded and, worse still, is suffering from a
disease that has broken out in the army. “The dysentery, you mean?” Euclides
answers in the affirmative. The interlocutors agree that Theaetetus’ death will
be a tremendous loss. Euclides calls the dying man “kalos kagathos,” adding
that he has just heard Theaetetus being highly praised for his performance on
the battlefield. “But why did he go on, instead of stopping in Megara?” “He
insisted on getting home,” Euclides explains. “I entreated and advised him to
remain, but he would not listen to me” (142a4–c3).
The opening exchange between Euclides and Terpsion is a standard ex-
ample of the zooming-in procedure routinely performed by Plato’s narrator
(above, pp. 20–21): by asking questions, thereby gradually absorbing the in-
formation which Euclides supplies, Terpsion transmits this information to the
reader (the first intradiegetic level).

6.2 Frame Narrator ii (Implicit): Euclides

Going back home from his meeting with Theaetetus, Euclides recalls that
shortly before his death Socrates told him about a conversation he had recent-
ly had with Theaetetus. Socrates was full of admiration for the gifted youth
and expressed the opinion that Theaetetus would most certainly become il-
lustrious if he reached manhood (142c7–d3). “The prophecy has certainly been
The Explicit and the Implicit Narrator Combined 99

fulfilled,” Terpsion says, “but what was the conversation? Can you tell me?”
(142d4–5). This is where Euclides’ narration begins in earnest.
Euclides took notes (hupomnēmata) of Socrates’ account as soon as he got
home.

And later I filled them up from memory, writing them out at leisure; and
whenever I went to Athens, I asked Socrates about any point which I had
forgotten, and on my return I made corrections; thus I have nearly the
whole conversation (logos) written down. (143a2–5)

As with Apollodorus in the Symposium, the consultation with Socrates guaran-

tees the reliability of Euclides’ account. Like Apollodorus again, Euclides also
edited the original account as he saw fit.54
Here however the similarity with the Symposium ends. Euclides’ account
is unique in that it is presented as a written composition. Does this mean
that Euclides’ narratorial authority is to be negotiated differently from that of
Plato’s other narrators? For example, should he be considered the author of
everything found beyond the dialogue’s dramatic frame?55 Or is he just a pas-
sive transcriber of what Socrates communicated to him?56 A comparison with
Plato’s other narrators seems appropriate here.
Euclides’ activity does not differ in principle from the activity of Socrates,
the narrator of Republic 1, who “edited for smoothness” his conversation with
Thrasymachus (pp. 37–38), or from the activity of Apollodorus, the secondary
narrator of the Symposium, who edited Aristodemus’ report stylistically and
with respect to the scope (pp. 94–95). Moreover, there is good reason to sup-
pose that the Parmenides’ switching from narrated to dramatic form, analogous
to what we find in the Theaetetus, was also conceived by Plato as having been
initiated by the secondary narrator, Antiphon (above, p. 41). Euclides is thus
neither the author of the story he delivers nor a slavish transcriber of Socrates’
words: the scope of his authority does not seem to deviate significantly from
that of Plato’s other narrators.57

54 For a detailed discussion of Euclides’ editorial work see Kaklamanou and Pavlou 2016:
118–27.
55 See, e.g., Morgan 2003: 105: “Euclides in the Theaetetus performs the part of an author of
Sokratikoi logoi.” See also Tarrant 1996: 133; Halliwell 2009: 16; Capuccino 2014: 122–26.
56 So Blondell 2002: 306: “Eukleides is utterly incapable of producing an oral account, focus-
ing rather on obtaining an accurate verbatim transcription.” See also Halperin 1992: 112
n. 19; Giannopoulou 2014: 44–48. For a discussion see Kaklamanou and Pavlou 2016.
57 For a detailed discussion of the scope of the narrator’s authority in Plato see Chapter 6.
100 Chapter 4

6.3 Narrator-Hero (Implicit): Socrates

On the third intradiegetic level – whose actual status, as Genette has point-
ed out, is blurred (Chapter 1) – the Theaetetus proceeds as a regular dra-
matic dialogue with Socrates as its implicit narrator. Socrates is also the
dialogue’s leading character, the other two being Theodorus of Cyrene and
Theaetetus.
As we saw in the preliminary analysis of this part of the dialogue in Chapter
3, the story begins with Socrates gradually focalizing the information about
Theaetetus that Theodorus supplies, until he eventually fixes his attention on
Theaetetus himself (p. 51). Socrates’ position in the Theaetetus closely resem-
bles the one he holds in the narrated Charmides and Lysis (above, pp. 32–33).
In these two dialogues, Socrates the narrator tells of how he was introduced
to beautiful and promising youths highly praised by their friends and admir-
ers. The irony however is that promising though he certainly is, Theaetetus is
anything but beautiful: he looks very much like Socrates himself, and, just like
Socrates, his lack of physical appeal conceals an extraordinary beauty of mind
and soul. In particular, the entrance of Theaetetus and his friends is staged
as a mirror-image of the entrance of Charmides in the eponymous dialogue
(144b9–c3, cf. above, p. 32). Moreover, Theodorus’ “if he had been a beauty
(kalos) I should have been afraid to praise him, lest you should suppose that
I was in love with him” (143e6–7) is the complete reversal of what we find in
the Charmides.58 All in all, it is hard to avoid the impression that here, tongue
in cheek, Plato revisits this earlier dialogue, with Alcibiades’ Silenus speech
looming large in between.59

6.4 Juxtaposition of the Frame and the Narrative

The Theaetetus is therefore Socrates’ story, as communicated to Euclides, about
how Socrates made Theaetetus’ acquaintance and caused the gifted youth to
deliver his first intellectual offspring. But it is also Euclides’ story of Theaete-
tus’ death as a citizen-soldier as communicated to Terpsion, who absorbs it
together with the reader. As in the Symposium, the juxtaposition of the main
story and the dramatic frame creates a comprehensive retrospective of the life
of the dialogue’s protagonist.

58 Cf. e.g. Chrmd. 154a. See also Chrmd. 154bc, Socrates’ reaction to Charmides’ beauty, quot-
ed above, p. 33.
59 The present interpretation is not related to Thesleff’s hypothesis that the Theaetetus as
we have it is a revised version of a narrated dialogue belonging to Plato’s early period, see
Thesleff 1982: 232–33, 300–304; Tarrant 2010; Schultz 2015. On parallels with the Sympo-
sium see also Blondell 2002: 261.
The Explicit and the Implicit Narrator Combined 101

7 Conclusions

As mentioned earlier, the mixed dialogues were most likely Plato’s own inven-
tion. What could make him open what looks like a regular narrated dialogue
(the Theaetetus being the only exception) with a dramatic episode that not
always even features those who participate in the subsequent discussion?
The answer that immediately comes to mind is that just as in the case of the
Theaetetus discussed in Chapter 1, by dramatizing the act of narration Plato
economizes on one narrative level, thus avoiding the cumbersome arrange-
ment that he tried to pursue in the Parmenides. And indeed, we have seen that
despite their structural complexity, the majority of the mixed dialogues ap-
parently contain only two narrative levels: the pseudo-extradiegetic level em-
bodied in the dramatic frame and the intradiegetic level of the main narrative
(two intradiegetic levels in the case of the Symposium). In fact, however, what
these dialogues present as an extradiegetic act of narration is, as we have seen,
nothing but an additional intradiegetic level in disguise. Just as in the dramatic
dialogues, the extradiegetic act of narration is suppressed here as well.
However, economizing on narrative levels is only one aspect of what Plato
has achieved by introducing the dramatic frame. Consider the following. In-
sofar as in the narrated dialogues the unpersonalized narratee merges with
the story’s implied reader/listener, these dialogues, just as any first- or third-
person narrative, are open-ended by definition: the primary act of narration
always belongs to the extradiegetic level (above, pp. 17–18). By suppressing the
primary act of narration and personalizing the narratee as a full-fledged char-
acter, Plato effectively abolishes the open-endedness inherent in other forms
of narrative by making both the narrator and the narratee intradiegetic. Again,
the main narratives of the Euthydemus, the Phaedo, and the Theaetetus are ad-
dressed not to us but to Crito, to Echecrates, and to Terpsion, just as the main
narratives of the Protagoras and the Symposium are addressed to unnamed
fifth-century Athenians rather than to the humanity at large. To be sure, here
and there Plato compensates for the loss of universal appeal that this arrange-
ment implies – for example, in Crito’s and Echecrates’ metaleptic remarks in
the Euthydemus and the Phaedo respectively (above, p. 92 and n. 42) or in the
anonymity of the narratees of the Protagoras and the Symposium. The ques-
tion however is what he gains instead.
The feature shared by all dialogues of this group is the complex perspective
resulting from juxtaposition of the dramatic frame and the framed narrative.
Everything indicates that this complex perspective was part and parcel of the
message Plato intended to deliver. In the Euthydemus and the Protagoras the
juxtaposition of the frame and the framed allows the reader to contemplate
102 Chapter 4

the discrepancy between what the characters do and say on each one of the
two levels of the dialogue and to reach his or her own conclusions concerning
the complex message which emerges as a result (see also below, p. 115). In the
Phaedo, the Symposium, and the Theaetetus the same juxtaposition creates a
spatiotemporal gap, which allows the reader to establish an overall perspec-
tive of an event (Socrates’ death in the Phaedo) or the protagonist’s life (the
life stories of Socrates and Theaetetus in the Symposium and the Theaetetus
respectively).60 Our ability to establish such a perspective derives from the
temporal and spatial “closing” of the dialogues effected by the introduction of
a personalized narratee.
Above all, however, it was Plato’s lifelong tendency to privilege the dramatic
over the narrated form that apparently triggered the introduction of the hybrid
genre of the mixed dialogue. This tendency will be the focus of our attention
in the next chapter.

60 Cf. Thesleff 1967: 41: “[I]t seems that Plato introduced the ‘frame’ (Prt.) to give an addi-
tional dimension to the work, experimented with it as a throughout counterpoint (Euthd.,
Phd.), then (Smp., Prm.) used it for justifying the increasing perspective of the second-
hand and the third-hand report, and finally (Tht.) to justify his return to the dramatic
form.” Cf. also Capuccino 2014: 293–94: “I proemi esterni … invitano il lettore a consid-
erare la tensione constitutiva della prosa platonica, riflesso della sua natura filosofica e
non retorica” (original emphasis).
 Part 2
The Interpretation

∵
 Chapter 5

Plato’s Experiments with Narrative Voice

1 Preliminary Remarks

In the course of his long career as a writer Plato never stopped experiment-
ing with narrative voice. Assuming that his main concerns were philosophical
rather than literary, it is amazing how much effort he invested in perfecting,
adjusting, and revising this aspect of his literary art. We saw him reducing the
multiple voices of the Euthyphro, the Crito, and the Menexenus to a single nar-
rative voice of the other dialogues; replacing the impartial narrator-observer of
the Hippias Minor with heavily biased ones; abandoning the contrasting narra-
tive voices of the Protagoras and the Euthydemus in favor of the unison voices
of the Phaedo, the Symposium, and the Theaetetus. We also saw him experi-
menting with multi-level narratives by introducing such literary strategies as
dramatic frame and metalepsis. Last but not least, we saw him abandoning the
narrated form and thematizing this strategic decision.
As we have seen, apart from the Euthyphro, the Crito, and the Menexenus, all
Plato’s dialogues are negotiated as narratives. These can be presented as dis-
tributed on a vertical axis. The dramatic and the narrated dialogues except the
Parmenides have two narrative levels, the intradiegetic and the extradiegetic,
but the dramatic dialogues suppress the extradiegetic act of narration, hence
are presented as possessing a single narrative level. Of the so-called mixed dia-
logues, the Protagoras, the Euthydemus, and the Phaedo have three narrative
levels: – two intradiegetic and the extradiegetic, but they suppress the extradi-
egetic act of narration, hence are presented as possessing two narrative lev-
els. Finally, the Symposium, the Theaetetus, and the Parmenides are located at
the top of the pyramid, having four narrative levels: – three intradiegetic and
the extradiegetic. However, the Symposium suppresses the extradiegetic act of
narration and presents itself as having three narrative levels, whereas the The-
aetetus suppresses both the extradiegetic and all the intradiegetic levels, thus
transforming itself into a wholly dramatic dialogue. Of the dialogues of this
group, only the Parmenides makes all its narrative levels explicit.
Since the narrative voice in Plato’s dialogues has been analyzed above in the
categories of contemporary literary theory, and since Plato could not have had
such categories in mind, one must ask what he was aiming at by experimenting
with the narrative voice as he did. For at least an approximate answer, in this

© koninklijke brill nv, leiden, ���9 | doi:10.1163/9789004390027_006

106 Chapter 5

chapter I shall trace what transpire as the major turning points in Plato’s career
as a writer. As far as I can judge, three such turning points emerge:
(a) introduction of a single focus of perception;
(b) introduction of multi-level narratives;
(c) abandonment of narrated form.
Let us examine the directions Plato’s literary composition has taken as a result.

2 A Single Focus of Perception

Early in his career as a writer, Plato removed from his dialogues all but one
focus of perception, thereby investing them with a single narratorial authority.
The latter is explicit in Plato’s narrated texts (the fully narrated dialogues and
the narrated parts of the mixed dialogues) and implicit in the dramatic ones
(the fully dramatic dialogues and the dramatic parts of the mixed dialogues).
Although the model of implicit narrator dealt with in this book was thema-
tized in full only as late as the Theaetetus, its relevance is supported by Plato’s
introduction in Republic 3 of the narrative category “diegesis through mimesis”
(Chapter 1) and by the evidence provided by the dialogues whose narrator is
explicit or explicitly accounted for as suppressed. In all dialogues in this cat-
egory, that is, the narrated dialogues, the narrated parts of the mixed dialogues,
and the Theaetetus, the source of narratorial authority has been shown to co-
incide with the focus of perception. In other words, Plato’s narrator is the one
who focalizes the narrative in its entirety.
This of course does not mean that characters other than the narrator do
not focalize. However, the characters’ focalization never affects the entire dia-
logue: although the characters may well contribute perceptions of their own,
the macro-focalization of the narratives in which they take part is effected by
the narrator (Chapter 2, with n. 37). Characteristically, the characters’ focal-
ization does not usually disagree with the narrator’s perception. Thus Meno
repeats almost verbatim Socrates’ observation that Anytus is getting angry;
Phaedrus intensifies Socrates’ references to the landscape, the time of day, and
the interlocutors’ appearance; Socrates echoes Eudicus’ words about the con-
versation’s setting; Polus and Gorgias reiterate the appeal to the audience initi-
ated by Chaerephon, and so on.1
As we have seen, the markers of focalization that regularly accompany the
explicit narrator of Plato’s narrated texts are (a) a description of the dialogue’s

1 Meno 99e2 and above, n. 30 to Chapter 3; Phdr. 229a9–b9, 242a3–5, 279b5–6 and above, n. 34
to Chapter 3; Hi.mi. 363a3–4, and p. 63 n. 39; Grg. 473e4–5, 506a8–b3, and n. 49 to Chapter 3.
Plato’s Experiments with Narrative Voice 107

setting; (b) a gradual focusing on a single object of attention (“zooming in”);

(c) registration of the characters’ appearance, movements, and emotional re-
sponses; (d) limitation of the scope of the narrative to the narrator’s personal
experience. Since these markers consistently characterize the narrator while
excluding any other character, we have good reason to regard them as signals
of narratorial authority. The same markers of focalization emerge in Plato’s
dramatic texts, where they also point to a single focus of perception. This in-
dicates that the markers of focalization should be approached as signals of
narratorial authority in the dramatic texts as well; accordingly, their bearers
should be identified as implicit narrators of the texts in question. This conclu-
sion allows us to treat narrative voice in the narrated, dramatic, and mixed
dialogues using the same criteria.
In the dialogues employing a narrator-hero – that is, those whose narrator
both focalizes the narrative and leads the conversation,2 – the introduction of
a single focus of perception results in the narrator finding himself in a unique-
ly privileged position. Not that no opinions except for those of the narrator are
admitted: some of Socrates’ opponents in this group of dialogues, for example,
Hippias, Protagoras, Callicles, or Thrasymachus, are very outspoken indeed in
their criticism of Socrates and his views. But all such dissident opinions, as well
as their bearers, pass through the filter of the narrator’s perception, and are ac-
cordingly represented on the narrator’s terms rather than their own. Kathryn
Morgan’s assessment of Socrates the narrator in the Charmides, the Republic,
and the Protagoras seems in place here:

Socrates’ intrusions into the narrative, however, make his own bias overt.
The reader becomes aware that Socrates selects and interprets in line with
his own interests and goals. One might ask how fair Socrates’ interpreta-
tions are (given that they mostly show his opponents in a bad light), and
even question the precision of his edited version of the conversations.3

The effect Morgan describes is entirely due to Plato’s skillful deployment of

focalization: Socrates focalizes Ion, Critias, Hippias, Anytus, Protagoras, Thra-
symachus, but their focalization of Socrates is “switched off” (cf. above, pp.
56–57, on the Ion); only Socrates’ own perception of their embarrassment and

2 The Ion, the Hippias Maior, the Charmides, the Lysis, the Protagoras, the Euthydemus, the
Cratylus, the Meno, the Republic, the Theaetetus, the Phaedrus, the Laws.
3 Morgan 2004: 364. Cf. also Bowery 2007: 106 n. 16: “In the narrated dialogues, each of these
Socratic characters is filtered through the lens of Socrates the narrator.”
108 Chapter 5

anger is registered.4 This leaves us with the spurious Cleitophon as the only
example of a criticism of Socrates not mediated by Socrates himself or by one
of his associates.5
Such unilateral focalization is even more pronounced in the dialogues
where Socrates’ interlocutors are young people whom he tutors, as it were, in
the art of dialectic. This situation, as Ruby Blondell put it, endows him with
“a much more authoritative pedagogical posture.” Blondell’s assessment of
Socrates’ interaction with Glaucon and Adeimantus in Republic 2–10 deserves
to be quoted in full:

Despite this collaborative tone, however, the dialectical enterprise has

now become unabashedly hierarchical. The elenctic Socrates did, of
course, dominate his conversations. But his dominance was both veiled
in self-deprecation and subject to the constraints imposed by the in-
terlocutor. In Books 2–10, by contrast, Socratic dominance is openly ac-
knowledged by all concerned.6

In itself, the hierarchical relationship Blondell refers to has nothing to do with

narrative voice: compare indeed the Symposium, the Parmenides, the Sophist,
or the Statesman, whose leading interlocutors – Socrates, Parmenides, the El-
eatic Stranger – apply the same conversational strategy in narratives delivered
by someone else.7 Nevertheless, it is noteworthy that conversations of this
kind are invariably accompanied by the almost complete removal of traces of
­perception – or indeed of any kind of emotional response – from the younger
interlocutors’ utterances s (cf. above, pp. 36–37, on the scarcity of emotions in
Republic 2–10). In sum, therefore, whether his interlocutors are antagonistic or
supportive, no focus of perception other than the narrator is permitted in the
dialogues whose narrator is also their leading character.
On the surface, the situation should be different in the dialogues that
employ a narrator-observer.8 Not being the dialogue’s leading character, the

4 Cf. Blondell 2002: 197, on Republic i: “Another function of Sokrates’ narrative role, as Plato
exploits it, is to undermine the legitimacy of Thrasymachos’ views by representing them as
inseparable from his peculiarly offensive character.” Cf. also Rowe 2007: 11; Schofield 2008: 63.
On Socrates’ descriptions of his opponents see also above, pp. 33, 35, 82.
5 Cf. Hutchinson 1997: 965: “It comes as quite a surprise to see a Platonic dialogue in which
Socrates is the target of attack and fails to have the last word.”
6 Blondell 2002: 200.
7 Cf. Zuckert 1998: 895, on application of this kind of strategy by Parmenides in the Parmenides;
cf. also Irwin 1986, who proposes a more nuanced approach.
8 The Hippias Minor, the Laches, the Gorgias, the Phaedo, the Symposium, the Parmenides, the
Sophist, the Statesman, the Timaeus, the Critias, the Philebus.
Plato’s Experiments with Narrative Voice 109

narrator-observer is positioned so as to enjoy a degree of impartiality that the

autodiegetic narrator can hardly afford. Yet, the only dialogue in which we
encounter an impartial narrator-observer is the early Hippias Minor. Eudicus
mediates between the opponents without explicitly siding with either; as a re-
sult, his position allows the dialogue to preserve the aspect of bifocality – the
very aspect, it should be added, which became responsible for the interpreters’
perplexity as regards Socrates’ views expressed in this dialogue.9 Considering
that Plato never repeated this kind of arrangement, it seems reasonable to sup-
pose that he ranked his first attempt at employing narrator-observer as noth-
ing short of failure.
The situation with the other dialogues employing a narrator-observer is dif-
ferent. Although these dialogues involve multiple participants who display a
wide array of conflicting opinions, the plurality of the speakers’ opinions is
never accompanied by a plurality of points of view. Whatever opinions the
participants may hold, only one perspective is available, that of the narrator.
As with Thrasymachus in Republic i, the result is that we do not really know
how the people expressing various opinions perceive their interlocutors or the
conversation in which they take part.10 Their points of view are communicated
to us, if at all, through the mediation of the narrator-observer.11
Although no longer the narrator, in the dialogues employing a narrator-
observer Socrates is still the leading character.12 Yet rather than focalizing the
narrative, now he himself is being focalized. We saw how this was done in the
Parmenides, the Laches, the Gorgias, the Phaedo, and the Symposium. Charac-
teristically, only in the dialogues employing a narrator-observer can we visu-
alize Socrates the man and hear his praises. The Symposium and the Phaedo
show us how Socrates looked and behaved in everyday life; the Parmenides
shows Zenon’s and Parmenides’ admiration for Socrates as a youth; the Laches
and the Symposium extol Socrates’ military record; the Gorgias presents clearly
and unambiguously his choice of the philosophical life; and the Phaedo deliv-
ers his intellectual biography and portrays in detail how he met his death. None
of this would have been possible had the narrator-hero been the only kind of

9 Above, p. 64. See also p. 66, on the difference between Eudicus of the Hippias Minor and
Lysimachus of the Laches.
10 Embedded narratives being the only exception; see e.g. Alcibiades’ reminiscences in the
Symposium.
11 Cf. above, p. 95, on Aristodemus’ selective presentation of speeches in the Symposium.
See also above, Chapter 2, on the Parmenides: the reaction of Parmenides and Zeno to the
youthful Socrates’ exposition of the Theory of Forms is explicitly presented in terms of
Pythodorus’ perception of it (pp. 42–43).
12 The dialogues employing Socrates himself as narrator-observer constitute a special case;
they are treated separately in Section 4 below.
110 Chapter 5

narrator Plato employed. The reason is simple: as we have observed more than
once, in the dialogues in which he assumes the role of narrator Socrates “sees”
in the Genettian sense13 without being “seen.”
It follows then that both the dialogues employing a narrator-hero and those
employing a narrator-observer unambiguously privilege Socrates and his argu-
ments. This seems to indicate that, contrary to the widespread view, Plato’s
dialogues cannot be considered multivocal, or “dialogic,” in Bakhtin’s sense.14
Malcolm Schofield’s observation that at least in some respects the Ciceronian
dialogues may claim to be more truly dialogic than the dialogues of Plato is a
helpful caveat in this respect:

In dialogues such as Academica, De Finibus, De Natura Deorum and De

Divinatione Cicero gives properly argued alternatives a real run for their
money, and adopts a variety of literary strategies for indicating that fur-
ther reflection on their merits and choice between them is left to the
reader.

However, no such opportunity is provided, for example, to the reader of Plato’s

Republic, where “[c]onversion to the Socratic viewpoint is the objective, not
balanced consideration of the alternatives.”15
Consider indeed the following. Focalized by Socrates himself or by one of
his associates, beliefs and opinions of others are habitually paraded at the
beginning of the dialogue, only to be subsequently demolished or rejected
outright: Laches’ view of courage, Lysis’ view of friendship, Charmides’ view
of temperance, Meno’s view of virtue, Gorgias’ view of rhetoric, Theaetetus’
view of knowledge, the first five speeches about love in the Symposium, Lysias’
speech in the Phaedrus, and so on.16 Only Socrates’ view is privileged. And al-
though it cannot be denied that Plato’s dialogues usually end in aporia (thus
in the earlier dialogues) or in indeterminacy (thus in the middle and the later
ones), neither is ever effected by the arguments of Socrates’ opponents.
This does not mean however that Schofield is fully justified in claiming that
no properly argued alternatives are presented to Plato’s reader. But rather than
from a plurality of voices, these alternatives issue from a single voice, that of

13 Genette 1980: 186.

14 For the most recent exposition of the thesis of Plato’s multivocality see Schur 2014: 19–40.
15 Schofield 2008: 63, 64. Cf. Rowe 2007: 32.
16 Cf. M. Frede 1992: 212, on the aporetic dialogues: “It is not his [the questioner’s] beliefs and
his authority which are under test, but the respondent’s.” The argument in Sheffield 2006
as regards Plato’s “endoxic method” in the Symposium seems to be applicable to other
dialogues as well.
Plato’s Experiments with Narrative Voice 111

the dialogue’s protagonist. Socrates’ two speeches in the Phaedrus, one neu-
tralizing the other and being neutralized in turn by the subsequent discussion,
immediately come to mind in this connection. The Phaedrus, however, is just a
particularly eye-catching example of numerous other cases where the protago-
nist presents his own alternative views of the same issue in one and the same
dialogue. This is the main cause of the indeterminacy with in Plato’s dialogues
frequently end.17 As Donald Davidson remarked apropos the Euthydemus, the
Hippias Maior, and the Lysis: “In these dialogues there is philosophical argu-
ment, but Socrates carries on essentially by himself, acting both as proposer
and as critic.”18
In view of this, it would be no exaggeration to say that the truly dialogic as-
pect of Plato’s narratives lies in the protagonist’s own mind, that is, ultimately,
in the mind of the author. The following passage from the Theaetetus speaks
volumes in this respect:

[By thinking (to dianoeisthai)] I mean the conversation (logon) which the
soul holds with herself in investigating any subject. I am now telling you
something which I do not really understand, but the soul when thinking ap-
pears to me to be doing nothing else as conversing (dialegesthai) – asking
questions of herself and answering them, affirming and denying. And
when she has arrived at a decision, either by a long circuit or by a sudden
leap, and has at last agreed, and does not doubt, this is called her opin-
ion (doxan). I say, then, that to form an opinion is to speak (legein), and
opinion is a word spoken (logon eirēmenon) – I mean, to oneself and in
silence, not aloud and to another.19

Michael Frede comments:

Notoriously Plato believes that thinking, that arriving at some view con-
cerning some question at issue, is a matter of internal dialogue one’s
reason engages in with itself (cf. Theaet. 189E, Soph. 263E). Thus, one
might think, the dialectical discussion and hence the written dialogue
are supposed to reflect, to be some kind of materialization of, the internal

17 See e.g. Ly. 218c, Resp. 343a, Prm. 144d, Tht. 195bc, Soph. 239b, Plt. 262c.
18 Davidson 2013: 6.
19 Tht. 189e6–190a6. Cf. Soph. 263e, Phlb. 38ce. As Ruth Scodel points out to me, Hippias
Maior 304d1-e2, where Socrates says that he is constantly questioned (304d2 ἀεί με
ἐλέγχοντος) by a person who is his closest relative and dwells in the same place, actually
amounts to the same thing. For a nuanced discussion of this passage see Long 2013: 46–63.
112 Chapter 5

dialogue of the soul, of reason, with itself. And it is no doubt that

in these terms Plato himself also thinks of dialectical argument, e.g. when
he makes Socrates say that we should follow the logos (i.e. the reason or
the argument) wherever it leads.20

The result is an interaction of multiple arguments played out in Plato’s mind.

The manner in which they are arranged seems to indicate that while Plato
­certainly privileged some arguments over others, he deliberately abstained
from endowing the argument he privileged with the status of sole embodi-
ment of absolute truth.
To recapitulate, at the beginning of his literary career Plato made a choice
between bifocal presentation, still manifest in some of his earlier dialogues,
and a single narrative perspective through which everything found in the text,
including diverse opinions expressed by the characters, is filtered. He predi-
cated this single perspective on the figure of the first-person narrator who fo-
calizes the text in its entirety.

3 Multiple Narrative Levels

Everything indicates that Plato’s multi-level narratives as represented by the

mixed dialogues and the Parmenides developed out of such simply structured
narrated dialogues as the Charmides and the Lysis (above, pp. 80, 101). Yet, by
introducing what is sometimes seen as hardly more than topping the narrated
dialogues with a dramatic, or in the case of the Parmenides a narrative frame,
Plato created narrative opportunities not provided by the dialogues with a
single narrator.
It may seem strange that a frame story, familiar to us from such classical
texts as The Thousand and One Nights, The Canterbury Tales, or The Decam-
eron, only emerges in Plato as late as the Parmenides and comes after a series
of dialogues that employ a frame drama instead, namely the Protagoras, the
Euthydemus, the Phaedo, and the Symposium. Moreover, Plato was obviously
not satisfied with his only attempt at producing a frame story, for he aban-
doned it immediately after the Parmenides (if not in the course of it; see above,
pp. 7–8) and the dramatic frame was restored, one last time, in the Theaetetus.

20 M. Frede 1992: 218. See also Sedley 2003: 1–2, on Plato’s dialogues as externalization of
his own thought-processes; contra Rowe 2007: 8 (with n. 20), 12, 33–35. For a book-length
treatment of the issue see Long 2013.
Plato’s Experiments with Narrative Voice 113

But Plato’s adoption of frame drama rather than frame story will look much
less peculiar if we take into account that the Greek literary tradition had no
precedent for a full-scale framed narrative. A likely candidate is of course the
so-called Apologue, Odysseus’ tale of his wanderings narrated to the Phaea-
cians in Books 9–12 of the Homeric Odyssey. But as distinct from The Thousand
and One Nights and other framed narratives, the Apologue’s narrative frame
does not embrace the entire Odyssey, and Odysseus is not the poem’s primary
narrator. It is rather a story within a story, so that a parallel with, for example,
Socrates’ tale of his meeting with Diotima in the Symposium would be more ap-
propriate here.21 It is reasonable, therefore, to credit Plato, among other things,
with creating the first full-scale frame narratives in Western literary tradition.
Judging by the brevity and simplicity of its dramatic frame, it is not out of
the question that the Protagoras was Plato’s first experiment with multi-level
narrative.22 As we have seen, the introduction of an additional narrative level
afforded Plato the opportunity to establish a perspective on the great sophist
that did not fully harmonize with the one found in the body of the dialogue.
This invested the Protagoras with narrative possibilities not available in the
single-level dialogues.
The Euthydemus, a dialogue whose handling of the dramatic frame is oth-
erwise very similar to what we find in the Protagoras (above, pp. 101–102), in-
augurated yet another breakthrough. Indeed, everything indicates that it was
in this dialogue that metalepsis, isolated examples of which had already been
present in the Charmides and the Protagoras (above, pp. 34, 82–83), was first
introduced as a full-fledged narrative strategy consistently applied through-
out the dialogue. The same strategy was pursued to even greater effect in the
Phaedo. In both the Euthydemus and the Phaedo metalepsis affects the entire
dialogue, becoming an integral part of the plot structure. Characters belonging
to the dramatic frame intrude into the story, collapsing the barrier between
different narrative levels and making the reception of the dialogue an integral
element of the plot (above, Chapter 4).

21 Cf. Wolf 2006: 183: “If it [the criterion of dominance] is employed in a quantitative sense
only, Homer’s Odyssey cannot be classed as a frame story, for Ulyssess’ embedded stories
only comprise parts of book vii and books ix to xii…” In Genette’s classification, Odys-
seus’ status in Odyssey 9–12 is that of the intradiegetic-homodiegetic narrator, “a narrator
in the second degree who tells his own story”; see Genette 1980: 248, cf. de Jong 2004: 2. On
the distinction between a story within a story and frame narrative see Fludernik 1996: 257.
On the distinction between Odysseus’ narrative and Socrates’ narrative in the Republic
see Ferrari 2010: 22.
22 Cf. Thesleff 1967: 41, quoted above, n. 60 to Chapter 4.
114 Chapter 5

As already pointed out, Plato’s practice of frame-breaking as pursued in his

mixed dialogues should not be confounded with breaking the boundaries of
the fictional world through intrusion of the real world into the fictional one,
or vice versa; when put into narratological terms, this would amount to inter-
penetration of the extradiegetic narrative level and the intradiegetic domain
(above, p. 22). In the ancient Greek literary tradition this kind of metalepsis
is best exemplified by parabasis, the chorus’ direct address to the audience,
which was an integral part of the comedies of Aristophanes and Old Attic
Comedy generally.23 But the only instances of Plato’s use of metalepsis that
involve the extradiegetic level are the Charmides – Socrates apostrophic ad-
dresses to his extradiegetic narratee purporting to express his admiration for
Charmides’ beauty, and Republic 1 – Socrates’ remark about the extradiegetic
act of narration in which he takes part (above, Chapter 2). It is hardly mere
chance that both these instances belong to the dialogues with only one intradi-
egetic level (see p. 105).
In the multi-level dialogues, the situation is different. Crito of the Euthyde-
mus and Echecrates of the Phaedo only represent the real-life narratees of
Socrates and Phaedo, respectively, who in turn only represent extradiegetic
narrators.24 Accordingly, the breaking of boundaries produced by Crito’ and
Echecrates’ intrusions into the narrated world is fully enclosed within the in-
tradiegetic domain. This can be illustrated by the following example of intradi-
egetic frame-breaking highlighted by Werner Wolf:

Where this device [metaleptic frame-breaks] is used, a paradoxical trans-

gression takes place between framing border and framed text. This can,
for instance, be seen in Michael Ende’s Die Unendliche Geschichte (1979),
where the principal character of the framing, a little boy reading a fairy
tale book, is suddenly able to enter the world of this fairy tale not only
metaphorically … but in person.25

23 On the parabasis as marking “a distinctive moment of frame-breaking” see Whitmarsh

2013: 13.
24 On the pseudo-extradiegetic status of the dramatic frame in Plato’s mixed dialogues see
above, pp. 77–78.
25 Wolf 2006: 201–202. Wolf proceeds as follows: “Often metalepses (which, of course, can
also take place between the extradiegetic and the intradiegetic levels) serve a recipient-
centred function (by undermining aesthetic illusion) as well as an implicitly metafiction-
al function: by stretching the (fantasy) potential of imaginary worlds to an unrealistic,
even ‘impossible’ extent, metalepses lay bare the ruling frame of such worlds which is
usually concealed, namely their fictionality.”
Plato’s Experiments with Narrative Voice 115

This is precisely what Crito of the Euthydemus and Echecrates of the Phaedo
do. Wolfgang Petersen’s 1984 film based on Ende’s book, The Never Ending Story,
in which the process of frame-breaking is vividly enacted, gives an excellent
illustration of the effect produced.
In all the cases involved, metalepsis works in Plato as a highly effective dis-
tancing strategy. The audience is reminded of the fictionality of the narrated
world and is prevented from fully immersing in it. Having several narrative
voices intrude into the narrated world of the Symposium and the Parmenides
(above, pp. 43, 96) is obviously meant to serve the same purpose. To quote
Richard Hunter, “Plato’s various experiments with dramatic frames show his
persistent concern to advertise and problematize the fictional status of his
dialogues.”26
Beyond metalepsis, there is no denying that in the Phaedo, the Symposium,
the Parmenides, and the Theaetetus Plato put the multi-level narrative to a dif-
ferent use from that in the Protagoras and the Euthydemus. We saw in Chapter 4
that in those two the dramatic frame on the one hand and the framed narrative
on the other introduce contrasting perspectives on Protagoras and Euthyde-
mus and their arguments, thus inviting readers to form their own opinion. In
a manner recalling the Euthyphro and the Crito, the frame and the narrative
of the Protagoras and the Euthydemus present diverse points of view that are
not mediated by one overarching narrative voice nor harmonized to produce
a uniform message. Not so with the other multi-level dialogues. In the Phaedo,
the Symposium, the Parmenides, and the Theaetetus Plato uses the Chinese-box
arrangement not in order to introduce contrasting perspectives but in order to
multiply the perspective of the main story to deepen its temporal dimension
and create a multi-layered but uniform picture of the dialogue’s subject.
Characteristically, in both the Protagoras and the Euthydemus the spatio-
temporal opportunities offered by the multi-level arrangement are left un-
explored. In the Protagoras, the encounter of the narrator and the narratee
occurs on the same day as the events of the narrated story, whereas in the Eu-
thydemus it occurs the next day. This indifference to time depth clearly indi-
cates that at least as regards these two dialogues the author’s agenda should be
sought elsewhere. The situation is different in the other multi-level dialogues.
The time gap between the dramatic frame and the framed narrative, in the
Phaedo still bridgeable by a single narrator, in the Symposium, the Parmenides,
and the Theaetetus becomes so substantial that a chain of several narrators has

26 Hunter 2004: 22, and above, p. 96 n. 50. For an analogous interpretation of frame-breaking
in Plato see McCabe 2008: 105–106. On Plato’s distancing strategies in general see Finkel-
berg 2006: 65–68.
116 Chapter 5

to be used; moreover, in the Phaedo and the Theaetetus the dramatic frame and
the framed narrative lie distant from each other in space as well as time.
Rather than offering two contrasting points of view, in the Phaedo, the
Symposium, the Parmenides, and the Theaetetus the frame and the framed are
fully in accord. Each successive level echoes and further enhances the signifi-
cance of the dialogue’s subject – be it Socrates’ life and death, the meeting of
Socrates and Parmenides, or the life story of Theaetetus – without questioning
the p­ erspective in which it is presented. Such a transition from diverse narra-
tive perspectives to a uniform one parallels a similar development which, as
we have seen, took place in the dramatic dialogues.
Yet, despite their giving rise to such indisputable masterpieces as the Pha-
edo and the Symposium, multi-level narratives were just an episode in Plato’s
literary career. They reached the peak of complexity with the Symposium, only
to be abandoned altogether soon after. The Parmenides, again, seems to have
played a decisive role in this development. It is indeed difficult to see this
dialogue’s structural arrangement as anything other than a failed attempt to
replace the dramatic with the narrative frame. Apparently purporting to em-
phasize the interval between the moment of narration and the chronologically
implausible encounter of Socrates and Parmenides, here Plato, for the first and
only time in his literary career, made all narrative levels explicit. Judging by the
sudden abandonment of reported speech in the second part of the dialogue,
already in the process of writing he came to the conclusion that this innovation
was not worth pursuing. In the Theaetetus, Plato’s last multi-level dialogue, the
dramatic frame was restored, but the narrated form was abandoned altogether.
To all intents and purposes the Theaetetus is a dramatic dialogue, as are the
dialogues that followed it.

4 Abandonment of the Narrated Form

4.1 The Transitional Dialogues

It has been suggested more than once that the Theaetetus’ programmatic aban-
donment of the narrated form signaled Plato’s adoption of dramatic dialogue
as his only literary medium. Indeed, all Plato’s late dialogues – the Sophist,
the Statesman, the Timaeus, the Critias, the Philebus, the Laws – are dramatic
through and through.27 That is not to say that the decisive turn to the dramatic

27 Cf. Thesleff 1967: 15–16, 37; Guthrie 1978: 64; Brandwood 1990: 1, 251; Sedley 2002: 16 and
n. 26; Morgan 2004: 360. Erler 2006: 53, also includes the Parmenides in this list, appar-
ently on account of its being partly dramatic; formally, however, the Parmenides is a nar-
rated dialogue, not to mention the fact that Tht. 183e, Socrates referring to his meeting
Plato’s Experiments with Narrative Voice 117

form was effected solely by the agency of the Theaetetus. We have seen that in
the narrated Parmenides, which must have been written in close proximity to
the Theaetetus, dramatic form also eventually prevailed.
The Parmenides stands midway between the dialogues that, like the Phaedo
and the Symposium, employ a narrator-observer who focuses his attention on
Socrates and the later ones, such as the Sophist and the Statesman, in which
Socrates himself assumes the role of narrator-observer. While in the narrated
part of the dialogue everything serves to showcase the young Socrates and his
accomplishments (above, pp. 43–44), in the dramatic part Socrates becomes
one of the silent witnesses of the dialectical procedure carried out by Par-
menides with the assistance of Aristoteles, his chosen interlocutor. This section
of the Parmenides brings us very close indeed to those late dialogues in which,
as in the Sophist and the Statesman, Socrates is presented as an observer and
silent witness of a dialectical inquiry led by someone else (above, pp. 72–73).
A similar twofold arrangement is also characteristic of the Phaedrus. Wheth-
er written before or after the Theaetetus (both opinions circulate in the schol-
arly literature), the prevailing consensus is that the Phaedrus belongs to the
cluster of the “late middle” dialogues, which also comprises the Republic (or
parts of it), the Parmenides and the Theaetetus.28 Both philosophically and sty-
listically the dialogue falls into two distinctly heterogeneous parts. According
to Charles Kahn, this heterogeneity is due to the Phaedrus being a transitional
“Janus-like” dialogue, looking backwards and forwards:

And whereas the myth of the first part is built around the doctrine of
Forms as known from the Phaedo and Republic, the philosophical center
of the second part is an account of dialectic that ignores the metaphysi-
cal Forms but announces a logical technique of definition by Collection
and Division – by determining unities and pluralities – that will be ex-
emplified in Plato’s later writings, the Sophist, Statesman, and Philebus.29

The dichotomy diagnosed by Kahn is neatly reflected in the way the dialogue is
presented to the reader. On the one hand, the first part of the Phaedrus is indis-
putably the visually richest text in the entire Platonic corpus. Its ­rendering of the
summer landscape along the banks of the Ilissus, put in Socrates’ mouth (quot-
ed above, p. 61), is justly acclaimed as one of the finest descriptions of ­nature

with Parmenides, strongly suggests that when this dialogue was written the Parmenides
already existed. Cf. Rutherford 1995: 273, and above, Chapter 1, n. 19.
28 See, e.g., Thesleff 1967: 8; Thesleff 1982: 153–63 (a historical survey), 317–26; Brandwood
1990: 250–51; Kahn 1996: 48, 380–82; Kahn 2002; Irwin 2008: 79.
29 Kahn 1996: 372. On the plot structure of the Phaedrus see also below, pp. 126–27.
118 Chapter 5

in the entire body of ancient Greek literature. But the Phaedrus’ ­privileging of
descriptive language concerns not only the landscape. As we have seen, the
interlocutors’ body language is described frequently and in detail (Chapter 3).
No less significantly, such descriptions are often used for plot building (more
below, pp. 125–27). But all this exuberance disappears as soon as we reach the
dialogue’s second part: the transformation in style mirrors that in the philo-
sophical content. The stylistic gap thus produced resembles that between
Republic 1 and Republic 2–10 or between the narrated and the dramatic parts
of the Parmenides. Plato’s ability to radically transform his style in one and the
same dialogue shows that contrary to what many are ready to admit, even in
his dialogues that may look as less rewarding aesthetically (the late dialogues
immediately come to mind) Plato was in full command of his literary art.30
As if trying to make a point, Plato demonstrates again and again in the Pha-
edrus’ first part that a dramatic dialogue is suited to accomplish everything
a narrated dialogue does, and even more. In the second part, however, he no
less convincingly drives it home that this is not the kind of dramatic dialogue
he ­intends or indeed recommends pursuing after all. “Here, as at the end of
the Tempest, the magician breaks his wand and embarks on a new course.”31
It is difficult to avoid the impression that the Theaetetus and the Phaedrus
complement each other here. While the Theaetetus makes a statement to the
effect that the explicit narrator is in fact redundant, the Phaedrus shows how
a text with no explicit narrator can nevertheless function as a full-fledged nar-
rative. That is why in everything regarding narrative voice the Phaedrus should
be seen as no less programmatic than the Theaetetus. Togeher with the Par-
menides they inaugurate Plato’s late period.32

4.2 The Late Dialogues

Plato’s late period saw intensive experimentation with narrative voice. First
and foremost, Plato radically reduced the amount of overtly marked focaliza-
tion, mainly by concentrating its markers in the dialogues’ openings: this is the
hallmark of both the Sophist-Statesman and the Timaeus-Critias (Chapter 3).
In the Laws the markers of focalization, though present in the body of the text,

30 Cf. Rutherford 1995: 276–81.

31 Kahn 1996: 382.
32 Thesleff 1967: 18, explicitly associates the Republic, the Theaetetus, and the Phaedrus with
Plato’s final abandoning of reported form. On the history of the study of relative chronol-
ogy of Plato’s dialogues, in which the establishment of the position of the Phaedrus had
an important role, see Thesleff 1982, esp. 153–63; Brandwood 1990; Kahn 1996: 42–48. For
a useful overview of the current scholarly consensus (“the standard view”) regarding the
chronological order of the dialogues see Irwin 2008: 77–84.
Plato’s Experiments with Narrative Voice 119

are scarce, especially considering the dialogue’s length. Even in the Philebus,
the most vivid of the late dialogues, the markers of focalization repeatedly in-
timate intellectual rather than sensual perception (see above, p. 69).
All this results in the late dialogues’ much more austere atmosphere. As
Benjamen Jowett put it in the Introduction to his translation of the Statesman:
“He has banished the poets, and is beginning to use technical language.”33 The
vividness of the early and middle dialogues, mainly due to the narrator’s visual
acknowledgement of the characters’ appearance, movements, and emotional
states, has given way to abstract language and almost impersonal descriptions
of predominantly intellectual responses. The single focus of perception is still
there, but we are never told what the Eleatic Stranger, the Younger Socrates,
Timaeus, Philebus, or even the youthful Protarchus look like or what their feel-
ings might be.
It is tempting to associate this change with the vision of the commendable
form of mimesis outlined in Republic 3. Let us quote it again:

For the sake of utility we for our part would employ the more austere and
less pleasant poet or storyteller, who will imitate for us the style of the
decent man, and will follow those models which we prescribed at first
when we began the education of our warriors.34

This change in the manner of presentation can be traced both to the exchange
between Parmenides and Aristoteles in the second part of the Parmenides and
to the discussion of speaking and writing in the second part of the Phaedrus
addressed in the previous section. Both contain in nuce such a hallmark of the
late dialogues as separation between the intense focalization at the beginning
and its almost complete absence in the body of the dialogue.
The Philebus inaugurated an innovation of a different kind. Plato reverses
the perspective of Socrates’ “pedagogical” conversations, in that he focalizes
the dialogue through Socrates’ young interlocutor rather than through Socrates
himself (above, pp. 69–70). Together with Protarchus and his friends, the read-
er takes an active part in the process of acquiring knowledge, which results in

33 Jowett 1892: 2. On “an unmistakable move towards blandness and lack of individuality in
the characterization of both the dominant speaker and his interlocutors” in the Sophist
and the Statesman see Blondell 2002: 316; on the lack of stylistic characterization in the
late dialogues in general see Thesleff 1967: 132; Rutherford 1995: 274–76.
34 Resp. 398a8–b4, cf. above, p. 4. Cf. also Resp. 396c7–d1 “he will be most ready to imitate the
good man when he is acting steadfastly and sensibly” and Plt. 286d4–6 (on the dialogue’s
length) “for we shall never require such a length as is suited to give pleasure, unless as a
secondary matter” and below, Chapter 6, with n. 31.
120 Chapter 5

a sweeping demonstration of Socrates’ beneficial influence on his audience as

explicitly acknowledged by Protarchus.35
Finally, the smooth and uneventful conversation of three old men depicted
in the Laws is unique in that it abolishes the pattern of both the “elenctic” and
“pedagogical” discussion. To quote R.B. Rutherford,

The relationship between the three speakers does not change: through-
out, the Athenian is the authoritative figure, and the other two follow his
lead, with only occasional dissent or surprise (much more often, enthu-
siastic approval). There are no sudden interventions by others, and the
environment makes little impact on their lengthy stroll.36

Considering that here, for the first time in Plato, Socrates is not found among
the interlocutors, it is difficult to avoid the conclusion that Plato must have
regarded the narrative form of the Laws as boldly experimental.
Let us return for a moment to the Sophist-Statesman and Timaeus-Critias.
In these four dialogues Socrates assumes a new role of narrator-observer – the
“theater,” as he refers to himself on one occasion (p. 73). However, this role
is negotiated differently from the other dialogues that employ the same type
of narrator. In the Gorgias, the Phaedo, or the Symposium the narrator’s me-
diating presence is felt all along. Even in the Laches, whose implicit narrator
withdraws into the background at a relatively early stage, he re-emerges at the
end of the dialogue, thereby emphasizing the continuity of his involvement
(above, p. 66). On the other hand, in the Sophist, the Statesman, the Timaeus,
and the Critias, after appearing in the opening episode Socrates ceases to par-
ticipate in the conversation. Formally one of the participants, now he in fact
takes on the role of the dialogue’s audience. Significantly, as the theater meta-
phor of the Critias makes clear, his reception of the dialogue is not only audial
but also visual. The implications of Plato’s placing the narrator in the position
of a spectator of his own narrative will be further explored in Chapter 7.
To sum up, Plato’s dissatisfaction with narrated form, first expressed in the
transitional dialogues, in the late dialogues not only crystallized but also led
to a series of literary experiments. It would perhaps be somewhat simplistic

35 In his concluding words in the eponymous dialogue Theaetetus also acknowledges the
positive impact that the conversation with Socrates has had on him; see Tht. 210b6–7: “By
Zeus, Socrates, I have said a good deal more than I ever had to say.” On the whole, however,
it is Socrates’ point of view as regards the beneficial influence of his maieutic skills that
prevails throughout the dialogue. On Socrates’ attitude to Theaetetus see Blondell 2002:
252–56.
36 Rutherford 1995: 302.
Plato’s Experiments with Narrative Voice 121

to suggest, together with Leonard Brandwood, that Plato’s abandoning the

narrated form was a by-product of the “wearisome repetition” of “I said” and
similar expressions in the Republic, whose last books were presumably written
at the same time as the transitional dialogues or immediately before them.37
However that may be, it is noteworthy that, as we shall see in Chapter 7, the
Republic, the transitional dialogues, and the late dialogues form a continuum
also in that all of them give expression to a train of thought, absent from Plato’s
earlier work, which equates the dialogue to a work of visual art.

5 Conclusions

It would be tempting to suggest that Plato started his literary career by com-
posing both narrated dialogues, a subgenre of the Socratic dialogue which he
explored alongside the other Socratics (pp. 28–30), and dramatic dialogues,
which were probably his own invention (pp. 49–50),38 and that he subsequently
modified the new subgenre of dramatic dialogue by introducing a single focus
of perception. But it is equally possible that for some time during his early
career Plato had simultaneously explored both the bifocal and the unifocal op-
tion, until the latter eventually prevailed. Be that as it may, this much is certain:
Plato’s treatment of narrative voice must have passed through the following
stages:
(a) At the first stage Plato introduced the bifocal dramatic dialogues, with or
without an impartial mediator, probably modelled on mimes (the Euthy-
phro, the Crito, the Menexenus, and the Hippias Minor) and modified the
narrated dialogues by adding a dramatic frame,39 thereby turning them
into multifocal multi-level narratives (the Protagoras and the Euthyde-
mus). This development did not involve the purely narrated dialogues
with their inbuilt single narrative voice (the Charmides and the Lysis).
(b) At the second stage Plato turned from a multiplicity of voices, clearly dis-
cernible in both the dramatic and the mixed dialogues of the previous
stage, to a single narrative voice in the dramatic dialogues (the Ion, the
Hippias Maior, and the Meno) and several unison voices in the multi-level
dialogues (the Phaedo and the Symposium). Moreover, in the dramatic

37 Brandwood 1990: 251.

38 A similar kind of development is suggested in Thesleff 1982: 202–207.
39 Cf. Thesleff 1982: 207; at the same time, I do not share Thesleff’s contention that the nar-
rated parts of the mixed dialogues previously existed as independent compositions; see
above, Chapter 4, with n. 5.
122 Chapter 5

dialogues he replaced the impartial mediator of the Hippias Minor with

the heavily biased one of the Laches and the Gorgias, thereby rendering
all his dialogues unifocal. As a result, the narrative voice of the dramatic
and the mixed dialogues, on the one hand, and of the narrated dialogues
(the Republic) on the other, came to be treated along the same lines.
(c) At the third stage Plato abandoned the subgenre of narrated dialogue,
explicitly defending the status of dramatic dialogue as a form of narrative
(the Theaetetus). At the same time, the strategy of a single narrative voice
or multiple narrative voices in unison continued to be strictly observed:
the former is exemplified by the Phaedrus, the latter by the multi-level
Parmenides (apparently the last narrated dialogue written by Plato)40
and by the dramatic but still multi-level Theaetetus.
(d) At the fourth stage Plato abandoned the multi-level arrangement al-
together. The strategy of the single narrative voice was retained, albeit
modified. Plato sharply reduced the narrator’s text, that is, the manifes-
tations of narrative voice as expressed in the markers of focalization,
mostly by restricting it to the dialogues’ openings (the Sophist, the States-
man, the Timaeus, and the Critias). At the same time, as we shall see in
the next chapter, he significantly increased the number of metanarrative
comments of the characters.
Although diachronically arranged, the developments described above are not
necessarily chronological, at least not before the Theaetetus. As we have seen
more than once, Plato was in full command of his literary style. He could com-
bine two entirely different stylistic registers in the same dialogue (the Pha-
edrus), to reproduce in the minutest detail the style and atmosphere of his
early dialogues (the Theaetetus), or to breathe life and humor into what was
supposed to be a rigorous dialectical inquiry (the Philebus). But the general
leaning to the exclusive use of dramatic form is unmistakable, and it can only
be explained as a matter of deliberate choice.
In a sense, Plato’s entire literary career can be presented as a series of at-
tempts to replace the literary form of narrated dialogue by that of dramatic
dialogue – without, at the same time, abandoning the latter’s identification
as a form of narrative. So much so that at some stage he conceived the idea of
providing the narrated dialogue with a dramatic frame, thus blurring its actual
mode of presentation. This process found its ultimate expression in what looks
like Plato’s radical re-evaluation, or rather in a manner reminiscent of Tolstoy’s

40 Cf. Rutherford 1995: 273–74: “the most obvious point is that narrated dialogues come to an
end with the Parmenides.”
Plato’s Experiments with Narrative Voice 123

late years, outright rejection of much of his earlier literary output, which even-
tually resulted in his complete abandonment of narrated form.
It seems to follow from this that Plato strived to make his dialogues behave
like narratives without looking like them. To achieve this purpose he aban-
doned the explicit narrator but kept the signals of his authority intact. It was
obviously important to him that his dialogues be filtered through a single focus
of perception. This key role was delegated by him to the implicit narrator.
Still, it cannot be denied that Plato’s treatment of purely mimetic texts –
and the dramatic dialogues are after all purely mimetic texts – as narratives is
problematic in more than one respect. We cannot, however, properly address
this issue until we have examined the place of the narrator, explicit or implicit,
in the inner fabric of Plato’s dialogues.
 Chapter 6

The Limits of Authority

1 Preliminary Remarks

The analysis of narrative voice in the previous chapters amply demonstrates

that the authority exercised by Plato’s narrator primarily concerns the way the
dialogue is presented to the reader: the narrator is the eyes, as it were, through
which everything happening in the dialogue is perceived and cognitively ab-
sorbed. Occasional references to the narrator’s ability to edit the text, stylisti-
cally and in respect of scope (Socrates in Republic 1, Phaedo in the Phaedo,
Apollodorus in the Symposium, Euclides in the Theaetetus), accentuating as
they do his responsibility for the form in which the narrative is delivered, also
belong to the sphere of presentation. Does the authority of Plato’s narrator
reach beyond this sphere, to the story and the plot of the dialogue?1
For the “story,” the answer readily suggests itself. The narrator is not respon-
sible for the dialogue’s story because it is conceived as an argument or a se-
ries of arguments raised in a conversation that is negotiated as having really
occurred. Plato’s narrators are invariably presented as eyewitnesses reporting
real events, or as those who transmit such reports further on, so the narrator
cannot be deemed responsible for the dialogue’s setting, its cast of characters,
or the arguments they make. For example, the party at Agathon’s house and the
participants’ speeches can in no way be seen as invented by Aristodemus, an
eyewitness, or moreover by Apollodorus, who retells the story. The influence
of both narrators on the story’s final form may well be considerable (above,
pp. 94–96), but neither Aristodemus nor Apollodorus is envisaged as having
created it ex nihilo.
The narrator’s relation to the dialogue’s plot however is a different matter. It
is often taken for granted that Plato’s narrator not only presents the story to the
reader but also fashions the plot of the dialogue, that is, that he assumes the
role of the author. This impression seems corroborated, as we have observed
more than once, by Plato frequently locating the narrator’s text at structurally
important junctions (pp. 45–46, 75). “Frequently,” however, is not the same as
“regularly,” a circumstance that has prevented us from regarding the strategic

1 On the terms “story” and “plot” as used in this book see n. 25 to Chapter 1 and below in this
section.

© koninklijke brill nv, leiden, ���9 | doi:10.1163/9789004390027_007

The Limits of Authority 125

placement of the narrator’s text as one of the signals of narratorial authority

(p. 75). The reason will become clear once we consider that the narrator’s text
is by no means the only tool with which Plato’s dialogues are structured, or
emplotted.2
As I argued in Chapter 1, in the literary genre of Socratic dialogue the sub-
ject matter that is organized into the “plot” consisits of arguments rather
than events: it is the arguments that constitute the dialogue’s “story.” In this
chapter I will argue that the story is emplotted not only by means of the narra-
tor’s text, that is, the descriptions of the setting and of the characters’ appearance,
movements, and feelings, but also by such means as the change of interlocutor
and metanarrative comments voiced by the characters. To all intents and pur-
poses, these are Plato’s plot-building tools. The systematic use of the tools in
question allows him not only effectively to emplot his dialogues but also to
privilege some arguments over others and thereby to guide the reader toward
certain conclusions regarding the messages he intends to deliver. Let us exam-
ine them one by one.

2 The Narrator’s Text

Narrative voice is a dimension of the text deduced from certain characteristics

the text displays. These are communicated to the reader through the narrator’s
text embodied in what we have referred to as markers of focalization, namely,
segments of text containing descriptions of the setting and of the characters’
appearance, movements, and emotional responses presented by the explicit or
implicit narrator.3

2.1 The Setting

The primary function of the narrator’s text is to introduce the dialogue’s setting
and to draw the reader’s attention to it as the dialogue proceeds (the change
of setting in the dialogues’ openings is discussed in the next section). The set-
ting is normally introduced in the opening part of the dialogue. The only dia-
logues that do not follow this pattern are the Meno, where the implicit narrator
draws the reader’s attention to the setting only in the middle of the dialogue
(pp. 59–60), the Philebus, in which the presence of the audience is gradually

2 On emplotment see above, Chapter 1, with n. 70.

3 Cf. Bal 1997: 36: “Description is a privileged site of focalization, and as such it has great impact
on the ideological and aesthetic effect of the text.”
126 Chapter 6

­disclosed (p. 69), and the Cratylus, with its “peculiarly thin frame,”4 whose only
hint at the setting is given at the end of the dialogue (p. 59).
The role of the setting is never passive. Rather often than not its re-emer-
gence in the body of the dialogue helps to punctuate the conversation, to mark
the change of subject, to emphasize the important points. This may be done,
for example, by a reference to the reaction of the audience, whose presence in
the background is thereby revealed: we find this in the Gorgias, the Euthyde-
mus, the Protagoras, the Symposium, and the Philebus. In such cases the audi-
ence, to borrow Plato’s own metaphor, functions like the chorus in tragedy:
“At these words the followers of Euthydemus and Dionysodorus, of whom I
spoke, like a chorus at the bidding of their director, laughed and cheered.”5 Of
course not every “chorus” is as vociferous as that of the Euthydemus, but even
the appeals to a silent audience, for example in the Philebus, can be used for
the purpose of structuring.6 The same effect can be achieved by going back
to the dialogue’s locale. This may be a palaestra (the Lysis, the Charmides, the
Theaetetus), the Lyceum (the Euthydemus), a private house (the Gorgias, the
Protagoras, the Symposium, the Republic, the Parmenides), the Athenian State
prison (the Crito, the Phaedo), or a shady spot on the banks of the Ilissus or in
the vicinity of Cnosus (the Phaedrus, the Laws). The Phaedrus, which is espe-
cially remarkable in this respect, deserves special attention.
The narrative of the Phaedrus is punctuated by two major breaks, each tak-
ing us back to the pastoral setting that frames the dialogue and creating an op-
portunity to close the conversation. Socrates, who has just completed a speech
dedicated to rational love, decides to leave the place and to make his way back
to the city. Phaedrus tries in vain to persuade him to wait until the midday heat
is over. When Socrates is about to cross the Ilissus, the daemonium commands
him to amend the offence against the God of Love he committed in his speech.
This leads to the Palinode, Socrates’ second speech (242a–243e). When that
speech is over, another option for ending the dialogue arises. Still waiting for
the end of the heat, Socrates and Phaedrus converse about Lysias; their conver-
sation is accompanied by the singing of cicadas. To avoid surrendering to the
vulgar pastime of an afternoon nap, Socrates proposes to spend the rest of the
afternoon in conversation, supporting his suggestion with a story of the origin
of the cicadas (257c–259d). This is where the discussion about speaking and
writing begins; in it, Socrates comments on the three speeches delivered in the
first part of the dialogue, introduces the method of collection and division, and

4 Friedländer 1964: 196.

5 Euthyd. 276b6–c1. On the narrator’s descriptions of the audience reaction see also above,
pp. 67, 90–91.
6 Phlb. 12b, 15bc, 16a, 19c, 20a, 23b, 28d, 29b, 67b, and above, p. 69 n. 57.
The Limits of Authority 127

questions the benefits of writing. By the time the discussion ends, the heat is
over, and Socrates and Phaedrus leave.
The repeated introduction of the dialogue’s setting effectively divides the
Phaedrus into three sections. First, it separates the two speeches dedicated to
rational love, those of Lysias and Socrates, from the Palinode, the speech of
Socrates dedicated to irrational love and introducing the myth of the soul. Sec-
ondly, it separates the Palinode from the concluding discussion about speaking
and writing. As a result, the Palinode is enclosed by two major breaks, an ar-
rangement which foregrounds its position as the climax of the dialogue.7 The
Phaedrus is thus fully structured through its setting.8 A similar organization of
the plot by systematic re-introduction of the setting is also characteristic of the
Euthydemus, the Protagoras, the Symposium, and the Phaedo.

2.2 The Characters

The characters in Plato’s dialogues are not passive either. They come and go,
hurrying to an appointment (Euthyphro in the Euthyphro, Socrates in the
Meno, the Protagoras, the Theaetetus, the Philebus); they are about to leave
but in the end are persuaded to stay (Socrates in the Gorgias, the Protagoras,
the Phaedrus, Thrasymachus in Republic 1); they may become angry (Critias
in the Charmides, Hippias in the Hippias Maior, Protagoras in the Protagoras,
Callicles in the Gorgias, Anytus in the Meno, Thrasymachus in Republic 1) or
unhappy with their own arguments (Socrates in the Lysis, the Theaetetus, the
Phaedrus, the Eleatic Stranger in the Sophist and the Statesman); they blush
(e.g. Lysis and Charmides in the eponymous dialogues, Dionysodorus in the
Euthydemus, Thrasymachus in Republic 1) and laugh (passim).
Furthermore, the interlocutors do not just reply to questions addressed to
them by the leading speaker: they are also able to challenge the protagonist, to
interrupt his argument or even to redirect its course. Thus, the second part of
the Meno opens with Socrates’ proposal to enquire into the nature of virtue in
order to define it; however, Meno insists on returning to his original question

7 A similar strategy of foregrounding is employed in the Phaedo; see above, p. 91 and below,
p. 145. On the transition effected by the cicada story see Ferrari 1987: 25–34; Schenker 2006:
76. The widespread contention that the cicada episode constitutes the only major break of
the Phaedrus follows a preconceived view of the dialogue’s structure, ignoring a no less em-
phatic break that precedes the Palinode.
8 Cf. Ferrari 1987: 3–4: “what is particularly striking about this dialogue is that the background
will not stay where it belongs. It becomes a prominent topic of discussion and a direct cause
of the conversational action rather than, as one would expect, at most an indirect influence
on its course.” Cf. also Murley 1955: 281: “the recurrent themes of the cicadas and the stream
of the outdoor setting keep the tone constant.” On the Phaedrus’ use of dramatic interludes
see also Werner 2007: 116–17.
128 Chapter 6

as to whether virtue can be taught, thereby skipping the definition. The discus-
sion veers in the direction chosen by Meno, with momentous consequences
for the dialogue’s argument.9 Such navigation of the plot in a direction not
envisaged by the leading speaker is also characteristic of the Phaedo, where
the objections of Simmias and Cebes cause Socrates to expound the Theory of
Forms (below, pp. 144–46).
The interlocutors’ influence on the plot is especially prominent in the Re-
public. The objections of Glaucon and Adeimantus at the beginning of Book 2
re-open the discussion of justice held in Book 1;10 Glaucon’s intervention and
his complaint about the “city of pigs” later in the same book (372c–373b) mark,
as G.R.F. Ferrari pointed out, “a major turning-point in the work”;11 finally, the
interference of Polemarchus and others at the beginning of Book 5 (quoted
above, p. 36) results in a digression that spreads across Books 5–7 and proves to
be the central part of the dialogue.12
The same or closely similar narrative devices are employed over and over.
This is how, for example, Socrates’ abortive attempt to end the conversation
and to return to Athens is described in the Phaedrus:

When I was about to cross the stream, my friend, the divine sign which
usually appears to me appeared again … and I thought that I heard a
voice saying in my ear that I must not go away until I had made an atone-
ment. (242b8–c3)

This is essentially the same strategy as that used in the Protagoras, in the de-
scription of Socrates’ abortive attempt to leave Callias’ house after he and Pro-
tagoras have failed to agree on how their discussion should proceed: “Thus I
spoke and was rising from my seat, when Callias seized me by the right hand

9 Meno 86ce. Cf. Guthrie 1975: 264: “The message is clear. The present conclusion is not the
correct one, because Meno has made Socrates ask the wrong question” (original empha-
sis). Cf. also Blössner 2011: 65.
10 Cf. Schofield 2008: 63–64: “When in a sequence at the beginning of Book 2 of the dialogue
running to ten Stephanus pages Glaucon restates what he takes to be an improved version
of Thrasymachus’ views … with Adeimantus bringing further arguments in support, the
response Plato attributes to Socrates occupies the whole of the rest of the work (around
250 Stephanus pages) …” Cf. also Ferrari 2010: 12–14; Schur 2014: 88–92.
11 Ferrari 2010: 12: “[t]he need for Guardians in the ideal city can be traced back to this inter-
vention”; see also Ferrari 2010: 14, on the intervention of Glaucon at Resp. 471c–473b.
12 Cf. Clay 1988: 22: “It is this interruption, unexpected by Socrates, but planned by Plato,
that introduces the more difficult themes of the Republic.” See also Ferrari 2010: 23.
The Limits of Authority 129

…” (335c7–8, quoted in full above, p. 83). The comparison demonstrates that

in both the Protagoras and the Phaedrus Socrates’ movements are described
not for their own sake but to signalize a crisis and to mark the transition to the
next stage of the discussion.
It follows, then, that the descriptions of the setting and of the characters’
emotions and movements should not be taken, as they often are, merely as a
tribute to the dialogue’s mimetic form – an “embellishment,” as it were.13 They
are used to introduce new arguments, to signal the transitions from one stage
to another, to draw the reader’s attention to the dialogue’s turning points, and
to emphasize its climax. They are as important in structuring the dialogue as
the pronouncements relating to the argument proper.
The narrator’s text is the most widespread plot-building tool employed by
Plato. Explicitly or implicitly, it is present in all dialogues, and its role in laying
out the dialogues’ framework is indispensable. Nevertheless, the narrator’s text
is not the only means by which Plato’s dialogues are structured.

3 Change of Interlocutor

The material presence of the setting and the interlocutors does not depend on
the narrator and is not influenced by his perception. These are autonomous
elements whose alteration can be used as a plot-building tool with or without
the narrator’s text. The plot-building role of the change of setting, however,
is limited to the dialogue’s openings, which normally consist of two stages
divided by a clearly articulated spatial change: Socrates and Hippocrates meet
at Socrates’ house and go to the house of Callias (the Protagoras); Socrates
and Chaerephon meet Callicles in the street and enter Callicles’ house (the
Gorgias); Socrates and Glaucon meet Polemarchus and his company in the
street and go to Polemarchus’ house (the Republic); Socrates and Phaedrus
meet in the city and go outside the city walls (the Phaedrus); and so on. One
of Plato’s favorite patterns, it was meant to set the rhythm of the narrative
and to direct the reader to the subject of the discussion (above, pp. 27–30,
47–49, 76–79).
The change of interlocutor is much more widely applied. In the dia-
logues where the protagonist has more than one interlocutor, the change of

13 For criticism of the assumption “that the literary aspects of Platonic texts are decorative
and separable from philosophical content,” see Michelini 2003, 3–4, 10. See also Laborde-
rie 1978, 60–64; Johnson 1998, 577; Schur 2014: 3–40.
130 Chapter 6

­interlocutor is the simplest and the most effective way to organize the argu-
ment and to punctuate its progress – or lack of it. It is regularly accompanied
by a change of subject or by a transition to the next stage of the discussion. As
W.K.C. Guthrie remarked apropos of the Charmides, the change of interlocutor
is never mechanical.14
The interlocutors may succeed one another, like Charmides and Critias in
the Charmides, Hermogenes and Cratylus in the Cratylus, Gorgias, Polus and
Callicles in the Gorgias, or Cephalus, Polemarchus and Thrasymachus in Re-
public 1, but they may also take turns, like Lysis and Menexenus in the Lysis,
Simmias and Cebes in the Phaedo, or Glaucon and Adeimantus in Republic
2–10. In the dialogues with multiple interlocutors, other participants may tem-
porarily replace both the protagonist and the main interlocutor. Thus, at the
beginning of the Gorgias Chaerephon and Polus temporarily replace Socrates
and Gorgias (above, pp. 66–67); in the Euthydemus, Ctesippus or Cleinias re-
place Socrates in some of the exchanges with Euthydemus and Dionysodorus,
a recurrent feature that well suits the “ball-playing” atmosphere of this dia-
logue.15 A similar temporary replacement of Socrates the protagonist is also
attested in the pseudo-Platonic Eryxias (395e–396a, 396e–397c, Critias and
Eryxias).
Sometimes a cameo appearance of a character who hitherto has taken no
part in the conversation is used for emphasis. In the Phaedo the introduction
of the Theory of Forms is emphasized, among other things, by the fact that
Phaedo the narrator, who as yet has not spoken as a character, temporarily re-
places Simmias and Cebes as Socrates’ principal interlocutor;16 in the Theaete-
tus Theodorus replaces Theaetetus when the famous Digression – presumably
the climactic episode of the dialogue – is being delivered.17 On other occa-
sions an extra speaker emerges, whose main role is to intensify the position of
one of the interlocutors or to express an opinion that would be out of charac-
ter for the principal interlocutor: this is the role assumed by Ctesippus in the

14 Guthrie 1975: 168.

15 Euthyd. 277b3–5 “Dionysodorus took up the argument, like a ball which he caught, and
had another throw at the youth”; see further Euthyd. 255c–277d, 286b–288a, 298b–300d.
16 Phd. 89b–91c and below, Section 5. Cf. Sedley 1995: 8: “He [Phaedo] is not only the narrator
but also an interlocutor at one crucial point.”
17 Tht. 172c–177c. Cf. Blondell 2002: 303: “Here Sokrates provides an inspiring synoptic vision
of the place of human beings in an intellectual and moral cosmos … In this respect, the
digression is comparable to an image that lies at the heart – if not exactly in the center –
of the Iliad: the shield of Achilles in Book 18.” Cf. also Rutherford 1995: 283. On the Digres-
sion and the history of its interpretation see esp. Sedley 2004: 62–81; Bartels 2015 (with
bibliography).
The Limits of Authority 131

E­ uthydemus (above, p. 87), by Anytus in the Meno,18 by Cleitophon in Republic

1,19 and by Philebus in the Philebus.20
The change of interlocutor may also be staged as a violent interruption of
the ongoing discussion by a bystander: this is the case of Critias in the Char-
mides, of Callicles in the Gorgias, of Thrasymachus in Republic 1; characteristi-
cally, in such cases the change of interlocutor is usually accompanied by the
narrator’s description of the interlocutors’ feelings and actions as perceived by
him (see Section 5, on clustering). Here too, the narrative strategies follow a
standard pattern. Compare, for example, the following passages:

Critias has long been showing uneasiness, for he felt that he had a repu-
tation to maintain with Charmides and the rest of the company. He had,
however, hitherto managed to restrain himself, but now he could no lon-
ger forbear. (Chrm. 162c1–4).
Several times in the course of the discussion Thrasymachus had made
an attempt to get the argument into his own hands, and had been put
down by the rest of the company, who wanted to hear the end. But when
Polemarchus and I had done speaking and there was a pause, he could no
longer hold his peace. (Resp. 336b1–5)

Finally, new characters may suddenly arrive and interrupt the conversation or
channel it in a different direction: the tutors of Lysis and Menexenus in the
Lysis, Alcibiades in the Symposium.
In the multi-level dialogues, the transition from one narrative level to an-
other invariably takes the form of a change of speaker. As we have seen, insofar
as it is presented in the text as interplay between the dramatic frame and the
framed narrative, in the dialogues of this group the change of speaker creates
a clearly pronounced boundary between the frame and the framed (above,
pp. 101–102). In the Euthydemus and the Phaedo, characters belonging to the
dramatic frame intrude upon the main story, collapsing the barrier between

18 Athenian society, the principal topic of Socrates’ conversation with Anytus, is not suitable
for discussion with the foreigner Meno.
19 Cleitophon modifies Thrasymachus’ position, in that he tries to help him by eliminating
a potential ambiguity in Thrasymachus’ thesis that justice is the interest of the stronger
(Resp. 340ab). On Cleitophon’s intervention see Cross and Woozley 1964: 43–46; Annas
1981: 41–42.
20 The presence and grudging rematks of Philebus make it possible to mark dogmatic hedo-
nism as the extreme pole of the discussion yet to ignore the uncompromising position it
represents. On Philebus’ role in the dialogue see Davidson 2013 [1990]: 10; D. Frede 1996:
218–20; Meinwald 2008: 485.
132 Chapter 6

different narrative levels and producing ontological metalepsis;21 the change

of interlocutor by whose means metalepsis is signalized acts in these dialogues
as a powerful plot-building tool.22
Change of interlocutor is not as widely applied as the narrator’s text. It is
irrelevant as regards the dialogues with a single interlocutor (the Euthyphro,
the Crito, the Ion, the Hippias Maior, the Menexenus, the Phaedrus), and it
is significantly reduced in dialogues whose main arguments are delivered in
the course of a question-and-answer session (the Parmenides, the Sophist, the
Statesman and, to a degree, the Theaetetus and the Philebus) or as a monologue
(the Timaeus).

4 Metanarrative Comments

On numerous occasions the leading interlocutor – either Socrates or, as in

the later dialogues, another character (Parmenides in the Parmenides, the Ele-
atic Stranger in the Sophist and the Statesman, Timaeus and Critias in the
eponymous dialogues, the Athenian Stranger in the Laws) – comments on
the discussion, explicitly thematizing its stages and expressing his attitude
to the form it has assumed. Taken together, the characters’ comments on the
discussion in which they take part constitute a corpus of evidence whose sig-
nificance cannot be overestimated. The metanarrative character of such com-
ments is self-evident, and the fact that Plato found it necessary to introduce
them over and over indicates that he invested considerable effort in trying to
help his readers navigate the dialogues.
Although often issued by the narrator’s former self as a character in the sto-
ry (the narrator in such cases is invariably Socrates), the metanarrative com-
ments should not be confounded with the narrator’s text. That such comments
are also voiced by characters other than the narrator, namely, Parmenides, the
Eleatic Stranger, or Socrates in the dialogues where he does not act as the nar-
rator, shows unequivocally that they belong with character speech rather than
with the narrator’s text.
Everything about the ongoing discussion – the beginning and the end, fresh
starts, digressions, the argument’s length, the format it should assume – is

21 Metalepsis that affects the entire story level, see above, p. 23. On the distinction between
rhetorical and ontological metalepsis see Ryan 2006: 204–11; cf. Pier 2013: 2.1 [7]; Finkel-
berg, forthcoming.
22 On the Euthydemus see above, pp. 86–87; on the Phaedo see above, pp. 91–92, and below,
Section 5.
The Limits of Authority 133

open to this kind of metanarrative commentary.23 Like the narrator’s text and
the change of interlocutor, these comments are as a rule placed at strategic
junctions, separating arguments or punctuating the argument’s progress and
highlighting the turning points. Like the narrator’s text again, they often follow
standard patterns that recur in several dialogues. For example, the following
remarks signal the beginning of an argument in the Parmenides, the Sophist,
and the Philebus:

Where shall I begin? And what shall be our first hypothesis if I am to play
this laborous game? Shall I begin with myself, and with my own
hypothesis …?24
And where shall I begin the perilous argument? I think that the road
which I must take is …25
And where shall we begin this great and multifarious battle, in which
such various points are at issue? Shall we begin thus?26

Metanarrative comments are relatively infrequent in the so-called early and

middle dialogues. It is true, of course, that the Cratylus, in which the narrator’s
text is minimal and the change of interlocutor occurs only once, is entirely
structured on the metanarrative comments voiced by Socrates the character.27
But the Cratylus is the exception rather than the rule, for even the Meno, a dia-
logue very close to it in manner of presentation, issues a considerable portion
of the narrator’s text (above, pp. 59–60). In the later dialogues, on the other

23 The beginning and the end Grg. 505a–506a; Phd. 77c; Resp. 398b, 497e; Prm. 137ab; Soph.
242; Plt. 265b, 311b; Phlb.66c; Laws 824a, 960b; fresh starts Euthyphr. 15c; Prt. 318a; Euthyd.
275c; Meno 79c, e; Phd. 88d, 105b; Cra. 436e; Prm. 142b; Tht. 151d, 164c, 187a, 200a, d; Plt.
264b; Phlb. 34e; Laws 683e, 867c; digressions Grg. 497c; Euthyd. 288d; Cra. 437e; Phd. 78ab;
Resp. 503a, 543c, 572b; Prm. 136e; Tht. 177b; Soph. 241c; Plt. 263a, c; Phlb. 18a; Laws 682e,
812a, 864c; length and format Prt. 334c–336b; Grg. 449bc, 451d, 465e; Resp. 348a, 354a,
376d, 435cd, 484a, 530e, 534a, 548c; Tht. 184a, 187de; Phdr. 271e–272b; Soph. 217c; Plt. 265ab,
277a, 286c–287a; Phlb. 28cd, 36d; Laws 583c, 887b.
24 Prm. 137a7–b2, spoken by Parmenides.
25 Soph. 242b6–8, spoken by the Eleatic Stranger. “The perilous logos” is the Eleatic Strang-
er’s refutation of the thesis of Parmenides, the speaker’s spiritual father (241d, 242a), that
not-being does not exist. On the Sophist see also below, Chapter 7.
26 Phlb. 15d1–2, spoken by Socrates.
27 See e.g. Cra. 397a4–5 “And where would you have us begin, now that we have got a sort of
outline of the enquiry?”; 427c9–d2 “This is my view, Hermogenes, of the truth of names;
but I should like to hear what Cratylus has more to say”; 436e2 “And here let us revert to
our former discussion”; 437e8–438a1 “But let us have done with this question and return
to the point from which we digressed,” and so on.
134 Chapter 6

hand, metanarrative comments multiply so significantly that they practically

replace the narrator’s text, taking on some of its functions.
Republic 2–10, the Statesman, and the Laws are especially rich in such com-
ments. Their abundance should largely be attributed to the dialogues’ length,
which is also subject to comment in the dialogues themselves. For example,
in the Statesman, at the point of transition from the model of weaving he has
just introduced to the application of this model to the statesman, the Eleatic
Stranger embarks on an explanation of the prolixity of both the previous dis-
cussion and the one to follow:

I mainly wanted to avoid the discomfort caused by verbosity (tēn makro-

logian), of which we complained in the discussion about weaving, as well
as in the one about the reversal of the universe and about the Sophist and
the existence of not-being. We noticed that they were too long, and we
blamed ourselves for this, fearing that they might be not only lengthy but
also irrelevant (perierga).28

In Benjamin Jowett’s uncharitable reading of the dialogue, “Plato apologizes

for his tediousness, and acknowledges that the improvement of his audience
has been his only aim in some of his digressions.”29 Yet the same issue is also
addressed in similar terms in the Republic, the Theaetetus, the Phaedrus, the
Philebus, and the Laws.30 The issue at stake here is how far a given text can be
considered relevant to the dialogue’s argument (cf. pros to prepon at Plt. 286d1–
2). As the Eleatic Stranger puts it later on in the same dialogue, adjusting the
dialogue’s length to its pleasurable effect should be considered a secondary
matter (parergon) in a philosophical discussion.31
In the Philebus, at the point of transition to a lengthy discussion of the possi-
bility of false pleasures (36c–42c), Socrates modifies the principle of relevance,
taking it further: “the point to be considered is whether the enquiry is relevant
(prosēkonta) … No overlong and irrelevant (para to prosēkon) discussion can
be allowed” (Phlb. 36d6–10). The issue of disproportionate length that troubled
Plato in the Statesman has thus been successfully resolved; yet, this is done not
to create a pleasurable effect but again, for the sake of clarity and efficiency of

28 Plt. 286b7–c1; my translation. Cf. also 277a6–b6: “like sculptors who, in their too great
haste, having overdone each part of their work, lose time, so too we … have taken up a
marvelous lump of myth, and have been obliged to use more than was necessary (μείζονι
τοῦ δέοντος).”
29 Jowett 1892: 2.
30 Resp. 376d, 435cd, 484a, 530e; Tht. 184a, 187de; Phdr. 272b; Phlb. 36d; Laws 887b.
31 Plt. 286d4–5 οὔτε γὰρ πρὸς τὴν ἡδονὴν μήκους ἁρμόττοντος οὐδὲν προσδεησόμεθα, πλὴν εἰ
πάρεργόν τι. Cf. also above, p. 119.
The Limits of Authority 135

the argument. This is the poetics of the late dialogues in the making – the same
poetics of austerity, it should be added, that was first displayed in Republic 3
(above, p. 119). It is pursued, among other things, by replacing the narrator’s
text with metanarrative comments strategically placed within the dialogues to
which they refer.

5 Distribution and Clustering: Three Case Studies

The plot-building tools discussed above can be furher reinforced by being

gathered into clusters. To obtain a fuller picture of how the narrator’s text, the
change of interlocutor and metanarrative comments interact in the context of
a single dialogue, let us dwell at some length on three dialogues – the narrated
Lysis, the dramatic Gorgias, and the mixed Phaedo. (The beginnings of the
plot-building passages are marked with asterisks.)

5.1 The Lysis

The standard two-stage opening involves a change of scene: Socrates encoun-
ters Ctesippus and Hippothales who invite him to visit a newly founded pa-
laestra (above, pp. 27–28). After the dialogue’s setting has been established,
Socrates meets his principal interlocutors, the boys Lysis and Menexenus. But
as the conversation starts to evolve, Menexenus leaves the stage. “At this mo-
ment Menexenus was called away by someone who came and said that the
gymnastic-master wanted him. … So he went away, and I asked Lysis some
more questions” (207d1–5). Menexenus’ temporary absence emphasizes the
central position of Lysis in the dialogue. The conversation begins.
(1) Socrates – Lysis (207d–210d). The conversation starts in the manner char-
acteristic of the elenctic dialogues. Socrates questions Lysis about what he
understands by friendship and demonstrates that Lysis’ idea of friendship is
inadequate.
*At this point, Menexenus comes back. His return is staged as a lively
interlude:

In the meantime Menexenus came back and sat down in his place by Ly-
sis; and Lysis, in a childish and affectionate manner, whispered privately
in my ear, so that Menexenus should not hear: “Do, Socrates, tell Menex-
enus what you have been telling me.” (211a1–5)

Wishing his friend to be let down as he himself was, Lysis bids Socrates ques-
tion Menexenus. Ctesippus complains about their talking in secret. Socrates
turns to Menexenus, and the discussion begins in earnest.
136 Chapter 6

(2) Socrates – Menexenus (212a–213d). The starting point of the discussion is

the question as to whether one who is friendly disposed to another or one to
whom another is friendly disposed should properly be called “friend” (philos).
Socrates finds it impossible to decide the issue.
*The ensuing change of speaker is accompanied by a lavishly elaborated
piece of the narrator’s text:

“But, o Menexenus! I said, may we not have been altogether wrong in our
conclusions?” “I am sure that we have been wrong, Socrates,” said Lysis.
And he blushed as he spoke, the words seeming to come from his lips
involuntarily, because his whole mind was taken up with the argument;
there was no mistaking his attentive look while he was listening. I was
pleased at the interest which was shown by Lysis, and I wanted to give
Menexenus a rest, so that I turned to Lysis and said … (213d1–e1)

(3) Socrates – Lysis (213e–215c). Socrates inquires of Lysis as to the respective

roles of the participants in the relationship. The first possibility, namely, that
like is friendly to like, is scrutinized and rejected on the grounds that since the
good man, insofar as he is good, will be sufficient unto himself, and since be-
ing self-sufficient he needs nothing, he will love nothing and thus will not be a
friend to anybody.
*Menexenus interposes a remark, and subsequently takes over from Lysis.
(4) Socrates – Menexenus (216a–218c). Now, the theory that it is rather op-
posites that are attracted to each other is examined. In this case, however, it
will be necessary to postulate absurd things, for example, that friendship is
a friend of enmity, and so on. This theory does not withstand scrutiny and is
rejected like the previous one. Now Socrates proposes distinguishing between
three rather than two classes: – good, evil, and that which is neither good nor
evil. Since it has already been proved that good cannot be a friend of good, or
evil of evil, it is agreed that friendship can only exist between good and what
is neither good nor evil.
*At this point, a crisis occurs:

They both agreed and entirely assented, and for a moment I rejoiced like
a huntsman just holding fast his prey. But then a most unaccountable
suspicion came across me, and I felt that what we had agreed upon was
untrue. I was pained, and said: “Alas! Lysis and Menexenus, I am afraid
that we have been grasping at a dream (onar) only.” (218c2–8)

Socrates expresses the suspicion that just like human impostors, they may have
fallen victim to impostor arguments (218d2–4): note that the narrator’s text of
The Limits of Authority 137

Socrates is combined here with the words of Socrates as one of the characters
in the story. The narrator’s effusive comments on the emotional state of his for-
mer self emphasize the transition to what can justly be seen as the dialogue’s
central argument.32
(5) Socrates – Lysis and Menexenus (218d–221a). The rest of the discussion
is simultaneously held with both boys. The agreed solution fails because it
requires postulating that good “is loved on account of evil by us who are in-
termediate between evil and good, but in itself, and for itself, it is of no use”
(220d4–7). Socrates revises it to eliminate the evil, and arrives at the conclu-
sion that both friendship and love are nothing but the desire (epithumia) of
one who is deficient in something (221d-e); this conclusion is taken further in
the doctrine of love set forth in the Symposium.33
*The transition to the last installment does not involve a change of inter-
locutor. Instead, it is signaled by a piece of the narrator’s text with a highly
charged erotic subtext. Socrates’ inference that the love of a true lover should
be reciprocated by his beloved delights Lysis’ admirer Hippothales but em-
barasses the boys: “Lysis and Menexenus gave a faint assent to this, but Hip-
pothales changed into all manner of colors with delight” (222b1–2). As Charles
Kahn reminds us in his comment on this passage, an erotic advance is not sup-
posed to be reciprocated: “when courted, a boy is expected to show indiffer-
ence, even annoyance.”34 As his reaction earlier in the same passage indicates,
Lysis seems to have spotted the problem even before the argument reached its
conclusion: “’Yes, yes,’ said Menexenus. But Lysis fell silent (esigēsen)” (222a4).
(6) Socrates – Lysis and Menexenus (222b-d). The interlude also introduces a
new term to be discussed – “what belongs to us (oikeion)” (222a-b). This brings
us to the last section of the dialogue. Proceeding from the assumption that the
desire spoken about is a desire of something that belongs to us, we shall find
ourselves postulating that like is friendly to like, thus returning to the start-
ing point of the discussion – unless, of course, “what belongs to us” and “like”
amount to two different things.
*Socrates is eager to proceed, but the boys’ tutors suddenly arrive, “like some
demons,” and take them home.
Note that only in one case, Menexenus replacing Lysis at the end of (3), is
the change of interlocutor not accompanied by the narrator’s text; and only in
one case, at the end of (5), is the transition from one section to another effected
by the narrator’s text with no change of interlocutor. In the rest, the t­ ransitions

32 Cf. Kahn 1996: 290: “the implicit positive conclusion of the Lysis.” A closely similar reversal
is also introduced, to a similar effect, in the Phaedrus (see above, pp. 126–27).
33 Cf. Kahn 1996: 286–90; Finkelberg 1997: 236–37.
34 Kahn 1996: 264–65.
138 Chapter 6

are signaled by the combination of these two plot-building tools. Finally, the
transition to the central argument at the end of (4) is further emphasized by
the addition of a metanarrative comment about impostor arguments.
The Lysis lucidly demonstrates the range of emplotment strategies available
in the dialogues with alternating interlocutors (Republic 2–10 readily comes
to mind in this connection). The dialogue is short (just over ten Stephanus
pages), and because it comprises a single argument, its structure is relatively
simple. Even so, the clustering and interplay of plot-building tools result in that
the sequence of the argument’s stages is not only cumulative but also incorpo-
rates culmination and reversal, thereby exhibiting the same dramatic pattern
that underlies the much more complex Phaedo and Phaedrus.

5.2 The Gorgias

The standard two-stage opening is only hinted at in this dialogue; neverthe-
less, just as in the narrated dialogues, it involves a change of scene. Socrates
and Chaerephon meet Callicles in the street, and after an opening exchange
(above, p. 66) enter Callicles’ house where Gorgias and Polus already are found.
The conversation proper opens with a preliminary discussion between Chae-
rephon and Polus.35 Three major discussions follow, each involving a different
interlocutor; each contributes a new turn in the argument and significantly
modifies the discussion. As R.B. Rutherford has pointed out, this is a cumula-
tive sequence:

Essentially, the sequence is of cumulative force: each exchange is longer

and more intense, even vehement; each new speaker goes further, draws
Socrates out more and challenges him more profoundly; in each case
Socrates reacts by rising to greater rhetorical heights, by the end adopt-
ing a form of discourse more solemn and intense than is usual for him.36

The Gorgias is the third longest dialogue in the Platonic corpus, and its plot
is much more complex than that of the Lysis. The three major parts are di-
vided into sections, whereas in the Socrates-Callicles exchange, the longest of
the three, the division is even more ramified. The change of interlocutor, the

35 Similarly in the Sophist, where such a preliminary exposition of the dialogue’s concerns is
called paradeigma; see Soph. 218d–221c. That in his translation of the Sophist F.M. Corn-
ford omitted this section altogether is symptomatic of the position according to which
literary aspects of Plato’s dialogues should be treated as mere embellishment (cf. n. 13
above). For criticism of Cornford’s attitude to Plato’s dialogues see Bloom 1991: xiv–xx.
36 Rutherford 1995: 142.
The Limits of Authority 139

­narrator’s text, metanarrative comments – all tools at Plato’s disposal are em-
ployed to emplot this intricate and dramatically charged dialogue.
(1) Socrates – Gorgias
(1.1) The discussion between Socrates and Gorgias falls into two parts. The
first, a search for the definition of rhetoric (449a–457c), ends in a crisis, and the
interlocutors are ready to end the conversation.
*Chaerephon’s appeal to the audience and the support he receives from Cal-
licles (457c–458e, above, p. 67) resolve the crisis.
(1.2) The discussion resumes, now taking up the issue of the morality of
rhetoric (458e–461a).
*The end of the verbal duel between Socrates and Gorgias is accompanied
by a metanarrative exchange concerning the format of both the preceding and
the forthcoming discussion. Socrates insists that, as with Gorgias, the conver-
sation with Polus should proceed through questions and answers, whereas Po-
lus prefers delivering a speech (461e–462a). This is only one of many examples
of the opposition between makrologia and brachulogia, used to the same effect
elsewhere in the Gorgias (below) and in other dialogues, first and foremost the
Protagoras.37
(2) Socrates – Polus
(2.1) Polus replaces Gorgias, and the discussion reverts to the definition of
rhetoric (461b–463e). This is Gorgias’ theme, and he intervenes, requesting
clarification. Socrates responds with a speech culminating in the conclusion
that rhetoric is an art of flattery (464b–465d).
*Socrates’ metanarrative comments on his own makrologia (465e–466a)
signal the re-emergence of Polus and the change of subject.
(2.2) First to be discussed is Socrates’ thesis that the unjust are miserable
(468e–473e).
*The apparent absurdity of Socrates’ conclusion makes Polus laugh, and like
Chaerephon before him he appeals to the audience. Socrates reacts with a spir-
ited diatribe against “counting the votes,” adducing his recent service on the
Athenian Council as an example (474a5–b1).
(2.3) The discussion resumes, leading to the paradoxical conclusion that to
do wrong is worse than to suffer wrong (474c–481b).
*Callicles is appalled. He turns to Chaerephon for clarification, but Chaere-
phon suggests he ask Socrates instead (above, p. 68): this is how the Socrates-
Callicles exchange, the longest of the three, begins. It is focused on two speech-
es, Callicles’ “speech of Zethus” and Socrates’ “speech of Amphion,” which

37 Grg. 449bc, 465e; Prt. 320c, 329b, 334c-e, 335b, 336bc. See also Hp. mi. 373a; Symp. 194d;
Resp. 348ab; Phdr. 271e; Soph. 217c.
140 Chapter 6

evoke the alternative ideals of the active and the contemplative life as exhibit-
ed in the agōn between Zethus and Amphion in the lost Antiope of Euripides.38
(3) Socrates – Callicles
(3.1) “The Speech of Zethus”39
(3.1.1) Callicles’ speech, which opens the discussion (482c–486d), attacks the
conventional approach to justice and endorses its opposite, “the justice of na-
ture” (484b1).
*In commenting on the speech, Socrates favorably compares Callicles’ out-
spokenness to the elusiveness of Gorgias and Polus. He finds in Callicles a suit-
able interlocutor for “the most beautiful enquiry,” namely, “What ought a man
to be, and what his pursuits, and how far is he to go – both in maturer years and
in youth?” (487e9–488a2).
(3.1.2) When the subsequent discussion reaches the conclusion that plea-
sure is not the same as the good, Callicles loses his ground and ceases to coop-
erate (489a–497a).
*Gorgias interferes, and urges Callicles to proceed: “Nay, Callicles, answer, if
only for our sakes: we should like to hear the arguments out.”40
(3.1.3) The situation recurs when Socrates leads his interlocutor to the conclu-
sion that restraint is better for the soul than intemperance. “I do not underst­and
you, Socrates,” Callicles says, “and I wish you ask someone who does …” (505c1–2).
*A series of metanarrative comments follows, in which Plato goes to great
lengths to emphasize that at this point we have only arrived at the middle of
the discussion.41 Socrates sees no other way out than completing the argument
on his own or leaving.42 Gorgias interferes again to save the discussion from

38 Grg. 485e–486c, 506b. Cf. Taylor 1960 [1926]: 122: “Socrates and Callicles stand respectively
for two antithetical ideals in life, the one for the ‘life of philosophy,’ the other for the ‘life
of action’ as followed by a man of affairs in the Athenian democracy.” See also Rutherford
1995: 166–68; Clay 2000: 143–45. For a detailed discussion see Nightingale 1995: 69–79.
39 Grg. 485e3–7 “Now I, Socrates, am very well inclined toward you, and my feeling may be
compared with that of Zethus toward Amphion, in the play of Euripides … for I am dis-
posed to say to you what Zethus said to his brother, that you, Socrates, are careless about
the things of which you ought to be careful …”
40 Grg. 497b4–5 ἀλλ’ ἀποκρίνου καὶ ἡμῶν ἕνεκα, ἵνα περανθῶσιν οἱ λόγοι.
41 Grg. 505c7–8, Socrates to Callicles: “Well, what shall we do then? Break off our logos in
the middle? (μεταξὺ τὸν λόγον καταλύομεν;)”; 505c10–d2 “Well, they say that it is not right
to leave even tales in the middle, but we should fit a head on them, that they may not go
about headless. Give us the rest of the answers, then, that our logos may acquire a head
(ἵνα ἡμῖν ὁ λόγος κεφαλὴν λάβῃ)”; 505d6–7 “You know, we must not leave the logos incom-
plete (μὴ γάρ τοι ἀτελῆ γε τὸν λόγον καταλίπωμεν).”
42 Grg. 506a6–7 “But I say this only if you think the logos should be carried through to the
end (εἰ δοκεῖ χρῆναι διαπερανθῆναι τὸν λόγον); but if you think otherwise let us leave off and
go our ways.”
The Limits of Authority 141

ending prematurely: “I do not think, Socrates, that we ought yet to depart, but
you should carry through the logos, and I think the others too agree with me.
I myself am anxious to hear you go through what remains” (506a8–b3).43 E.R.
Dodds’ comment on this episode deserves to be quoted in full:

The interlude emphasizes the importance of the positive doctrine which

Socrates is now about to establish, as Gorgias’ previous intervention, at
497b4, emphasized the importance of the refutation of hedonism. It also
enables Plato to get rid of Callicles for a while, and thus (a) to avoid mak-
ing him to assent to propositions which we know to be against his deep-
est convictions (cf. on 513c5); (b) to let Socrates to present a continuous
exposition of his point of view (the question-and-answer form is dropped
from 507c onwards).44

The interlude signals the transition to a solo performance by Socrates, his

“speech of Amphion,” set as a counterpoise to Callicles’ “speech of Zethus”
(507a–509c).
(3.2) “The Speech of Amphion”45
(3.2.1) Although still signaling his eagerness to end the conversation,46 Cal-
licles is gradually drawn back into it. A discussion of the “true political art,” of
which Socrates claims to be the only practitioner, follows (510a–522e).
*The discussion concludes by foreshadowing Socrates’ trial and death
(521c-d), thus preparing the ground for the myth of afterlife. The latter is ex-
plicitly set as the final installment of the dialogue: “Since you have finished all
else, you may finish this too,” Callicles says to Socrates,47 and the latter begins
his story.
(3.2.2) The myth of afterlife (523a–527e).
As the summary shows, while the change of interlocutor neatly divides the
dialogue into three, this overall division is further emphasized by the use of
additional plot-building tools. Thus, the transition from the first to the second

43 Tr. W.D. Woodhead. Cf. Prt. 336e3–4, spoken by Critias: “…let us rather unite in entreat-
ing both of them not to break up the discussion in the middle (μὴ μεταξὺ διαλῦσαι τὴν
συνουσίαν).” Cf. also Prt. 335d, 347c, 348a.
44 Dodds 1959: 331.
45 Grg. 506b4–6 “I myself too, Gorgias, would have liked to continue the argument with Cal-
licles here, until I had paid him back with the speech of Amphion in reply to that of
Zethus”; cf. also 485e–486c.
46 See esp. Grg. 510a1–2 “Granted, Socrates, so that you will only have done with the discus-
sion (ἵνα διαπεράνῃς τὸν λόγον).”
47 Grg. 522e 6–7 Ἀλλ’ ἐπείπερ γε καὶ τἆλλα ἐπέρανας, καὶ τοῦτο πέρανον.
142 Chapter 6

part is indicated not only by Polus taking over from Gorgias but also by meta-
narrative comments on the discussion’s format; in the transition to the third
part, Polus’ replacement by Callicles is accompanied by the narrator’s text
(Chaerephon mediates between Callicles and Socrates).
Furthermore, the three major parts are further divided into sections and
sub-sections by the same means. The exchanges between Socrates and Gorgias
and Socrates and Polus are punctuated by the narrator’s text (Chaerephon and
Polus appeal to the audience) and metanarrative comments (Socrates com-
ments on the length of his own speech); the exchange between Socrates and
Callicles is marked by the narrator’s text (Callicles’ withdrawal, Gorgias’ inter-
vention and appeal to the audience) and by metanarrative comments signal-
ing the possibility of a premature conclusion of the discussion. Together, all
these form a dynamic layout that wholly emplots the array of arguments dis-
played in the dialogue.

5.3 The Phaedo

The Phaedo is a mixed dialogue whose dramatic frame, Phaedo telling
Echecrates about Socrates’ last conversation and death, encompasses the en-
tire narrative (above, Chapter 4). Phaedo’s account of Socrates’ death emplots
a series of arguments for the immortality of the soul and introduces the Theory
of Forms, all put into the mouth of Socrates and discussed with two alternat-
ing interlocutors, Simmias and Cebes. As in the Lysis and the Gorgias, the se-
quence is cumulative. To quote A.E. Taylor,

A possible misconception which would be fatal to a real understanding

of the dialogue is to look upon the members of the series of arguments
for immortality as so many independent substantive “proofs,” given by
the author or the speaker as all having the same inherent value. Any care-
ful study will show that they are meant to form a series of “aggressions”
to the solution of the problem, each requiring and leading up to the com-
pleter answer which follows it.

Taylor also emphasizes that “Plato is careful, by skillful use of dramatic by-play
and pauses in the conversation, to let us see what he regards as the critical
points in the argument.”48
The standard two-stage opening of the framed narrative that understand-
ably does not involve a change of scene, is punctuated by modification in the
cast of characters, gradually focusing our attention on Socrates and his circle.

48 Taylor 1960 [1926]: 177. Cf. also McCabe 2006: 41.

The Limits of Authority 143

The conversation proper starts when the last of the outsiders, Socrates’ wife
Xantippe, leaves the stage.49
Socrates has just been released from chains (59c), a circumstance clearly
meant to be understood in more than one sense. The discussion proper is pre-
ceded by a general conversation focused on Socrates having spent his time in the
prison composing poetry. It leads to Socrates’ assertion that a true philosopher
should be willing to die, although it would be unlawful for him to take his own life.

Here he changed his position, and put his legs off the couch on to the
ground, and during the rest of the conversation he remained sitting.
“Why do you say,” enquired Cebes, “that a man ought not to take his own
life, but the philosopher will be ready to follow the dying?” (61c10–d5)

Socrates’ change of position signals the beginning of the discussion. Cebes,

who asked the question, will converse with him first.
(1) Socrates – Cebes: It is unlawful to commit suicide (61d–63d). Humans are the
possession of gods, who are their guardians, and are not at liberty to take their
own lives. Again Cebes points out the ostensible contradiction between this be-
lief and the philosopher’s willingness to die (62d).
*Simmias interferes to express his support of Cebes, and replaces him as
Socrates’ interlocutor. But before the Socrates-Simmias exchange begins,
Socrates has to take care of what he obviously sees as a minor distraction:

“I will do my best,” replied Socrates. “But you must first let me hear what
Crito wants; he has long been wishing to say something to me.” “Only
this, Socrates,” replied Crito, “the attendant who is to give you the poison
has been telling me, and he wants me to tell you, that you are not to talk
much, talking, he says, increases heat, and this is apt to interfere with
the action of the poison; persons who excite themselves are sometimes
obliged to take a second or even a third dose.” “Then,” said Socrates, “let
him mind his business and be prepared to give the poison twice or even
thrice if necessary; that is all.” (63d3–e5)

(2) Socrates – Simmias: The philosopher lives the death (61d–63d). Socrates
explains that during his life the philosopher prepares himself for the moment
of death, looking forward to the highest reward he will receive in the afterlife.

49 Phd. 60a9–b3 “And some of Crito’s people led her away, crying out and beating herself,
whereas Socrates, sitting up on the couch, bent and rubbed his leg, saying, as he was rub-
bing …”
144 Chapter 6

*Cebes intervenes, expressing his apprehension that the soul may not survive
after death, and subsequently takes Simmias’ place as Socrates’ interlocutor.
(3) Socrates – Cebes: The first argument for the immortality of the soul (69e–
73a). The argument proceeds from “an ancient doctrine” (palaios … tis logos)
that the souls “go from hence into the other world, and returning hither, are
born again from the dead” (70c5–8). Cebes accepts Socrates’ argument, and
remarks that Socrates’ own doctrine that knowledge is recollection leads to the
same conclusion.
*“‘But remind me, Cebes,’ said Simmias, interposing (hupolabōn), ‘what
proofs are urged in favor of this doctrine of recollection. I am not sure that at
the moment I remember them’” (73a4–6).
(4) Socrates – Simmias: The second argument for the immortality of the soul
(73a–77c). Since the process of acquiring knowledge can be shown to amount
to recollection of what we have already known (cf. the Meno), our souls must
have accumulated this knowledge before our birth. Simmias is satisfied with
the proof insofar as the existence of the soul before birth is concerned. “But
that after death the soul will continue to exist is not yet proven even to my own
satisfaction” (77b1–2).
*Cebes supports the objection, claiming that only about half of what was
required has been proven (77b-c). Socrates responds that the proof is already
given if the two foregoing arguments are put together. Cebes laughs and asks for
further reassurance: “Then, Socrates, you must argue us out of our fears (77e3–
4).” He also expresses his apprehension that, when Socrates is gone, no one will
be left to charm away the fear of death. Socrates disagrees, claiming that there
are people in Greece and elsewhere fit to fulfill this mission. The digression
ends with Cebes’ call to return to the point at which they digressed (78a10–b1).
(5) Socrates – Cebes: The third argument for the immortality of the soul (78a–
84c). Socrates argues that the soul is immortal because, as distinct from the
body, it is akin to the unchangeable and the divine (79d–80b). This returns him
to the vision of the afterlife and the philosopher’s reward in it (80c–84c).
*This is followed by a pause:

When Socrates had done speaking, for a considerable time there was
silence; he himself appeared to be meditating, as most of us were, on
what had been said. Only Cebes and Simmias spoke a few words to one
another. And Socrates observing them asked what they thought of the
argument, and whether there was anything wanting. (84c1–6)

Until now, each stage in the conversation was introduced by a change of in-
terlocutor, and each change of interlocutor, except that between (2) and (3),
The Limits of Authority 145

was accompanied by the narrator’s text.50 But now, a major crisis occurs.51 Sim-
mias remarks that the idea of the immortality of the soul will be at variance
with the doctrine that the soul is a harmony, or attunement, which dissolves
after death, whereas Cebes suggests that, having “worn” many bodies, the soul
may perish nevertheless (85e–88b). The transition from the first to the second
objection is again punctuated by the narrator’s text, which allows us to visual-
ize the entire scene: “Socrates looked fixedly at us as his manner was, and said
with a smile …” (86d5–6).
The objections of Simmias and Cebes act as a transition to the final argu-
ment for the immortality of the soul. The transition is further foregrounded (a)
by frame-breaking (88d–89a, Echecrates’ first interruption of Phaedo’s story,
quoted above, p. 91); (b) by a change of interlocutor (89b, Phaedo replaces Sim-
mias and Cebes); (c) by the narrator’s text (89bc, the description of Socrates’
playing with Phaedo’s hair), and (d) by a metanarrative digression on misolo-
gists, “the haters of arguments” (89d–91c), which is addressed to Phaedo.52 This
prepares the ground for Socrates’ response to the objections and for the intro-
duction of the Theory of Forms.
(6) Socrates – Simmias: Socrates’ reply to Simmias (91d–95a). The doctrine
that the soul is a harmony contradicts the theory of recollection, already ac-
cepted by Simmias. Asked to choose between the two, Simmias chooses the
theory of recollection.
*Now Socrates turns to Cebes, recapitulating the latter’s objection (95b-e).
But instead of dealing with it directly, he embarks on a detailed account of his
life-long search for truth. The sudden transition is marked by the narrator’s text:
“Socrates paused awhile, and seemed to be absorbed in reflection” (95e7–8).
(7) Socrates – Cebes: The Theory of Forms (96a–102a). Socrates’ account of his
intellectual biography (96a–100a) culminates in his exposition of the Theory of
Forms (100b–101e).
*“‘What you say is most true,’ said Simmias and Cebes, both speaking at
once” (102a2). This is immediately followed by yet another frame-breaking,
namely, Echecrates’ expression of his equally enthusiastic support (102a3–9,
quoted at p. 91 above). “But what followed?” Echecrates asks, and Phaedo
resumes his story. After a short digression introducing the objection of an

50 In the transition from (3) to (4), the narrator’s text consists of only one word, “interposing”
(hupolabōn).
51 Cf. Rowe 1993: 200: “the equivalent of the reversal of fortune, or peripeteia (Aristotle, Poet-
ics Ch. 6 etc.) in a tragedy.”
52 Sedley 1995: 14: “the very important methodological passage.” For a similar treatment of a
digression in the Theaetetus see above, p. 130. On the significance of digressions in Plato
see Brumbauch 1988; Rutherford 1995: 283; Schur 2014: 81–98.
146 Chapter 6

a­ nonymous ­bystander and again accompanied by a description of Socrates’

body language,53 the narrator arrives at the final argument for the immortal-
ity of the soul. The transition is punctuated by Socrates addressing Cebes: “At
the same time, turning to Cebes, he said: ‘Are you at all disconcerted, Cebes, at
our friend’s objection?’” (103c2–4). There is no change of interlocutor, which
emphasizes the fact that the forthcoming argument for the immortality of the
soul flows naturally from the Theory of Forms.
(8) Socrates – Cebes: The fourth and the final argument for the immortality of
the soul (103c–107a). Soul is what makes a body alive; life is therefore the soul’s
essential property. It cannot admit the opposite of life, death, without ceasing
being soul. The argument has reached its conclusion:

“Then, Cebes, beyond question, the soul is immortal and imperish-

able, and our souls will truly exist in another world!” “I am convinced,
Socrates,” said Cebes, “and have nothing more to object; but if my friend
Simmias, or anyone else, has any further objection to make, he had better
speak out …” (106e9–107a5)

*Simmias has nothing to say nor any reason for doubt, but he still feels un-
certain when he thinks of the greatness of the subject and the frailty of man.
Socrates agrees, remarking that insofar as one sees the initial premises (tas …
hupotheseis tas prōtas) as sound, nothing remains but to follow the course of
the argument (logos), “and if that be plain and clear, there will be no need for
any further enquiry” (107b8–9).
(9) Socrates – Simmias: The myth of the afterlife (107a–114c). But it seems that
the argument alone is nevertheless not quite enough and, as in the Gorgias
and the Republic, the concluding muthos of the Phaedo counterbalances the
preceding logos.54
*The return to the setting and the description of Socrates’ death is followed
by Phaedo’s concluding words addressed to his narratee Echecrates (118a15–17).

...
The most remarkable feature of the interaction of plot-building tools in a sin-
gle dialogue is the dynamics of their clustering. The plot-building elements as

53 Phd. 103a11–b1 “Socrates inclined his head to the speaker and listened. ‘I like your courage,’
he said, ‘in reminding us of this.’”
54 Cf. the words with which Socrates introduces the eschatological myth at the end of the
Gorgias (Socrates to Callicles): “Give ear, then, as they say, to a very fine logos, which you,
I suppose, will consider muthos, but I consider logos” (523a1–2).
The Limits of Authority 147

a rule accumulate at the point where the dialogue’s main argument is to be

introduced. Thus in the Lysis, the transition to the central argument is empha-
sized by the narrator’s description of the emotional state of his former self, by
a metanarrative comment on the unreliability of the preceding argument and
by a change of interlocutor. In the Gorgias, the importance of Socrates’ “speech
of Amphion” is foregrounded by a series of metanarrative comments relating
to the prospect of finishing the discussion and by the narrator’s text describing
Socrates’ intention to leave, followed by Gorgias’ intervention. The clustering is
especially dense in the Phaedo. Here, the culminating argument of the Theory
of Forms is preceded not only by the highly developed narrator’s text and a
digression on misologists but also by frame-breaking and the emergence of an
interlocutor who previously has taken no part in the discussion. The suspense
is intensified step by step, and its purpose is unmistakable: it is a carefully con-
structed crescendo that marks the climax of the dialogue.
Like zooming-in, clustering is a literary device attested already in Homer. As
Norman Austin showed years ago, the density, length, and location of Homeric
digressions stand in direct ratio to the significance of the episode they intro-
duce: the more important the episode, the more extensive the digressions an-
ticipating it; for example, the first encounter of the Achaean and Trojan armies
at the beginning of Iliad 3 is foregrounded in the concluding part of Iliad 2
by a series of five similes (2.455–83), by an elaborate invocation of the Muses
(2.484–93), and by two extended catalogues (2.494–877); the importance of
Patroclus’ first intervention in the course of events, which is the turning point
in the poem’s plot, is emphasized by an especially long reminiscence by Nestor
(11.642–805), and so on.55 Since such clusters typically signal the approach of a
strategic point in the evolving plot, this must have been a narrative device that
Plato’s audience had no difficulty interpreting.

6 Conclusions

Although the authority vested in Plato’s narrator is far from negligible, it does
not spread to the story and the plot of the dialogue. Again, the narrator c­ annot
be responsible for the story because it is conceived as an argument or a se-
ries of arguments made in a conversation that is presented as having really
occurred. Nor can he be responsible for the plot because, as we have seen in
this chapter, the narrator’s text is only one of the tools by whose means the
dialogue becomes emplotted. This shows unequivocally that as regards the
dialogue’s plot the narrator does not hold a position of authority.

55 Austin 1966; cf. Edwards 1987: 48.

148 Chapter 6

The principal contribution of Plato’s narrator is his establishing a certain

perspective of the events described and in communicating it to the reader.
That is, the narrator is the one who focalizes the story and gives his own ver-
sion of it. He is entitled to edit the story as he sees fit and to tell it in the lan-
guage and style he chooses. But it is not by the narrator that the dialogue’s
fictional universe is fashioned. G.R.F. Ferrari’s assessment of the narrative of
the Republic further illuminates this point:

Plato, we can begin to see, is at pains in the Republic to distinguish his

narrative control from that of Socrates, its internal narrator. Plato is the
author of a fiction on an epic scale, who must bear in mind the complex
structure of the whole as he writes, and who can weave into his web not
only grand themes but also smaller motifs that may recur over long in-
tervals … The character Socrates, on the other hand, is neither relating a
fiction, nor is he even engaged in artistic speech. He is presenting to an
unnamed person or persons a philosophical conversation that he had the
day before …56

The narrator is thus subordinated to an authority that lies beyond the author-
ity at his disposal. This ultimate authority finds its expression in the strategies
of emplotment, that is, the way the subject matter is organized: it is predomi-
nantly in these strategies that the authorial design becomes manifest. To quote
Jean Laborderie, “It is always in the presentation, the arrangement, and the
composition that the author’s presence manifests itself in the most interest-
ing manner.”57 This raises the question about the role played by the ultimate
authority responsible for the dialogue as a whole.

56 Ferrari 2010: 18–19; cf. Ferrari 2010: 28–29.

57 Laborderie 1978: 364; my translation. See also ibid.: “La récurrence de procédés de style,
de langue ou de composition permet de mieux saisir la place que tient la personalité de
Platon dans les Dialogues.”
 Chapter 7

The Narrator and the Author

1 Poetry and Painting in Republic 10

As we have seen, in Republic 3 Plato classifies mimesis as a subdivision of nar-

rative discourse (diēgēsis) (Chapter 1). This is not yet to say that the idea of mi-
mesis as an autonomous category was alien to Plato. To see that, it will suffice
to turn to Republic 10.
In the last book of the Republic the discussion of mimesis is resumed but
taken in a different direction. While Republic 3 approached mimesis from the
perspective of narrative voice and its manifestations in different forms of dis-
course (lexis), Republic 10 explores the ontological dimension of mimetic art as
a whole. Besides obvious links between the two books, some notable disconti-
nuities are also evident.1 Thus, while in Republic 3 mimesis is partly accepted in
the ideal state (see Chapter 1), in Republic 10 it is dismissed altogether, only the
non-mimetic hymns to gods and praises of good men being allowed.2 More to
the point, the classification of literary genres from the standpoint of narrative
voice as introduced in Book 3 is ignored, and the discussion makes a fresh start.
The renewed discussion of mimesis, which occupies the first half of Repub­
lic 10, rests on analogy between poetry and painting.3 The painter (zōgraphos)
is capable of creating everything with his art, as if he is turning a mirror
round and round; his creations, however, are nothing more than appearances
(phainomena), which stand at the third remove from reality. The painter is
therefore just an imitator (mimētēs) of things that others make. This is pre-
cisely the position occupied by the mimetic poet (596d-597e). Furthermore,
neither painters nor poets know anything about the objects of their imitation,
and their creations, though deceiving the unsophisticated public, cannot be
considered faithful to their originals:

1 See, e.g., Annas 1981: 335–44; Finkelberg 1998: 4–6 (with bibliography); Capuccino 2014:
55–65.
2 See esp. Resp. 595a5 “It [poetry] should absolutely not be admitted insofar as it is mimetic (τὸ
μηδαμῇ παραδέχεσθαι αὐτῆς ὅση μιμητική).”
3 On Plato’s ambivalent attitude toward painting see Keuls 1974; Demand 1975; cf. Friedländer
1958: 119–20. For a discussion of Plato’s treatment of painting and poetry in Republic 10 from
the standpoint of general aesthetics see Halliwell 2002: 118–47.

© koninklijke brill nv, leiden, ���9 | doi:10.1163/9789004390027_008

150 Chapter 7

The poet is like a painter who, as we have already observed, will make a
likeness of a cobbler though he understands nothing of cobbling; and his
picture is good enough for those who know no more than he does, and
judge only by colors and figures (ek tōn chrōmatōn kai schēmatōn).4

Finally, painting (hē graphikē) and poetry are equally removed not only from
truth but also from reason (phronēseōs), in that they act on the lower, namely,
the irrational, part of the soul (603a-605a). In sum, the poet is the painter’s
complete counterpart (antistrophon 605a9).
We saw that in Republic 3 Plato leaves no classificatory option to mimetic
discourse not mediated by narrative. His placing mimetic poetry next to paint-
ing in Republic 10 overturns this classification. This brings us back to the view
of mimesis as a non-narrative form of expression.

2 The Body of the Dialogue

By itself, Plato’s association of poetry with painting may well be seen as part of
his singularly vituperative attack against poetry in Republic 10 and therefore not
strictly relevant to his view of his own dialogues. Yet this impression vanishes
once we take a closer look both at the Republic itself and at the dialogues that
followed it – the Phaedrus, the Statesman, the Timaeus, the Critias, the Philebus,
the Laws. References to painting, almost all of them self-referential, emerge
with increasing frequency both in the second part of the Republic and in these
later dialogues. Let us start by considering two passages from Republic 6:

“No city would ever prosper unless painters using a divine pattern (para­
deigma) drew its lines (diagrapseian). … Taking the city and the habits of
men like a tablet, first they wipe it clean … And then, after this, wouldn’t
they outline (hupograpsasthai) the sketch (schēma) of the constituion?

4 Resp. 600e6-601a2; cf. also 597d-599a. Burnyeat 1999: 302–305, does not agree that the viewers
described by Plato were indeed meant to be “deceived into thinking that the painted car-
penter is a real carpenter” and offers an alternative interpretation. Yet, this is not to take into
account illusionistic trends in fifth- and fourth-century bce painting and sculpture: their
cultivation of the trompe l’oeil effect was famously thematized in the anecdote of a painting
contest between Zeuxis and Parrhasius (Plin. HN 36). Plato’s criticism of mimetic poetry and
painting in Republic 10 and of mimetic sculpture in the Sophist (below) emerged in response
to these contemporary developments. On painting see Richter 1987: 277–78; on sculpture see
Pollitt 1972: 157–59, 174–78; on Plato having in mind contemporary trompe l’oeil painting see
Halliwell 2002: 134–35.
The Narrator and the Author 151

… And then, in finishing off their work (apergazomenoi) they would look
frequently in both directions … blending and mixing (summeignuntes te
kai kerannuntes) the image of man from customs and usages. … And they
would wipe away portions, and paint them again until they had rendered
human ways as pleasing to the gods as possible.” “Indeed,” he said, “in no
way could a fairer picture (hē graphē) be produced.”5
And of the virtues too we must behold not the outline (hupographē)
merely, as at present – nothing short of the most finished picture (tēn
teleōtatēn apergasian) must satisfy us. (504d6-8)

Note that terms associated with painting, such as “picture” (graphē), “out-
line” (hupographē, hupograpsasthai), “sketch,” or “figure” (schēma), “blending”
(summeignuntes) and “mixing” (kerannuntes) [sc. of colors], and “finishing off”
(apergasia, apegrazesthai) – some of them the same as those used for the dis-
paragement of mimetic art in Republic 10 (above, p. 150) – are simultaneously
applied both to the creators of the ideal state and to the discourse about that
state, that is, the dialogue itself.6 This becomes even more obvious from the dis-
cussion of the timocratic state in Book 8, where the same terms are employed:

Such is the origin and such the character of this State, of which we have
only outlined a verbal sketch (logōi schēma … hupograpsanta) and not
finished it off in every detail (mē akribōs apergasasthai), for even an out-
line (ek tēs hupographēs) is enough to discern the most just and most
unjust. (548c9–d2)

Note also that, whether the picture under consideration is verbal or visual, its
creation is invariably envisaged as a two-stage process. At the first stage, only
a schematic outline is drawn; at the second stage, a full-fledged picture is pro-
duced by this outline being filled with colors.

5 Resp. 500e2-501c3. Tr. Nancy Demand, slightly adapted. Cf. also Resp. 600e6-601a2 (quoted in
Section 1 above) and Ti. 37c (quoted below, p. 152). For paradeigma as applied to the painter’s
work see also Resp. 472d.
6 Cf. Halliwell 2002: 130, on Plato intimating that the Republic itself is “a kind of philosophi-
cal world-picture.” At the same time, I do not share Halliwell’s sweeping generalization that
“philosophers are painters in another medium, in the sense that they endeavor to give vivid
realization or embodiment to ideas conceived in and held before their minds” (ibid.). This
would admittedly be true only of those philosophers who, like Plato, embody their ideas in
literary form through the medium of artistic representation.
152 Chapter 7

The same terms and the same two-stage process re-emerge in the States­
man. Here, however, they are taken a step further. Somewhere around the mid-
dle of the dialogue the Eleatic Stranger announces to his audience:

Although we may have described a sort of royal figure (schēma), we have

not as yet accurately finished off (ou … apeirgasmenoi… di’ akribeias) the
statesman … And that is what we must do, if we do not mean to bring
disgrace upon the discourse (ton logon) at its close. (268c5-d2)

And, several pages later:

And our discourse (logos), just like some picture of a living creature
(zōion), seems to be sufficient in broad outline (tēn exōthen perigraphēn),
but somehow has not yet acquired the vividness (enargeian) which is
given by dyes (tois pharmakois) and the blending of colors (tēi suggkrasei
tōn chrōmatōn). Now to those endowed with higher abilities (tois duna­
menois) any picture of a living being (zōion) had better be rendered by
language and argument (lexei kai logōi) rather than by painting or any
work of art (graphēs de kai sumpasēs cheirourgias): to the rest by works of
art. (277c1-6; cf. 277ab)

The distinction drawn here between hoi dunamenoi, “those endowed with
higher abilities,” and hoi alloi, “the rest,” correlated as it is with the levels of
text addressing each of the two types of audience, is noteworthy, and we shall
return to it shortly.
In the Timaeus, Socrates uses similar terms in referring to the version of the
Republic that he delivered a day before:

I should like, before proceeding further, to tell you how I feel about the
State which we have described. I might compare myself to a person who,
on beholding beautiful animals (zōia kala) either created by the painter’s
art (hupo graphēs eirgasmena) or, better still, alive but at rest, is seized
with a desire of seeing them in motion or engaged in some struggle or
conflict to which their forms appear suited; this is my feeling about the
State which we have been describing. (19b3-c2)

This compares well with the following passage from Book 6 of the Laws:

You know the endless labor which painters expend upon their pictures
of living creatures (tōn zōiōn) – they are always putting in or taking out
The Narrator and the Author 153

colors, or whatever be the term that artists employ; they seem as if they
would never cease arranging (kosmousa) their works, which are always
being made brighter and more beautiful. (769a7-b3)

Yet another comparison of the process of literary composition to the process

of painting can be found in the Critias, whose protagonist states unequivocally
that “we may observe the same thing [as in painting] to happen in discourse
(kata tous logous)” (107d5-6).7
In the passages cited, Plato repeatedly exploits the double meaning of the
Greek noun zōion, which designates both a living creature and a picture of a
living creature as painted by zōgraphos, the painter – literally “one who draws
pictures of living creatures.” As far as I can see, this supplies a broader context
to the famous passage in the Phaedrus, which has come to be generally re-
garded as the standard definition of organic unity:

Every composition (logos) ought to be formed (sunestanai) like a zōion,

having a kind of body (sōma ti) of its own, so as not to be devoid of either
a head or feet, but be provided with both middle parts and extremities,
drawn (gegrammena) so as to fit one another and the whole. (264c2-5)

Just as in the Timaeus passage about “beholding zōia kala either created by the
painter’s art or, better still, alive but at rest” (above, p. 152), the Phaedrus use of
the noun zōion simultaneously evokes both a living creature and a picture of it.
An explicit comparison of writing to painting emerges in another key pas-
sage of the Phaedrus:

I think writing (graphē) has this strange feature, which makes it like
painting (zōgraphiai). The offspring of painting stand there as if alive
(zōnta), but if you ask them something, they preserve a quite solemn si-
lence. Similarly with written words (hoi logoi)…8

Only words written together with knowledge in the soul of the learner know
to whom they should speak and to whom they should say nothing. “You mean
the discourse, living and endowed with soul (logon… zōnta kai empsuchon), by
the man who knows (tou eidotos),” Phaedrus remarks, “of which the written

7 See also Criti. 107b-e, esp. “all that is said by any of us [about the gods] can only be mime-
sis and likeness (μίμησιν μὲν γὰρ δὴ καὶ ἀπεικασίαν τὰ παρὰ πάντων ἡμῶν ῥηθέντα χρεών που
γενέσθαι)” (107b5-7).
8 Phdr. 275d4-7; tr. C.J. Rowe, slightly changed.
154 Chapter 7

one (ho gegrammenos) would rightly be called a kind of phantom (eidōlon).”

“Absolutely.”9
Clearly, the Phaedrus opposition of “the living discourse endowed with soul”
to “the written one,” which is nothing but a phantom, amounts to much the
same as the Statesman’s distinction between “any picture of a living being ren-
dered by language and argument” on the one hand, and a picture rendered by
“any painting or work of art” on the other.10 Moreover, in both the Phaedrus
and the Statesman the verbal and the visual media address two different kinds
of audience: a select few in the former case and a mass audience in the latter.
Note also that, as in the case of the zōion, Plato plays on the double meaning of
the noun graphē, which designates both painting and writing.
To recapitulate this part of our discussion, in the Republic, the Phaedrus,
and the Statesman the dialogue is envisaged as consisting of two layers of pre-
sentation that correspond to two successive stages in the process of composi-
tion: (a) the general outline, presumably an argument drawn in words and (b)
its vivid elaboration in material form, visual or written, both rendered by the
term graphē.11 Each of the two layers of presentation addresses a different kind
of audience: the chosen ones are satisfied with “language (lexis) and argument
(logos)” (the Statesman), or with “the living discourse (logos) endowed with
soul” (the Phaedrus), whereas the rest need a “painting (graphē) or any work
of art” (the Statesman) or the “phantom” of the written discourse (graphē) (the
Phaedrus). The latter are of course the same people referred to disparaging-
ly in Republic 10 as capable of judging the painter’s work “only by colors and
figures.”12
However, the opposition of the living logos endowed with soul to the eidōlon
embodied in a picture or a written text is far from Plato’s last word on the

9 Phdr. 276a5-b1, cf. 277a.

10 Cf. also drawing a “verbal sketch” (schēma) as opposed to finishing it off in detail at Resp.
548cd (quoted in full above, pp. 150–51).
11 Cf. Phlb. 39b3-7 on the mental process of creating images: “I must bespeak your favor also
for another craftsman (dēmiourgos), who is busy at the same time in the chambers of the
soul.” “Who is he”? “The painter, who, after the scribe has done his work, paints in the soul
images of the things which he has described (τῶν λεγομένων εἰκόνας ἐν τῇ ψυχῇ τούτων
γράφει).” Cf. Davidson 2013 (1990): 354 with n. 22.
12 Resp. 601a2 ἐκ τῶν χρωμάτων δὲ καὶ σχημάτων, quoted in full above, p. 150. See also the so-
called philosophical digression of the Seventh Letter, where Plato sets out to throw light
on the part of his teaching that cannot be committed to writing because it is too danger-
ous for the general public and can be mastered only by those few who “are capable of
discovering the truth for themselves with a little guidance” (341e2-3). This is the very part
of Plato’s doctrine that is defined by him as “the greatest” and “the most serious” (341b1,
344c6).
The Narrator and the Author 155

­matter. This is made clear from the Philebus and the Timaeus. Towards the end
of the former, Socrates addresses Protarchus:

And now you and Philebus must tell me whether anything is still wanting
in this blending (tēi sugkrasei tautēi),13 for to my mind, just as some incor-
poreal arrangement (kosmos tis asōmatos), which is going to hold fair rule
over a body endowed with soul (empsuchou sōmatos), the present dis-
course (logos) appears to have been finished off (apeirgasthai). (64b5-8)

The language of the passage recalls the Republic and the Statesman (see es-
pecially apeirgasthai). But it is a comparison with the Phaedrus that proves
particularly rewarding. As in the Phaedrus, the discourse (logos) is treated here
as incorporeal – or, more precisely, as equivalent to an “incorporeal arrange-
ment” (presumably the mixture of knowledge and pleasure); but as distinct
from the Phaedrus, the analogy suggests that the logos is now seen not as sepa-
rated from the body but as governing it, apparently acting as its soul. Thus,
unlike Phaedrus 264c, we find the entire dialogue made analogous not just to
the body, but to the body and the soul united as one. This effectively super-
sedes the Phaedrus’ dichotomy between “the living discourse endowed with
soul,” on the one hand, and the “phantom” of its material incarnation, on the
other. Thus, what in the Republic and the Statesman was presented as a two-
stage process of drawing the outline and its subsequent elaboration by filling it
with colors, and in the Phaedrus and the Statesman again as an opposition of
the verbal argument to its vivid visual or written presentation (see especially
enargeian at Plt. 277c2), is treated in the Philebus as a unity of the soul and the
body of the dialogue.14
It is certainly more than a mere coincidence that the Philebus applies the
same language to the entire universe:

And whence comes this soul, my dear Protarchus, unless the body of the
universe (to ge tou pantos sōma), which contains elements like those in
our bodies but in every way fairer, had also be endowed with soul (em­
psuchon)? (30a5-7; cf. 29e1-3, 30a3)

The Timaeus rendering of the process of creation by the Demiurge is set in

similar terms:

13 That is, in the mixture of knowledge and pleasure arrived at in the course of the discussion.
14 It is hard to avoid the impression that Aristotle’ treatment of the dramatic plot (muth­
os) as the “soul” of tragedy is a development of this specific set of ideas. See Arist. Poet.
1450a37-8.
156 Chapter 7

For which reason, [when he was framing the universe] having put intel-
ligence (noun) in soul, and soul in body, he was fashioning the whole in
such a way that he might have finished off (eiē … apeirgasmenos) a work
(ergon) which was by nature fairest and best. Wherefore, using the lan-
guage of probability, we may say that this world (kosmon) became a living
creature truly endowed with soul and mind (zōion empsuchon ennoon te)
by the providence of God. (30b4-c1)

Further on, the Demiurge’s work of “finishing off” the created universe is
described:

When the father and creator (ho gennēsas patēr) saw it [the created
universe] moving by itself and alive, the created image (agalma) of the
eternal gods, he rejoiced, and in his joy determined to finish it off (aper­
gasasthai) so as to make it still more like the pattern (to paradeigma).
(37c6-38a1)

The Timaeus passage is also closely similar to the description given in Republic
6 of “finishing off” (apergazomenoi) a verbal picture of the ideal state with the
assistance of the divine paradeigma (above, p. 150).
In view of these smooth transitions from the fashioning of a dialogue to the
fashioning of the universe and back, it is hard to avoid the conclusion that at
some stage Plato came to see the dialogue as an artifact fashioned in the same
way as the universe itself. As Donald Davidson put it,

The created universe, and everything in it which belongs to the world of

becoming, is for Plato a product of art, both those things which are cre-
ated by man and those which are not. There is therefore nothing in the
created universe which cannot be explained and analyzed in terms of an
art.15

This of course makes the “father and creator” of the dialogue an agency analo-
gous with the divine craftsman. The reality he fashions, however, is of a differ-
ent order.

15 Davidson 2013 (1990): 133; on the creator as craftsman in the Philebus and the Timaeus see
Davidson 2013 (1990): 269–71. Cf. Halliwell 2002: 71: “In some of Plato’s later writing this
second perspective [mimesis as representation] is expanded by a sense that the world
itself is a mimetic creation, wrought by a divine artist who, at one point in the Timaeus
(55c6), is expressly visualized as a painter.”
The Narrator and the Author 157

3 The Gardens of Adonis

In the Phaedrus, immediately after Phaedrus’ remark about the living logos en-
dowed with soul, of which the written logos could only be a phantom (above),
Plato tells a parable:

Then tell me this: the sensible farmer who had some seeds he cared about
and wanted to bear fruit – would he sow them with serious purpose dur-
ing the summer in some garden of Adonis, and delight in watching it be-
coming beautiful within eight days, or would he do that for the sake of
play (paidias … charin) and the festival, when he did it at all; whereas for
the purposes about which he was in earnest, he should make use of the
science of farming and sow them in appropriate soil, being content if
what he sowed reached maturity in the eighth month?16

In the same way, he who has some knowledge of “good and beautiful things” will
not be serious when writing them “in black water” and sowing them through
a pen: “his gardens of letters (en grammasi kēpous) he will sow and write for
the sake of play (paidias charin), when he does write … and he will be pleased
when he watches their tender growth” (276c3-d5). Although much preferable
to other forms of play (paidia), for example, the symposia, these writings still
cannot be equated with sowing into a fitting soul, with the assistance of the art
of dialectic, of “words accompanied with knowledge,” that is, with the direct
teacher-disciple interaction (276d5-e7). This, of course, addresses once again
the hierarchy of oral and written discourse discussed in the previous section.
Nevertheless, we should not be misled by Plato’s ostensibly dismissive lan-
guage when he speaks of his writings. In Plato, especially in the later dialogues,
the term paidia (whose usual translation “amusement” does it little justice)
serves to designate a wide range of cultural activities, from music,17 sports18
and symposia19 to liberal education20 and human life itself. The reference to
the latter deserves to be given in full:

16 Phdr. 276b1-8, tr. C.J. Rowe, with slight changes. “The gardens of Adonis”: miniature artifi-
cal gardens produced by sowing seeds of quickly germinating plants into a basket or a
bowl. Rapidly growing but short-living, they were placed on the roofs during the popular
Athenian festival of the god Adonis.
17 Lach. 188d4; Euthyd. 277d9; Resp. 424d5, 424e6; Plt. 268b2; Laws 656c3, 657c4, 657d3,
659e4, 673c9, 673d4, 764e4, 771e6, 796b7.
18 Laws 796d4, 829b7, 830e5, 832d6, 834c3, 834d3.
19 Phdr. 276 d 6; Laws 666b5, 671e6, 673e8, 844d6.
20 Euthyd. 278b2, 278b3; Resp. 572c3; Plt.308d3; Laws 819b3, 819c2, 820d5.
158 Chapter 7

I say that about serious matters (spoudaion) a man should be serious,

and about matters which are not serious he should not; and that God is
the natural and worthy object of our most serious and blessed endeavors,
for man, as I said before, is designed (memēchanēmenon) to be a kind
of plaything (ti paignion) of God. And this, truly considered, is the best
of him; wherefore every man and woman should live their lives follow-
ing this route and making their play as noble as possible (paizonta hoti
kallistas paidias). … And what is the right way of living? We should pass
our lives playing certain games (paizonta … tinas dē paidias) – sacrificing,
singing, dancing, so as to be able to propitiate the gods…21

As far as I can see, the fact that culture and peaceful life in general are treated
here as a kind of play (the serious activity to which it is favorably contrasted is
war), provides the missing context for Plato’s calling writing a variety of paidia,
still more as several times he applies this term directly to his own dialogues or
their parts. In the Symposium, Agathon presents his speech as a mixture of play
(paidia) and seriousness (spoudē) offered to the God of Love (Symp. 197e7); in
the Parmenides, Parmenides refers to his forthcoming exposition of the dialec-
tical method as “playing a laborous paidia;”22 the admixture of myth to seri-
ous discourse in the Statesman is also termed paidia (Plt. 268d8, 268e5); in the
Timaeus, one’s moving from the certainties of the eternal being to the prob-
abilities of becoming is recognized as a moderate and thoughtful sort of paidia
capable of filling one’s entire lifetime;23 finally, the conversation of three old
men in the Laws is twice referred to as “thoughtful paidia.”24
This is the background against which Plato’s recurrent treatment of mimet-
ic activities as paidia should be approached – so much so that in the Sophist
he styles mimesis as the quintessential paidia. “And is there any more sophis-
ticated (technikōteron) or refined (chariesteron) kind (eidos) of paidia than the
mimetic one?” the Eleatic Stranger asks Theaetetus. The latter replies: “Never.
For you have named here a very broad category (eidos) which embraces many
different things.”25

21 Laws 803c2-e3.
22 Prm.137b2 πραγματειώδη παιδιὰν παίζειν. On the playful character of Parmenides’ argument
in this dialogue, already recognized as such in the Early Academy, see Zuckert 1998: 876, 894.
23 Ti. 59d1-2 μέτριον ἂν ἐν τῷ βίῳ παιδιὰν καὶ φρόνιμον ποιοῖτο.
24 Laws 685a7-8 περὶ νόμων παίζοντας παιδιὰν πρεσβυτικὴν σώφρονα; 769a1 ἡ πρεσβυτῶν
ἔμφρων παιδιά.
25 Soph. 234b1-4; my translation. See also Resp. 396e2, 602b8; Soph. 234a6, 234a9, 235a6; Plt.
288c9; Laws 889d1; cf. Epin. 975d2 and Chapter 1, n. 9. Cf. Laborderie 1978: 106–7; Capra
2014: 178.
The Narrator and the Author 159

Notably, the spheres Plato subsumes under the category of paidia are re-
peatedly treated as standing for phenomena of a secondary order: becoming
in contrast to being, sports in contrast to the authentic battle experience,
mimesis in contrast to reality, men in contrast to gods. It speaks volumes
about Plato’s attitude to polis religion that he places communal rituals,
among them religious festivals and sacrifice (above, with n. 21), in the sphere
of play. Nevertheless, as he emphasizes over and over, these are the very ac-
tivities that please the gods. “And our virgin lady,” says the Athenian Stranger
in the Laws,

delighting in the playfulness of the dance (tēi tēs choreias paidiai),

thought it not fit to play (athurein) with empty hands; she must be closed
in a complete suit of armor, and in this attire go through the dance; and
youths and maidens should in every respect imitate (mimeisthai) her, es-
teeming highly the favor of the Goddess, both with a view to the necessi-
ties of war, and to festive occasions.26

This is the background against which Plato’s comparison of his writings to the
Gardens of Adonis should be placed. Artificial gardens created for the celebra-
tion of a religious festival are of course only an imitation of real fields grow-
ing real crops, but they reproduce these fields on a miniature scale, and they
please the god. Likewise the fashioning of the microcosm of the dialogue,
which as we have seen reproduces in all particulars the fashioning of the uni-
verse. Belonging as it does to the second-order sphere of human activity, it
reproduces in miniature the workings of the divine demiurge, thus creating a
reality sui generis. The latter is made especially clear from the direction which
Plato’s discussion of the ontological status of mimetic art, first introduced in
Republic 10, takes in the Sophist.

4 Mimesis and Reality

In Book 10 of the Republic Plato sets out to prove that any artistic representa-
tion lies at a third remove from reality. The work of fine arts should be placed
first after the idea of the artifact and second after the work of the craftsman
who produces the artifact; by the same token, the work of poetry should be
placed first after the idea of justice and second after the activity of the king.

26 Laws 796b6-c4. Cf. also Laws 796b3, referring to military dances as “imitations”
(mimēmata) of war.
160 Chapter 7

But did Plato really mean, as is often supposed, that the picture of a couch is
an exact copy of the real couch made by the craftsman? In fact, his argument
concerning the relation between the work of mimetic art and reality is not as
straightforward as many are ready to admit.27
Judging by what Plato says in Republic 10, he does not think mimesis capable
of producing exact replicas of its models. This holds not only for the painter
who reproduces the couch made by the craftsman but also for the kind of mi-
mesis in which the craftsman himself, reproducing the one and only idea of the
couch, is involved: the work of both is only “a dim adumbration (amudron ti)
in comparison with truth” (Resp. 597a10-11). The premise from which Plato pro-
ceeds in this and similar contexts is that any mimetic representation would
involve distortion of its model, thus turning every form of reproduction into
nothing more than “a dim adumbration.” This would be true even of the kind
of reproduction supplied by the object’s reflection in a mirror: as Plato em-
phasized more than once, since it exchanges left for right and vice versa, the
reflection in the mirror is actually a standard example of distortion of reality.28
The second part of Plato’s argument concerns only mimetic art. Whereas
a real couch is always the same, although it looks different “according as you
view it from the side or the front or in any other way,” the only thing that the
picture of a couch is able to reproduce is this very appearance. Accordingly,
such a picture is nothing more than a phantom:

Then the mimetic art is far removed from truth, and this, it seems, is the
reason why it can create (apergazetai) everything, because it touches or
lays hold of only a small part (smikron ti) of each object and that a phan-
tom (eidōlon), as, for example, a painter, we say, will paint us a cobbler, a
carpenter, and other craftsmen, though he himself has no expertness in
any of these arts, but nevertheless if he were a good painter, by exhibiting
at a distance his picture of a carpenter he would deceive children and
foolish men, and make them believe it to be a real carpenter.29

No work of mimetic art can thus be regarded as an exact reproduction of the

object it imitates. We should conclude therefore that no such thing as the exact
copy or replica of a real thing is possible.
The argument is recapitulated and taken further on in the Sophist. In this
dialogue, which resumes the discussion of the ontological status of mimetic

27 Finkelberg 1998, 181–91 and 2014: 153–56.

28 Tht. 193c; Ti. 46a, 71b; Soph. 239d.
29 Resp. 598b6-c4, tr. P. Shorey, with slight changes; cf. also 598a and above, with n. 4.
The Narrator and the Author 161

art started in Republic 10, Plato is at pains to eliminate the claim, first presented
in the Republic, that the mimetic artist is able “to create everything” with his
art. Yet he cannot avoid the conclusion that, insofar as the objects of art can-
not be envisaged as replicas of already existing things, their presence poses
an ontological problem. This is especially true of sculptors or painters whose
works are of colossal size:

If they were to reproduce the true proportions of a well-made figure, as

you know, the upper parts would look too small, and the lower too large,
because we see the one at a distance, the other close at hand…. So artists,
leaving the truth to take care of itself, do in fact put into the images they
make, not the real proportions, but those that will appear beautiful.30

Further on in the same dialogue, Plato admits, again fully in accord with
Republic 10, that even the images created by mirror-like imitation, namely
those that do not deliberately distort their model’s proportions, cannot be re-
garded as this model’s truthful representations. The conclusion, again, is that
creating the exact likeness of a given natural object is impossible.
There is thus no reality of which the “imitation” can be considered a copy.
This fact turns the existence of the work of mimetic art into a real ontological
puzzle:

The truth is, my friend, that we are faced with an extremely difficult ques-
tion. The “appearing” or “seeming” without really “being,” and the saying
of something which yet is not true – all these expressions have always
been and still are deeply involved in perplexity (aporia). It is extremely
hard, Theaetetus, to find correct terms in which one may say or think that
falsehoods (pseudē) have a real existence, without being caught in a con-
tradiction by the mere utterance of such words. (236e1-237a1)

Still, as these “falsehoods” cannot be regarded as representations of truth – or,

in Plato’s own words, as “another truth” (talēthinon … heteron 240a8) – they
must be credited with some sort of autonomous existence. This is precisely the
conclusion Plato eventually reaches in the Sophist. His reasoning is quite sim-
ple: since images of art cannot be regarded as copies of really existing things
(and we have seen that Plato envisages no situation in which this can be pos-
sible), these images cannot be regarded as reflections of reality any more than
lies can be regarded as a reflection of truth. Or, in Plato’s words again,

30 Soph. 235e-236a, tr. F.M. Cornford.

162 Chapter 7

…anyone who talks of false statements or false judgments as being im-

ages (eidōla) or likenesses (eikones) or copies (mimēmata) or semblances
(phantasmata), or of any of the arts concerned with such things, can
hardly escape becoming a laughing stock by being forced to contradict
himself. (241e2-5)

Accordingly, just like the Gardens of Adonis and other kinds of cultural play,
images of art can only be treated as reality of a secondary order, “a dim adum-
bration” containing only a small element of truth (Resp. 597a10-11, 598b7-8).
This opens the way to Plato’s negotiating the status of his own dialogues along
similar lines. As we saw above in this chapter, he chose to present the reality he
himself created as a series of visually imprinted images, each functioning as a
microcosm, with the author playing the role of the demiurge.

5 Representation of Narration

Where does the narrator enter this picture? In Chapter 1 we arrived at the
conclusion that since the narrative voice in Plato’s dialogues never belongs to
the author, all the dialogues, their narrative mode notwithstanding, should be
identified as mimetic narrative, or “diegesis through mimesis,” a subcategory
of narrative discourse introduced in Book 3 of the Republic. Now, considering
the direction taken by the discussion of mimesis in Republic 10, this definition
should be adjusted and taken further.
Plato envisaged philosophical dialogue not only as a form of narrative, with
narrator, story, and plot of its own, but also as a representational artifact – a
picture, as it were.31 It seems more than mere coincidence that his insistent
equating writing with painting, which we saw applied in the second half of the
Republic, in the Phaedrus, the Statesman, the Timaeus, the Critias, the Philebus,
and the Laws, was paralleled by the adoption of dramatic form as his only liter-
ary medium. “I let them converse among themselves,” Euclides says in the The­
aetetus about his converting what was supposed to be a narrated dialogue into
dramatic form (above, p. 5). Cicero’s allusion to this passage is illuminating:

I committed the main points of the discussion to memory, and have set
them out in the present book in my own way; for I have, so to speak,
brought the actors themselves on the stage in order to avoid too frequent

31 Cf. Nightingale 2004: 107, about Plato’s conceiving of knowledge as involving both “saying”
and “seeing.”
The Narrator and the Author 163

repetition of “said I” and “said he,” and to create the impression that they
are present and speaking in person.32

Cicero’s choice of words unambiguously indicates that in his eyes at least the
outcome of the switch from narrated to dramatic form will be equivalent to
transforming a piece of reported speech into a live spectacle. As we have seen,
in the Critias Socrates already admits as much when comparing the conversa-
tion in which he acts as an implicit narrator to a theatrical performance with
himself as a spectator (p. 73).
Even more remarkable is the fact that such transformation of narrative
into a visual phenomenon does not make Euclides, Socrates, or indeed any
­other narrator portrayed by Plato disappear from the picture. Plato’s insistence
that the narrative act always be present in his dialogues obviously requires
explanation.
Since Plato envisaged the dialogue as a kind of picture, we are entitled to ask
what precisely the picture in question was supposed to represent. The simple
answer would be that it was supposed to represent a philosophical conver-
sation, something on a par with Jacques-Louis David’s The Death of Socrates,
demonstrably intended as a visual representation of the Phaedo. What is, how-
ever, missing in David’s representation of Socrates’ last conversation is the
narrative act by whose means this conversation is communicated in Plato’s
dialogue. The Death of Socrates is a Phaedo without Echecrates, which makes
it signally un-Platonic.
We have indeed seen that with few exceptions Plato does not negotiate his
dialogues as direct representations of philosophical conversations. The conver-
sations depicted in Plato’s dialogues are communicated to us, explicitly or im-
plicitly, through the mediation of a narrator. Hence in Plato we encounter not
a philosophical conversation as such, but a philosophical conversation filtered
through the act of narration. Rather than to David’s painting, it would be more
appropriate to compare Plato’s dialogues to Velázquez’s Spinners (The Fable of
Arachne), which places the producer of the tapestry representing the myth of
Arachne, as well as the act of production itself, in the picture’s foreground.33
In such multi-level dialogues as the Parmenides, the Symposium, or the
Theaetetus we encounter a receding row of perspective-building narrators,
each placed within the picture: again, Velázquez’s play in the Spinners with a

32 Cic. Amic. 1.3. Tr. W.A. Falconer.

33 As distinct from the Meninas and other old master paintings which incorporate the figure
of the painter, the Spinners is unique in that here the producer of the representation of
the Arachne myth is not the same person as the creator of the painting.
164 Chapter 7

r­ eceding row of pictorial representations and their producers comes to mind.

On the tapestry produced by the spinners Arachne shows Athena her own tap-
estry, which reproduces Titian’s Rape of Europa. However, we cannot get to Ti-
tian’s painting without first absorbing the interior of the workshop where the
tapestry depicting the myth of Arachne is being produced; secondly we must
take in the representation of the Arachne myth, of which the representation
of the Rape of Europa is made part. Small wonder that for centuries the subject
matter of Velázquez’s painting was misidentified.34
It follows, then, that philosophical dialogue as negotiated by Plato should
be approached as a representation (mimēsis) of perspectively organized narra-
tion (diēgēsis) of a philosophical conversation. What the reader of the dialogue
immerses into is therefore not the fictional universe by and of itself but the
fictional narrative act whereby this universe is communicated.35
Why was the narrator’s presence so important to Plato? As we have seen,
in the formula “diegesis through mimesis,” which covers the entire genre of
Socratic dialogue, the diegesis and the mimesis alike belong to the author.
But the author, Plato, never speaks in his own voice in his dialogues. He is the
creator of a fictional universe whose fullness is not communicated through
the author’s personified voice.36 So if Plato insisted, as he apparently did, on
presenting all his dialogues as narratives without at the same time assuming
the role of their narrator, the introduction of the figure of narrator, explicit or
implicit, was a required step. For Plato, therefore, the narrator was a given, the
only question being how his presence is to be negotiated.
There is more than one indication in Plato’s work that he was acutely aware
that any representation is perspectively organized.37 As we have seen, this was
the main argument in his rejection of the claim that a mimetic r­ epresentation

34 The Spinners had been believed to represent an everyday scene in a tapestry workshop; it
was conclusively identified as The Fable of Arachne only in the middle of the 20th century,
with the discovery of a 17th-century inventory quoting this title.
35 Cf. Schaeffer 2013: 45, on the target domain of narrative immersion in the simulation
theories of fiction: “does the reader or spectator immerse into a (fictional) world, or into a
narrative act depicting a world? Does narrative fiction induce immersion through mimet-
ic primers feigning descriptive utterances, or simply through a perspectively organized
mentally centered and phenomenologically saturated presentation of a universe?.”
36 Cf. Rimmon-Kenan 2002 [1983]: 90: “Thus, while the narrator can only be defined circu-
larly as the ‘narrative voice’ or ‘speaker’ of a text, the implied author is – in opposition
and by definition – voiceless and silent. In this sense the implied author must be seen as
a construct inferred and assembled by the reader from all the components of the text.”
37 On more general implications of this aspect of Plato’s thought see Nightingale 2002. Cf.
also Nightingale 2004: 111: “Plato’s theoretical philosopher only attains a partial vision of
the Forms and therefore does not achieve a perfect ‘frontal’ view.”
The Narrator and the Author 165

can faithfully reproduce reality. Thus, we saw him speaking in Republic 10 of

the couch appearing diversely according to the angle from which it is observed,
and of a picture of the couch as bound to reproduce this limitation of our per-
ception, whereas in the Sophist we saw him musing about sculptors who have
to falsify the proportions of human body to make the colossal statues appear
proportional to the viewers.38 More generally, we saw him claiming in the
­Republic, the Phaedrus, and the Statesman that the adequate interpretation of
both a picture and a dialogue ultimately depends on how the individual view-
er or reader is cognitively positioned.39 This does not necessarily mean that
Plato saw the single narrative perspective that he pursued in his dialogues
along the same lines, only that he was aware of the nature of such phenomena
and the role they play. Otherwise, it would be difficult to explain why, for exam-
ple, he assigns to Socrates the role of spectator even in the dialogues where, as
in the Sophist, the Statesman, the Timaeus, and the Critias, his function as the
implicit narrator is minimalized (above, pp. 70–74); or why, as we witnessed
more than once, he invested so much effort in “switching off” any focus of per-
ception different from the narrator’s.
No less telling are Plato’s attempts, discussed in Chapter 4, to place both
the narrator and the narratee of the mixed dialogues in the fictional universe
he created.40 We saw indeed that Plato goes to great lengths to demarcate the
boundaries of the fictional universe in such a way that the narrative act in its
entirety may be presented as placed within it. The reason is obvious: by plac-
ing not only the narrator but also the narratee within the fictional universe,
he aimed to control the way his dialogues will be read.41 Equally remarkable
is his life-long struggle, described in detail in Chapter 5, to make his fictional
universe as seamless as possible by hiding the narrator behind the apparently
innocent mimetic façade of dramatic dialogues, thereby rendering him as in-
visible as the author himself.
All these factors point unambiguously to the position assigned to the nar-
rator in Plato’s dialogues. Plato’s narrator is an inseparable part of the fictional
universe created by the author. Proceeding from the vantage point designed
for him by the author, the narrator in turn draws a picture of his own, which
assumes the form of a perspectively organized report of real events. The single
narrative perspective is thus an integral component of philosophical dialogue

38 Resp. 598a-b; Soph. 235e-236a, cf. 234b. See also above, pp. 160–61.
39 Resp. 600e-601a; Phdr. 275d-e; Plt. 277a-c. See also above, p. 154.
40 Cf. above, Chapter 4, with n. 43.
41 This is made especially obvious in the Euthydemus and the Phaedo, see Chapter 4, with n.
42, and Collins 2012. Cf. also Chapter 5, p. 114, on the intradiegetic character of metalepsis
in Plato’s multi-level dialogues.
166 Chapter 7

as conceived by Plato. Everything indicates that its main purpose is unobtru-

sively to regulate the readers’ reception and response.
This is not to say that the only thing readers of Plato’s dialogues are sup-
posed to do is follow the narrator’s lead. As we have seen, Plato’s dialogues
are plotted in such a way that their overall design delivers a message of its
own, which lies outside the narrator’s authority. We have also seen that this
message is not reducible to any single argument appearing in a given dialogue
or to their mechanical sum.42 As a result, Plato’s readers have to work hard at
putting things together for themselves. This is obviously what Plato meant us
to do.

6 Conclusions

The complex fabric of Plato’s dialogues is governed by a holistic vision of the

dialogue as a quasi-living artifact – a microcosm, as it were – brought to life
by the playful activity of mimetic representation. The author is the demiurge
of this microcosm, he who creates a fictional universe by fashioning a plotted
narrative out of the arguments he purports to set forth (Chapter 6). However,
the reader can only enter this universe through the mediation of the narrator.
The narrator is the dialogue’s gatekeeper, a filter through whose agency it is
communicated to the reader. His main function is to control how the dialogue
is to be received by the reader by sustaining a certain perspective of it. There is
little room for doubt about whose perspective it ultimately is. The narrator is
a trusted agent of the author and the principal communication tool at his dis-
posal. Small wonder, then, that Plato’s narrators, explicit and implicit alike, are
either Socrates’ close associates – his friends and disciples Crito, Chaerephon,
Aristodemus, Apollodorus, Phaedo, Echecrates, Euclides, Terpsion – or, above
all, Socrates himself.43
In each of the three groups of his dialogues, Plato uses the narrator in a
similar way. In the narrated dialogues, the narrator openly exercises his con-
trol over the narrative by directly transmitting his perspective to the reader by
means of markers of focalization embodied in the narrator’s text (Chapter 2).

42 Cf. Griswold 2002: 97–98: “The whole of the drama is greater than the parts and cannot be
explained through them alone. What is not explained within the dialogue seems to show
us the intentions of Plato.”
43 Only three of Plato’s named narrators do not belong to this category: the impartial Eu-
dicus of the Hippias Minor (implicit; see further above, pp. 64, 109); the initially ignorant
but eventually enthusiastic Lysimachus of the Laches (implicit; see above, p. 66); and the
thoroughly sympathetic Pythodorus of the Parmenides (explicit; see pp. 41–43).
The Narrator and the Author 167

In the dramatic dialogues the act of narration is suppressed, and the narra-
tor becomes implicit as a result; nevertheless, his perspective is transmitted,
again, through the same markers of focalization attached to one character to
the exclusion of the others (Chapter 3). In the mixed dialogues, which also sup-
press the act of narration, the implicit narrator of the dramatic frame imposes
his control on this part of the dialogue in the same way as the narrator of the
dramatic dialogues, whereas the explicit narrator of the body of the dialogue
controls the main narrative by the same means as the narrator of the narrated
dialogues (Chapter 4).
To judge by the diachrony of the narrative voice in Plato’s dialogues as dis-
cussed in Chapter 5, Plato came to see the model of implicit narrator, the one
that allows the dialogue to act like a narrative without looking like one, as the
optimal solution for the issue of narrative voice. Highly sophisticated as it cer-
tainly is, this solution, along with zooming-in, metalepsis, and clustering, is
rooted in traditional narrative techniques attested as early as Homer. Consider
the following simile, which emerges at the end of Book 8 of the Iliad:

As when in the sky the stars about the moon’s shining,

are seen in all their glory, when the air has fallen to stillness,
and all the high places of the hills are clear, and the shoulders out-jutting,
and the deep ravines, as endless bright air spills from the heavens
and all the stars are seen, and the shepherd’s heart rejoices.44

Rather than being presented as a panoramic picture standing entirely on its

own, the nocturnal landscape described in the simile is focalized and emotion-
ally reacted upon by a shepherd: it is through the shepherd’s eyes that we are
supposed to view it.45 “We should remind ourselves,” Wayne C. Booth wrote
in The Rhetoric of Fiction, “that any sustained inside view, of whatever depth,
temporarily turns the character whose mind is shown into a narrator.”46 The
anonymous shepherd of Homer’s simile is the implicit narrator of the moment
of stillness of nature that he beholds. At the same time, he is himself part of the
picture drawn by the author. This is precisely how Plato’s narrator is positioned.

44 Il. 8.555–59, tr. R. Lattimore, slightly adapted; my emphasis.

45 Cf. de Jong 1987: 134: “the position the N[arrator]F[ocalizer]1 chooses is that of the shep-
herd.” Il. 8.555–59 is far from being the only example of this kind, cf. Il. 13.493–94, 21.346–
47; Od. 6.102–106, 22.302–306, with Kirk 1990: 341: “the herdsman’s reaction is typical.”
46 Booth 1983: 164.
Bibliography

Abbot, H. Porter. 2011. “Narrativity.” In Hühn et al.

Annas, Julia. 1981. An Introduction to Plato’s Republic. Oxford: Clarendon Press.
Arieti, James A. 1991. Interpreting Plato: The Dialogues as Drama. Savage, MD: Rowman
and Littlefield.
Austin, Norman. 1966. “The Function of Digressions in the Iliad.” GRBS 7: 295–312.
Bal, Mieke. 1981. “Notes on Narrative Embedding.” PT 2: 41–59.
Bal, Mieke. 1997 [1985]. Narratology. Introduction to the Theory of Narrative. 2nd ed.
Toronto, Buffalo, London: University of Toronto Press.
Baldick, Chris. 2001. The Concise Oxford Dictionary of Literary Terms. 2nd edn. Oxford:
Oxford University Press.
Bartels, Felix. 2015. “Zur Deutung der Digression des Theaitet (172c–177c).” Philologus
159: 29–72.
Beversluis, John. 2000. Cross-examining Socrates: A Defence of the Interlocutors in Pla-
to’s Early Dialogues. Cambridge: Cambridge University Press.
Blondell, Ruby. 2002. The Play of Character in Plato’s Dialogues. Cambridge: Cambridge
University Press.
Bloom, Allan. 1991 [1968]. “Preface.” In The Republic of Plato. Ed. Allan Bloom. 2nd edn.
Harper Collins, xi–xxiv.
Blössner, Norbert. 2011. “The Unity of Plato’s Meno. Reconstructing the Author’s
Thoughts.” Philologus 155: 39–68.
Booth, Wayne C. 1983. The Rhetoric of Fiction. 2nd edn. Chicago – London: University
of Chicago Press.
Bowery, Anne Marie. 2007. “Know Thyself: Socrates as Storyteller.” In Gary Alan Scott.
Philosophy in Dialogue. Plato’s Many Devices. Evanston, Illinois: Northwestern Uni-
versity Press, 82–110.
Brandwood, Leonard. 1990. The Chronology of Plato’s Dialogues. Cambridge: Cam-
bridge University Press.
Brooks, Peter. 1992 [1984]. Reading for the Plot. Design and Intention in Narrative. Cam-
bridge, MA: Harvard University Press.
Brooks, Cleanth and Robert Penn Warren. 1959 [1943]. Understanding Fiction. Engle-
wood Cliffs: Prentice-Hall.
Broombauch, Robert S. 1988. “Digression and Dialogue: The Seventh Letter and Plato’s
Literary Form.” In Griswold 1988, 84–92.
Burnyeat, Myles F. 1999. “Culture and Society in Plato’s Republic.” Tanner Lectures on
Human Values 20: 217–324.
Capra, Andrea. 2014. Plato’s Four Muses. The Phaedrus and the Poetics of Philosophy.
Washington, DC: Harvard University Press.
170 Bibliography

Capuccino, Carlotta. 2014. ARCHE LOGOU. Sui proemi platonici e il loro significato fi-
losofico. Florence: Olschki.
Chance, Thomas H. 1992. Plato’s Euthydemus. Analysis of What Is and Is Not Philosophy.
Berkeley – Los Angeles: University of California Press.
Charalabopoulos, N.G. 2012. Platonic Drama and Its Ancient Reception. Cambridge:
Cambridge University Press.
Clay, Diskin. 1983. “The Tragic and the Comic Poet of the Symposium.” In J.P. Anton and
A. Preus (eds.). Essays in Ancient Greek Philosophy. Vol. 2. Albany, NY: State Univer-
sity of New York Press, 321–39.
Clay, Diskin. 1988. “Reading the Republic.” In Griswold 1988, 19–33.
Clay, Diskin. 1992. “Plato’s First Words.” YCS: 29: 113–29.
Clay, Diskin. 1994. “The Origins of the Socratic Dialogue.” In Paul A. Vander Waerdt
(ed.). The Socratic Movement. Ithaca – London: Cornell University Press, 23–47.
Clay, Diskin. 2000. Platonic Questions. Dialogues with the Silent Philosopher. University
Park: Pennsylvania State University Press.
Cohn, Dorit. 1990. “Signposts of fictionality.” PT 11: 753–804.
Cohn, Dorit. 2001. “Does Socrates Speak for Plato? Reflections on an Open Question.”
NLH 32: 485–500.
Cohn, Dorit. 2005. “Métalepse et mise en abym.” In Pier and Schaeffer, 121–30.
Collins, J.H. 2012. “Prompts for Participation in Early Philosophical Texts.” In Elizabeth
Minchin (ed.). Orality, Literacy and Performance in the Ancient World. Leiden –
Boston: Brill, 151–82.
Collins, J.H. 2015. Exhortations to Philosophy: The Protreptics of Plato, Isocrates, and Ar-
istotle. Oxford – New York: Oxford University Press.
Collobert, Catherine. 2013. “La litérarité platonicienne: instances et modes narratifs
dans les dialogues.” RMM 4: 463–76.
Cornford, Francis M. 1937. Plato’s Cosmology. The Timaeus of Plato translated with a
running commentary. London: Routledge and Kegan.
Cotton, A.K. 2014. Platonic Dialogue and the Education of the Reader. Oxford: Oxford
University Press.
Cross, R.C., and A.D. Woozley. 1964, Plato’s Republic. A Philosophical Commentary. Lon-
don: Macmillan.
Danzig, Gabriel. 2010. Apologizing for Socrates. How Plato and Xenophon Created Our
Socrates. Lanham, MD: Lexington Books.
Danzig, Gabriel. 2014. “The Use and Abuse of Critias: Conflicting Portraits in Plato and
Xenophon.” CQ 64: 507–24.
Danzig, Gabriel. 2017. “Xenophon’s Symposium.” In Michael A. Flower (ed.). The Cam-
bridge Companion to Xenophon. Cambridge: Cambridge University Press, 132–51.
Danzig, Gabriel. 2018. “Introduction.” In Gabriel Danzig, Dave Johnson, Donald Mor-
rison (eds.). Plato and Xenophon. Comparative Studies. Leiden – Boston: Brill, 1–30.
Davidson, Donald. 2013 (1990). Plato’s Philebus. London – New York: Routledge.
Bibliography 171

de Jong, Irene J.F. 1987. Narrators and Focalizers. The Presentation of the Story in the
Iliad. Amsterdam: Grüner.
de Jong, Irene J.F. 2001. A Narratological Commentary on the Odyssey. Cambridge: Cam-
bridge University Press.
de Jong, Irene J.F. 2004. “Narratological Theory on Narrators, Narratees, and Narrative.”
In De Jong, Nünlist, and Bowie, 1–10.
de Jong, Irene J.F. 2009. “Metalepsis in Ancient Greek Literature.” In Grethlein and Ren-
gakos, 87–115.
de Jong, Irene J.F. 2014a. Narratology and Classics. A Practical Guide. Oxford: Oxford
University Press.
de Jong, Irene J.F. 2014b. “Diachronic Narratology (The Example of Ancient Greek Nar-
rative).” In Hühn et al.
de Jong, Irene J.F., René Nünlist, and Angus Bowie (eds.). 2004. Narrators, Narratees,
and Narratives in Ancient Greek Literature. Leiden – Boston: Brill.
Demand, Nancy. 1975. “Plato and the Painters.” Phoenix 29: 1–20.
Dodds, E.R. (ed.). 1959. Plato. Gorgias. A Revised Text with Introduction and Commen-
tary. Oxford: Clarendon Press.
Edmiston, William F. 1989. “Focalization and the First-Person Narrator: A Revision of
the Theory.” PT 10: 729–44.
Edwards, Mark W. 1987. “Topos and Transformation in Homer.” In J.M. Bremer, Irene J.F.
de Jong, and J. Kalff (eds.). Homer: Beyond Oral Poetry. Amsterdam: Grüner.
Erler, Michael. 2006. Platon. Munich: Beck.
Erler, Michael. 2013. “Plasma und Historie: Platon über die Poetizität seiner Dialoge.” In
Erler and Hessler 2013, 59–83.
Erler, Michael. 2015. “Argument and Context: Adaptation and Recasting of Positions in
Plato’s Dialogues.” In Nails and Tarrant, 91–105.
Erler, Michael. Forthcoming. “Argumentative aporia and dramatic euporia. The clo-
sures of and in the Protagoras and the Euthyphro.” In Kaklamanou, Pavlou, and
Tsakmakis, forthcoming.
Erler, Michael, and Erik Hessler (eds.). 2013. Argument und Literarische Form in antiker
Philosophie. Berlin – Boston: de Gruyter.
Ferrari, G.R.F. 1987. Listening to the Cicads. A Study of Plato’s Phaedrus. Cambridge:
Cambridge University Press.
Ferrari, G.R.F. 2000. Review of Press 2000. BMCR 00.11.10.
Ferrari, G.R.F. 2010. “Socrates in the Republic.” In Mark L. McPherran (ed.). Plato’s Re-
public. A Critical Guide. Cambridge: Cambridge University Press, 11–31.
Fine, Gail (ed.). 2008. The Oxford Handbook of Plato. Oxford: Oxford University Press.
Finkelberg, Margalit. 1997. “Plato’s Language of Love and the Female.” HThR 90: 231–61.
Finkelberg, Margalit. 1998. The Birth of Literary Fiction in Ancient Greece. Oxford: Clar-
endon Press.
Finkelberg, Margalit. 2006. “Aristotle and Episodic Tragedy.” G&R 53: 60–72.
172 Bibliography

Finkelberg, Margalit. 2014. “Diagnosing Fiction: From Plato to Borges.” In Anders Cull-
hed and Lena Rydholm (eds.). True Lies Worldwide: Fictionality in Global Contexts.
Berlin – Boston: de Gruyter, 153–65.
Finkelberg, Margalit. Forthcoming. “Frame and Frame-Breaking in Plato’s Dialogues.”
In Kaklamanou, Pavlou and Tsakmakis, forthcoming.
Fludernik, Monika. 1996. Towards a “Natural” Narratology. London – New York:
Routledge.
Fludernik, Monika. 2001a. “The Establishment of Internal Focalization in Odd Pro-
nominal Contexts.” In van Peer and Chatman, 103–13.
Fludernik, Monika. 2001b. “New Wine in Old Bottles? Voice, Focalization, and New
Writing.” NLH 32: 619–38.
Fludernik, Monika. 2003. “Scene Shift, Metalepsis, and the Metaleptic Mode.” Style 37:
382–400.
Fludernik, Monika. 2008. “Narrative and Drama.” In Pier and Garcia Landa, 353–81.
Ford, Andrew. 2008. “The Beginnings of Dialogue: Socratic discourses and fourth-cen-
tury prose.” In Goldhill, 29–44.
Ford, Andrew. 2010. “ΣΩΚΡΑΤΙΚΟΙ ΛΟΓΟΙ in Aristotle and Fourth-Century Theo-
ries of Genre.” CP 105: 221–35.
Forster, Edward M. 1972 [1927]. Aspects of the Novel. Harmondsworth: Penguin.
Frede, Dorothea. 1996. “The Hedonist’s Conversion: The Role of Socrates in the Phile-
bus.” In Gill and McCabe, 213–48.
Frede, Michael. 1992. “Plato’s Arguments and the Dialogue Form.” In Klagge and Smith,
201–19.
Frede, Michael. 1996. “The Literary Form of the Sophist.” In Gill and McCabe, 135–51.
Friedländer, Paul. 1958. Plato 1. An Introduction. Tr. Hans Meyerhoff. New York: Pan-
theon Books.
Friedländer, Paul. 1964. Plato 2. The Dialogues. First Period. Tr. Hans Meyerhoff. New
York: Random House.
Friedländer, Paul. 1969. Plato 3. The Dialogues. Second and Third Periods. Tr. Hans Mey-
erhoff. Princeton: Princeton University Press.
Fritz, John H. 2016. Plato and the Elements of Dialogue. Lanham, MD: Rowman and
Littlefield.
Gagarin, Michael. 1969. “The Purpose of Plato’s Protagoras.” TAPA 100: 133–64.
Genette, Gérard. 1972. Figures III. Paris: Éditions du Seuil.
Genette, Gérard. 1980. Narrative Discourse. An Essay in Method. Tr. Jane E. Lewin. Ithaca –
New York: Cornell University Press.
Genette, Gérard. 1990. “Fictional Narrative, Factual Narrative.” PT 11: 755–74.
Giannopoulou, Zina. 2014. “Authorless Authority in Plato’s Theaetetus.” In Martinez,
43–58.
Bibliography 173

Gill, Christopher and Mary Margaret McCabe (eds.). 1996. Form and Argument in Late
Plato. Oxford: Clarendon Press.
Goldhill, Simon (ed.). 2008. The End of Dialogue in Antiquity. Cambridge: Cambridge
University Press.
Gray, V. 2004. “Xenophon.” In de Jong et al. 2004, 377–88.
Grethlein, Jonas and Antonios Rengakos (eds.). 2009. Narratology and Interpretation.
The Content of Narrative Form in Ancient Literature. Berlin – New York: de Gruyter.
Griswold, Charles L., Jr. (ed.). 1988. Platonic Writings, Platonic Readings. New York –
London: Routledge.
Griswold, Charles L. 2002. “Irony in the Platonic Dialogues.” Ph&Lit 26: 84–106.
Guthrie, W.K.C. 1975. A History of Greek Philosophy IV. Plato: The Man and His Dia-
logues: Earlier Period. Cambridge: Cambridge University Press.
Guthrie, W.K.C. 1978. A History of Greek Philosophy V. The Later Plato and the Academy.
Cambridge: Cambridge University Press.
Halliwell, Stephen. 1997. “The Republic’s Two Critiques of Poetry.” In Otfried Höffe
(ed.). Platon: Politeia. Berlin: Akademie-Verlag, 313–32.
Halliwell, Stephen. 2002. The Aesthetics of Mimesis. Ancient Texts and Modern Problems.
Princeton: Princeton University Press.
Halliwell, Stephen. 2009. “The Theory and Practice of Narrative in Plato.” In Grethlein
and Rengakos, 15–41.
Halperin, David M. 1992. “Plato and the Erotics of Narrativity.” In Klagge and Smith,
93–129.
Hanebeck, Julian. 2017. Understanding Metalepsis. The Hermeneutic of Narrative Trans-
gression. Berlin – Boston: de Gruyter.
Haslam, Michael. 1972. “Plato, Sophron, and the Dramatic Dialogue.” BICS 19: 17–38.
Heath, Malcolm. 1989. “The Unity of Plato’s Phaedrus.” OSAP 7: 151–173.
Hösle, Vittorio. 2012. The Philosophical Dialogue. A Poetics and a Hermeneutics. Tr. Ste-
phen Rendall. Notre Dame, IN: University of Notre Dame Press.
Hühn, Peter, Jan Christoph Meister, John Pier, Wolf Schmid (eds.). the living handbook of
narratology. Hamburg: Hamburg University Press. http://www.lhn.uni-hamburg.de/.
Hühn, Peter, and Roy Sommer. 2013. “Narration in Poetry and Drama.” In Hühn et al.
Hunter, Richard. 2004. Plato’s Symposium. Oxford: Oxford University Press.
Hunter, Richard. 2012. Plato and the Traditions of Ancient Literature: The Silent Stream.
Cambridge: Cambridge University Press.
Huss, Bernhard. 1999. Xenophons Symposion. Ein Kommentar. Stuttgart – Leipzig:
Teubner.
Hutchinson, Douglas S. 1997. An introductory note to the translation of the Clitophon.
In John M. Cooper, Douglas S. Hutchinson (eds.). Plato. Complete Works. Indianapo-
lis – Cambridge: Hackett, 965–66.
174 Bibliography

Irwin, T.H. 1986. “Coercion and Objectivity in Plato’s Dialectic.” RIPh 40: 49–74.
Irwin, T.H. 2008. “The Platonic Corpus.” In Fine, 63–87.
Iser, Wolfgang. 1993. The Fictive and the Imaginary. Charting Literary Anthropology. Bal-
timore – London: Johns Hopkins University Press.
Jahn, Manfred. 2001. “Voice and Agency in Drama: Aspects of a Narratology of Drama.”
NLH 32: 659–79.
Johnson, William A. 1998. “Dramatic Frame and Philosophic Idea in Plato.” AJP 119:
577–98.
Jowett, Benjamin. 1892. “Introduction and Analysis.” In Benjamin Jowett (tr.). States-
man by Plato. Electronic edition. Global Grey 2014.
Joyal, Mark. 2014. “‘Genuine’ and ‘Bastard’ Dialogues in the Platonic Corpus: An Inquiry
into the Origins and Meaning of a Concept.” In Martinez, 73–93.
Kahn, Charles H. 1996. Plato and the Socratic Dialogue. The Philosophical Use of a Liter-
ary Form. Cambridge: Cambridge University Press.
Kahn, Charles. 2002. “On Platonic Chronology.” In Julia Annas and Christopher Rowe.
New Perspectives on Plato, Modern and Ancient. Washington, DC: Harvard Univer-
sity Press, 93–127.
Kahn, Charles H. 2013. Plato and the Post-Socratic Dialogue. The Return to the Philosophy
of Nature. Cambridge: Cambridge University Press.
Kaklamanou, Eleni, and Maria Pavlou. 2016. “Reading the Prooemium of Plato’s The-
aetetus: Euclides in Action.” GRBS 56: 410–37.
Kaklamanou, Eleni, Maria Pavlou and Antonis Tsakmakis (eds.). Forthcoming. Fram-
ing the Dialogues: How to Read Openings and Closures in Plato. Leiden – Boston: Brill.
Keuls, Eva. 1974. “Plato on Painting.” AJP 95: 100–27.
Kirk, G.S. (ed.). 1990. The Iliad: A Commentary. Vol. 2: books 5–8. Cambridge: Cam-
bridge University Press.
Klagge, James C. and Nicholas D. Smith (eds.). 1992. Methods of Interpreting Plato and
his Dialogues. Oxford: Clarendon Press.
Klauk, Tobias, and Tilmann. Köppe. “Telling vs. Showing.” In Hühn et al.
Kosman, L.A. 1992. “Silence and Imitation in the Platonic Dialogues.” In Klagge and
Smith, 73–92.
Kukkonen, Karin. 2014. “Plot.” In Hühn et al.
Laborderie, Jean. 1978. Le dialogue platonicien de maturité. Paris: Les Belles Lettres.
Long, A.G. 2013. Conversation and Self-Sufficiency in Plato. Oxford: Oxford University
Press.
Martinez, Javier (ed.). 2014. Fakes and Forgers of Classical Literature. Leiden – Boston:
Brill.
McCabe, Mary Margaret. 2006. “Form and the Platonic Dialogues.” In H.H. Benson
(ed.). A Companion to Plato. Malden, MA – Oxford: Blackwell.
McCabe, Mary Margaret. 2008. “Plato’s Way of Writing.” In Fine, 88–113.
McHale, Brian. 1987. Postmodernist Fiction. London: Routledge.
Bibliography 175

McKeon, Michael. 2002. The Origins of the English Novel, 1600–1740. The 15th anniver-
sary edition. Baltimore – London: Johns Hopkins University Press.
Meinwald, Constance C. 2008. “The Philebus.” In Fine, 484–503.
Michelini, Ann N. 2003. “Plato as Author: The Rhetoric of Philosophy. Introduction.” In
Ann N. Michelini (ed.). Plato as Author: The Rhetoric of Philosophy. Leiden – Boston:
Brill, 1–15.
Morgan, Kathryn. 2000. Myth and Philosophy from the Presocratics to Plato. Cambridge:
Cambridge University Press.
Morgan, Kathryn. 2003. “Plato’s Dream: Philosophy and Fiction in the Theaetetus.” In
Stelios Panayotakis, Maaike Zimmerman, Wytse Keulen (eds.). The Ancient Novel
and Beyond. Leiden – Boston: Brill, 101–14.
Morgan, Kathryn. 2004. “Plato.” In de Jong et al. 2004, 357–76.
Murley, Clyde. 1955. “Techniques of Modern Fiction in Plato.” CJ 6: 281–87.
Nails, Debra. 2002. The People of Plato. A Prosopography of Plato and Other Socratics.
Indianapolis: Hackett.
Nails, Debra, and Harold Tarrant (eds.). 2015. Second Sailing: Alternative Perspectives on
Plato. Helsinki: Societas Scientiarum Fennica.
Niederhoff, Burkhardt. 2011a. “Focalization.” In Hühn et al.
Niederhoff, Burkhardt. 2011b. “Perspective – Point of View.” In Hühn et al.
Nightingale, A.W. 1995. Genres in Dialogue. Plato and the Construct of Philosophy. Cam-
bridge: Cambridge University Press.
Nightingale, A.W. 2002. “Distant Views: ‘Realistic’ and ‘Fantastic’ Mimesis in Plato.” In
Julia Annas and Christopher Rowe (eds.). New Perspectives on Plato, Modern and
Ancient. Washington, DC: Harvard University Press, 227–47.
Nightingale, A.W. 2004. Spectacles of Truth in Classical Greek Philosophy. Theoria in its
Cultural Context. Cambridge: Cambridge University Press.
Nünning, Ansgar. 2001. “On the Perspective Structure of Narrative Texts: Steps toward a
Constructivist Narratology.” In van Peer and Chatman, 207–23.
Nünning, Ansgar. 2008. “Diegetic and Mimetic Narrativity: Some Further Steps towards
a Narratology of Drama.” In Pier and Garcia Landa, 329–52.
Nussbaum, Martha. C. 2001 [1986]. The Fragility of Goodness. Luck and Ethics in Greek
Tragedy and Philosophy. Updated edition. Cambridge: Cambridge University Press.
Pentassuglio, Francesca. 2017. Eschine di Sfetto, Tutte le testimonianze. Turnhout:
Brepols.
Pfister, Manfres. 1988. The Theory and Analysis of Drama. Tr. John Halliday. Cambridge:
Cambridge University Press.
Phelan, James. 2001. “Why Narrators Can Be Focalizers—and Why It Matters.” In van
Peer and Chatman, 51–64.
Pier, John. 2008. “After This, Therefore Because of This.” In Pier and García Landa,
109–40.
Pier, John. 2013. “Metalepsis.” In Hühn et al.
176 Bibliography

Pier, John. 2014. “Narrative Levels.” Revised version. In Hühn et al.

Pier, John, and J.Á. García Landa (eds.). 2008. Theorizing Narrativity. Berlin: de Gruyter.
Pier, John, and J.-M. Schaeffer (eds.). 2005. Métalepses. Entorses au pacte de la représen-
tation. Paris: Éditions de l’École des Hautes Études en Sciences Sociales.
Pollitt, J.J. 1972. Art and Experience in Classical Greece. Cambridge: Cambridge Univer-
sity Press.
Press, D.A. (ed.). 2000. Who Speaks for Plato? Studies in Platonic Anonymity. Lanham –
Boulder – New York: Rowman and Littlefield.
Puchner, Martin. 2010. The Drama of Ideas. Platonic Provocations in Theater and Phi-
losophy. Oxford – New York: Oxford University Press.
Richter, Gisela M.A. 1959. A Handbook of Greek Art. A Survey of the Visual Arts of Ancient
Greece. Oxford: Phaidon Press.
Ricoeur, Paul. 1984 [1983]. Time and Narrative. Vol. 1. Tr. Kathleen McLaughlin and Da-
vid Pellauer. Chicago – London: The University of Chicago Press.
Ricoeur, Paul. 1985 [1984]. Time and Narrative. Vol. 2. Tr. Kathleen McLaughlin and Da-
vid Pellauer. Chicago – London: The University of Chicago Press.
Rimmon-Kenan, Shlomith. 2002 [1983]. Narrative Fiction. Contemporary Poetics. 2nd
edn. London – New York: Routledge.
Rossetti, Livio. 2011 [2003]. “Le dialogue socratique in statu nascendi.” In Livio Rossetti.
Le dialogue socratique. St-Just La Pendue: Les Belles Lettres, 23–53.
Rowe, C.J. 1986. “The Argument and Structure of Plato’s Phaedrus.” PCPS 32: 106–25.
Rowe, C.J. 1989. “The Unity of the Phaedrus.” OSAP 7: 175–88.
Rowe, C.J. (ed.). 1993. Plato. Phaedo. Cambridge: Cambridge University Press.
Rowe, C.J. 1996. “The Politicus: Structure and Form.” In Gill and McCabe, 153–78.
Rowe, C.J. 2000. Plato: Phaedrus, with introduction, translation and commentary. Re-
vised edition. Warminster: Aris and Phillips.
Rowe, C.J. 2007. Plato and the Art of Philosophical Writing. Cambridge: Cambridge Uni-
versity Press.
Rutherford, R.B. 1995. The Art of Plato. Ten Essays on Platonic Interpretation. Cambridge,
MA: Harvard University Press.
Ryan, Marie-Laure. 2006. Avatars of Story. Minneapolis: University of Minnesota Press.
Schaeffer, Jean-Marie. 2013. “Fictional vs. Factual Narration.” In Hühn et al.
Scheffel, Michael. 2013. “Narrative Constitution.” In Hühn et al.
Schenker, D.J. 2006. “The Strangeness of the Phaedrus.” AJP 127: 67–87.
Schmid, Wolf. 2008. Elemente der Narratologie. 2nd edn. Berlin – New York: de Gruyter.
Schofield, Malcolm. 2008. “Ciceronian Dialogue.” In Goldhill 2008, 63–84.
Schultz, Anne-Marie. 2013. Plato’s Socrates as Narrator. A Philosophical Muse. Lanham –
Boulder – New York: Lexington Books.
Schultz, Anne-Marie. 2015. “Listening to Socrates in the Theaetetus: Recovering a Lost
Narrator.” In Nails and Tarrant, 107–24.
Bibliography 177

Schur, David. 2014. Plato’s Wayward Path. Literary Form and the Republic. Washington,
DC: Harvard University Press.
Scolnicov, Samuel. Plato’s Parmenides. Translated with Introduction and Commentary.
Berkeley – Los Angeles – London: University of California Press.
Sedley, David. 1995. “The Dramatis Personae of Plato’s Phaedo.” Proceedings of the Brit-
ish Academy 85: 3–26.
Sedley, David. 2003. Plato’s Cratylus. Cambridge: Cambridge University Press.
Sedley, David. 2004. The Midwife of Platonism. Text and Subtext in Plato’s Theaetetus.
Oxford: Oxford University Press.
Sermamoglou-Soulmaidi, Georgia. 2014. Playful Philosophy and Serious Sophistry. A
Reading of Plato’s Euthydemus. Berlin – Boston: de Gruyter.
Sheffield, F.C.C. 2001. “Alcibiades’ Speech: A Satyric Drama.” G&R 48: 193–209.
Sheffield, F.C.C. 2006. “The Role of the Earlier Speeches in the Symposium: Plato’s En-
doxic Method?” In James H. Lesher, Debra Nails, Frisbee Sheffield (eds). Plato’s Sym-
posium: Issues in Interpretation and Reception. Cambridge, MA: Harvard University
Press, 23–47.
Shen, Dan. 2003. “Difference Behind Similarity: Focalization in Third-Person Center-
of-Consciousness and First-Person Retrospective Narration.” In Carol Jacobs and
Henry Sussman (eds.). Acts of Narrative. Stanford, CA: Stanford University Press,
81–92.
Tarrant, Dorothy. 1955. “Plato as Dramatist.” JHS 75: 82–89.
Tarrant, Harold. 1996. “Orality and Plato’s Narrative Dialogues.” In Ian Worthington
(ed.). Voice into Text. Orality and Literacy in Ancient Greece. Leiden – Boston: Brill,
129–47.
Tarrant, Harold. 2000. Plato’s First Interpreters. Ithaca – New York: Cornell University
Press.
Tarrant, Harold. 2010. “The Theaetetus as a Narrative Dialogue.” Australian Society for
Classical Studies Proceedings 31: 1–17.
Taylor, A.E. 1960 [1926]. Plato. The Man and His Work. 7th edn. Trowbridge – London:
Methuen.
Thesleff, Holger. 1967. Studies in the Styles of Plato. In Thesleff 2009, 1–142.
Thesleff, Holger. 1982. Studies in Platonic Chronology. In Thesleff 2009, 143–382.
Thesleff, Holger. 2003. “A Symptomatic Text Corruption: Plato, Gorgias 448a5.” In
Thesleff 2009, 551–56.
Thesleff, Holger. 2009. Platonic Patterns. A Collection of Studies. Las Vegas – Zürich –
Athens: Parmenides Publishing.
Van Peer, Willie, and Seymour Chatman (eds.). New Perspectives on Narrative Perspec-
tive. Albany: SUNY.
Vlastos, Gregory. 1983. “The Socratic Elenchus.” OSAP 1: 27–58.
Werner, D. 2007. “Plato’s Phaedrus and the Problem of Unity.” OSAP 32: 91–137.
178 Bibliography

Whitmarsh, T.J.G. 2004. “Lucian.” In de Jong, Nünlist, and Bowie, 465–76.

Whitmarsh, T.J.G. 2013. “Radical Cognition: Metalepsis in Classical Greek Drama.” G&R
60: 4–16.
Wilamowitz-Moellendorff, Ulrich von. 1920. Platon II. Beilagen und Textkritik. Berlin:
Weidman.
Wolf, Werner. 2006. “Framing Borders in Frame Stories.” In W. Wolf and W. Bernhart
(eds.). Framing Borders in Literature and Media. Amsterdam: Rodopi, 179–206.
Zuckert, Catherine H. 1998. “‘Parmenides’: A Dramatic Reading.” Review of Metaphysics
51: 875–906.
Index of Passages Cited

Aeschines 437e8–438a1 133n27

Alcibiades 440e3–5 59
ssr vi A 43 28 Cri.
Miltiades 43a1 54
ssr vi A 76 28 43a4–6 54
Aristotle Criti.
On Poets 107b5–7 153n7
fr. 72 Rose 9n21 107d5–6 153
Poet. 108b2–7 73
1447a28–b13 9 Epist. 7
1447b14–15 9n23 341e2–3 154n12
1447b17–20 9 Euthyd.
1451b2–3 9n22 271a1–b8 85
275c5–6 86n22
Cicero 276b6–c1 67, 126
Amic. 276d1–2 67
1.3 162–63 277b3–5 130
283a1–2 86n22
Diogenes Laertius 290e1–2 87
3.48 30 291a4 87
3.50 50 291b7–c2 11
291d5–6 86n22
Homer 291e2 86n22
Il. 292d5–6 86n22
8.555–59 167 292e8–293a1 86n22
Od. 294d7–8 86n22
9.228–29 37n31 303a4–5 86n22
303b1–2 86n22
Plato 303b3–7 67
Alc. ii 304b6–7 86n22
138a1–3 10 307b6–c4 88
Chrm. Euthyphr.
153a1–6 31 2a1–4 53
153d2–154b7 32 2b8–11 53
154b8 34 3a7–8 53
155c5 34 3c3–4 53
155d3–4 33, 34 10e7–8 53
162c1–4 131 15e3–4 53
162d2 60n31 Grg.
169c7–d1 33 447b1–3 66
Cra. 458c3–7 67
383a1–5 59 481b6–9 68
397a4–5 133n27 485e3–7 140
427c9–d2 133n27 487e9–488a2 140
436e2 133n27 497b4–5 140
180 Index of Passages Cited

Plato (cont.) Menex.

505c1–2 140 234a1–b8 54
505c7–8 140n41 235e1–8 55
505c10–d2 140n41 236d1–2 55, 63n39
505d6–7 140n41 Meno
506a6–7 40n42 89e9–10 59–60
506a8–b3 141 95a2 60
506b4–6 141n45 99e2 60n30
510a1–2 141n46 Phd.
522e6–7 141n47 57a1–2 88
523a1–2 146n54 58d7–9 89
Hp. mai. 59a8–9 90
281a1–2 57 59e8–60a2 89n33
291a6–7 57 60a9–b3 143n49
301c6–d1 57 60b1–3 89n35
304d2 111n19 61c10–d5 143
Hp. mi. 63d3–e5 143
363a1–5 63, 106 64a10–b1 90
364b8–9 63 70c5–8 144
373b4–5 64 73a4–6 144
Ion 77b1–2 144
530a1–b3 56 77e3–4 144
La. 84c1–6 144
178a1–2 66 86d5–6 89n35, 145
180b7–c1 64–65 88c1–7 89–90, 90–91
181b5–c1 65 88c8–d3 91
189d1–3 66 89a9–b2 89n34
Leg. 95e7–8 145
625a6–b7 62 102a2 91, 145
625b8–c2 62 102a3–9 91, 92n42
685a7–8 158n24 102a10–b3 89n32
722c7–9 62 103a4–5 89n32
769a1 158n24 103a11–b1 89n35, 146n53
769a7–b3 152–53 103c2–4 146
796b6–c4 159 106e9–107a5 146
803c2–e3 158 107b8–9 146
811c7–9 62 117b5–6 89n35
Ly. 118a15–17 92
203a1–5 31 Phdr.
204e1–10 32 227a1–3 61
207d1–5 135 228d6–7 61
211a1–5 135 229a9–b2 61n32
213d1–e1 136 230b2–c4 61
218c2–8 136 234d3–5 61
220d4–7 137 237a4 61
222a4 137 242a1 61
222b1–2 137 242b8–c3 61, 128
Index of Passages Cited 181

243b6–7 61 314e3–315b8 81
264c2–5 153 316a3–b2 82
275d4–7 153 333e2–3 60n31
276a5–b1 153–54 334c7–8 67
276b1–8 157 335c7–d3 83, 128–29
276c3–d5 157 336b1–5 131
Phlb. 336d7 82
11a1–2 68 336e3–4 141n43
11c7–8 69n56 339d10–e5 82–83
12a9 69n56 361d2–5 84
12b3–4 69n57 362a4 54, 84
12b4–6 69n56 Resp.
15c8–9 69n56 327a1–c3 34–35
15d1–2 133 328b4–c1 35
16a4–5 69n57 336b1–5 131
20a4–5 69n57 336d5–6 35
29b1–2 69n57 350c12–d3 37
30a5–7 155 394b3–8 3, 5n12
36d6–10 134 396c5–e2 3–4, 119n34
39b3–7 154n11 396e4–7 4
45d6–7 69n55 398a7–b4 4, 119
49c6–7 69n55 449a7–b6 36
51d4–5 69n55 500e2–501c3 150–51
53e2 69n55 504d6–8 151
53e8 69n55 548c9–d2 151
57c6–7 69n55 595a5 149
64b5–8 155 597a10–11 160, 162
67b8–9 69n57 598b6–c4 160
Plt. 600e6–601a2 150, 154n12
268c5–d2 152 602b8 4n9
277a6–b6 134n28 Soph.
277c1–6 152 216a1–6 72
286b7–c1 134 216d3 73
286d1–2 134 234b1–4 158
286d4–6 119n34, 134n31 235e–236a 161
Prm. 236e1–237a1 161
126b1–7 40 240a8 161
126c1–2 43–44 241e2–5 162
127a7–b5 42 242b6–8 133
127c5–d5 42 Symp.
130a3–7 42 172a1–5 93
136e5–137b2 40–41, 133, 172a6–b3 44n39
158 173b1–4 94
137c2–5 7 178a1–5 94
Prt. 180c1–3 95
309a1–d3 80–81 185c4–6 95
310a2–4 81 198a1–2 67
182 Index of Passages Cited

Plato (cont.) Ti.

221c1–3 95–96 19b3–c2 152
222d3–4 12n34 20b1–2 71n67
223c2–d2 95 26e7–8 73
Tht. 27b7–9 73
142a4–c3 98 30b4–c1 156
142d4–5 98–99 37c6–38a1 156
143a2–5 99 59d1–2 158n23
143b5–c5 5, 162
143e6–7 100 P. Oxy.
144c5–d1 51 3219 49
157c4–6 52
169c1 51–52 Xenophon
177c3–4 52 Symp.
189e6–190a6 111 2–3 29
210b6–7 120
210d1–4 52
General Index

Act of narration See narrative act Alcibiades 10, 12, 44n39, 80–82, 86n21,
Aeschines of Sphettus 27–30 95–96, 100, 109n10, 131
Anger 35, 52, 57, 60, 82, 86, 106, 107, 127 Antiphon 7, 27–28, 38–44, 77–78, 94,
Annas, J. 28n2, 131n19, 149n1 98, 99
Apostrophic addresses 23, 34, 37, 86, 114 an unnamed Athenian (1) (Protagoras)
See also metalepsis 80–81, 101
Aristotle 9–10, 13, 23, 49–50 an unnamed Athenian (2) (Symposium)
Metaphysics 59n26, 71n63 93–94, 98, 101
Poetics 2n4, 9, 145n51, 155n14 Anytus 59–60, 75, 106–107, 127, 131
Audience (fictional) 21, 23, 45, 55, 66–67, Apollodorus 10, 44n39, 76–79, 81n8, 90,
73–74, 81n8, 82, 86, 89–90, 92, 95, 120, 93–96, 98–99, 124, 166
125–26 Aristodemus 10, 67, 68, 77–79, 93–96,
appeal to 67n49, 106, 126, 139, 142 98–99, 109n11, 124, 166
reaction of 67–68, 82, 90–91, 126 Aristophanes 95–96
See also setting Aristoteles 7, 40, 42, 117, 119
Author 99, 111, 115, 124, 162 Athenian Stranger 62, 74, 120, 132, 159
and the divine craftsman 16, 156, 159, Callias 29–30, 69n53, 79–81, 83, 128, 129
162, 166 Callicles 48, 66, 68, 107, 127, 129–31,
and the narrator 2, 15–16, 24, 148, 164–67 138–42, 146n54
Cebes 90–91, 128, 130, 142–46
Bal, M. 17n45, 17n46, 125n3 Cephalus of Clazomenae 7, 21, 27, 38–40,
Bifocality 53–56, 64, 109, 112, 121 43, 77–78, 98
See also dialogues; focalization; Cephalus of Syracuse 34–35, 61n33, 130
perspective Chaerephon 27, 48, 66–68, 74–75, 106,
Blondell, R. 14n41, 14n42, 36, 70n59, 72n70, 129, 130, 138–39, 142, 166
73n71, 99n56, 100n59, 108, 119n33, Charmides 27–30, 32–35, 81, 100, 110, 114,
120n35, 130n17 127, 130, 131
Bloom, A. 12n33, 14n40, 138n35 Cleinias the Cretan 62
Blushing 33, 35, 37, 82, 86, 127, 136 Cleinias the son of Axiochus 30, 85–87,
Booth, W.C. 167 130
Brandwood, L. 8n19, 116n27, 117n28, Cleitophon 35, 131
118n32, 121 Cratylus 49, 58–59, 130, 133n27
Brooks, P. 24, 76, 79n4, 92n43 Critias (1) 27, 29, 32–33, 60, 82, 107, 127,
130, 131, 141n43
Capuccino, C. 14n42, 99n55, 102n60, 149n1 Critias (2) 32n13, 71–73, 93n45, 132, 153
Change of interlocutor 125, 129–32, 133, Crito 21, 47, 49, 54, 58, 76, 79, 84–88, 90,
135–37, 141, 144–47 101, 114, 143, 166
See also plot-building tools Ctesippus 27, 31, 86–87, 130, 135
Characters 7, 28, 32–33, 40, 42, 45, 48, 52, Dionysodorus 21, 67, 76, 79, 85–87, 126,
53, 59, 61, 74, 78, 81, 95–96, 102, 106–107, 127, 130
119, 122, 124–25, 127–29, 130–32, 138, Echecrates 76, 79, 88–89, 91–92, 101, 114,
142, 167 142, 145–46, 163, 166
Adeimantus 3, 27, 35–36, 38–40, 44, 70, Eleatic Stranger 70–73, 108, 119, 127,
77, 108, 128, 130 132–34, 152, 158
Agathon 10, 44n39, 67, 76–77, 79, 93–97, Eudicus 47, 57, 63–64, 66, 74–75, 106, 109,
124, 158 166n43
184 General Index

Characters (cont.) Thrasymachus 27–28, 35, 37, 57, 94n48,

Euclides 5, 41, 51n13, 77, 79, 97–100, 124, 99, 107–109, 127–28, 130–31
162, 163, 166 Timaeus 71–73, 119, 132
Euthydemus 21, 67, 76, 79, 85–87, 115, Zeno 7, 27, 38–40, 42–44, 96n49, 109
126, 130 See also narrator’s text; Socrates
Euthyphro 47, 49, 53, 56, 58, 127
Glaucon 27, 34–36, 38–39, 44, 70, 77, Cicero, Ciceronian 110, 162–63
93–94, 108, 128–30 Clay, D. 2n5, 9n21, 10n24, 12n34, 14n42, 50n9,
Gorgias 48, 66–67, 106, 110, 130, 138–42, 128n12, 140n38
147 Closures 28, 49, 52, 79, 87, 92
Hermocrates 71, 73 Clustering 131, 135, 138, 146–47, 167
Hermogenes 29–30, 49, 58–59, 90, 130, See also plot-building tools
133n27 Cohn, D. 2n6, 13n37, 22n63, 23, 56n21
Hippias 47, 57–58, 63–64, 82, 107, 127 Collins, J.H. 13–14n39, 22n63, 78n2, 85n20,
Hippocrates 78–79, 82, 84, 129 87n26, 87n27, 92n42, 165n41
Hippothales 27, 31–33, 135, 137 Comedy 9, 30n8, 66n46, 87n26, 114
Ion 47, 56–58, 107 Cornford, F.M. 71n67, 138n35, 161n30
Laches 48, 64–65, 110
Lysimachus 48, 64–66, 72, 74, 109n9, 166 Danzig, G. 29n6, 29n8, 32n13, 58n24, 80n7
Lysis 27–28, 30, 32–33, 35, 110, 127, 130, Davidson, D. 69n54, 111, 131n20, 154n11, 156
131, 135–37 de Jong, I. 6, 8n20, 13n38, 17n46, 20–21, 22,
Menexenus 27–28, 30, 33, 47, 54–55, 130, 34n20, 48n3, 113n21, 167n45
131, 135–37 Diachrony See relative chronology
Meno 59–60, 106, 110, 127–28, 131n18 Dialogues, Platonic 47, 70, 108n5, 110, 125,
Nicias 48, 64–65 127, 129, 147, 156, 165–66
Parmenides 7, 27, 38–44, 96n49, 108–109, as literature 2n6, 9–15, 21, 24, 30, 82,
116–17, 119, 132–33, 158 105, 112–13, 118, 120–23, 129n13, 129n13,
Phaedo 68, 76, 79, 88–92, 114, 124, 130, 138n35, 151n6
142, 145–46, 166 biography-oriented 44, 96–97, 100, 102,
Phaedrus 48, 49, 61, 106, 126–27, 129, 145
153, 157 dramatic 2, 5–8, 14–16, 45, 47–51, 52–53,
Philebus 69, 119, 131, 155 55–56, 58, 62–63, 68, 72, 74–76, 80, 82,
Polemarchus 27, 29–30, 34–36, 128, 129, 97–98, 100–101, 105–107, 116, 118, 121–23,
130, 131 135, 165, 167
Polus 48, 66–67, 106, 130, 138–40, 142 elenctic 11, 135
Prodicus 82–83 “invention” of 28n3, 30, 49–50, 76,
Protagoras 60, 76, 79–84, 107, 115, 101, 121
127, 128 late, later 52n16, 70–74, 75, 116–17, 118–21,
Protarchus 68–70, 74, 119–20, 155 133–35, 150, 157
Pythodorus 7–8, 21, 27, 38–44, 59, 72, 74, mixed 2, 4, 13–15, 36, 50, 76–80, 85, 88,
75, 77–78, 95–96, 98, 109n11, 166 93, 97, 101–102, 105–107, 112, 113, 121–22,
Simmias 90–91, 128, 130, 142–46 135, 142, 165, 167
Socrates the Younger 71, 73, 119 narrated 2, 4–8, 13–16, 27–31, 39,
Terpsion 5, 77, 97–101, 166 44–45, 48, 49–51, 53, 55–56,
Theaetetus 5, 51–52, 70, 72–73, 77, 79, 66, 72, 74–75, 78, 80, 82, 100n59,
97–98, 100, 102, 110, 116, 120n35, 130, 101, 105–107, 112, 118, 121–22, 135,
158, 161 138, 166
Theodorus 5, 51–52, 70, 72–73, 79, 97, spurious 10, 27n1, 29–30, 35n22, 47, 64,
100, 130 66n46
General Index 185

transitional 116–18, 120–21 narrator 5, 18n49

visual aspects of 117, 120–21, 151, 154–55, reality 76
162–63 universe 11, 16, 18, 148, 164–66
See also multi-level; Socratic dialogue world 22, 113, 164n35
Diegesis, diēgēsis 1–6, 17n45, 149, 164 Fludernik, M. 20–22, 24n69, 42, 86n24,
“diegesis through mimesis” 1, 3–5, 77, 113n21
106, 162, 164 Focalization 16, 19, 39, 61, 63, 68, 80, 93, 96,
Diegetic See intradiegetic 107, 109–10, 118–19, 125n3, 167
Digressions 41, 52, 132–34, 145n52 by characters 42, 55–56, 60n30, 61, 64,
Homeric 147 106
in the Phaedo 144–47 by the narrator 15, 19–20, 31, 35, 40–42,
in the Republic 36, 128 51–52, 60–62, 66, 112, 148
in the Seventh Letter 154n12 external 43, 45, 59, 106
in the Theaetetus 11–12, 130, 145n52 gradual 20–21, 28, 30, 32–33, 45, 38,
Diogenes Laertius 29n5, 30, 49–50 51, 53, 56, 72, 74, 81, 85, 88, 98, 100, 107,
Dodds, E.R. 48n2, 66n47, 67n48, 68n51, 141 142
Drama 1, 14, 24n69, 55, 64, 77 markers of 38, 44–45, 50, 56, 62, 74–75,
See also comedy; tragedy 85, 106–107, 118, 122, 125, 167
Dramatic form 7, 8n20, 13–14, 39, 41, 50, reciprocal 53–56
76–77, 80n6, 94, 97, 99, 105, 117 unilateral 57–58, 108
privileging of 102, 122, 162–63 See also bifocality; perception; zooming-in
See also narrated form Frame, frame story 13, 16, 21–22, 35n21,
43–44, 50, 76, 78–81, 83, 86–89, 92–94,
Edmiston, W.F. 19–20 96–97, 102, 112–13, 116, 131
Emotions 33, 35–36, 38, 42–43, 45, 51–52, 60, narrative 39, 41, 43, 76–77, 112, 116
74, 82, 89–90, 95, 107–108, 119, 125, 129, dramatic 12n35, 43, 76–85, 87–88, 92,
131, 137, 147, 167 97n52, 99–101, 105, 112–13, 115–16,
See also blushing; anger; laughter 121–22, 142, 167
Emplotment 11, 23–24, 125, 138–39, 142, See also frame narrator(s); narrative levels
147–48 Frame-breaking See metalepsis
See also plot-building tools Frame narrator(s) 21–22, 23, 39–41, 43–44,
Erler, M. 14n42, 71n67, 84, 116n27 80, 85, 88, 90, 93–94, 96, 98
Euripides 28 See also narratorial authority
Antiope 140 Frede, D. 49, 69–70, 131n20
Medea 56 Frede, M. 11, 71n66, 110n16, 111–12
Telephus 66n47 Friedländer, P. 14n40, 47n1, 55n19, 84n19,
Extradiegetic 16, 76–78 87n25, 126n4, 149n3
level 22, 37, 77–78, 101, 105, 114
narratee 18, 77, 114 Genette, G. 2, 6, 13, 17–23, 41, 43n38, 68, 78,
narrative act 34, 77, 101, 105, 114 100, 110, 113n21
narrator 17–18, 22, 39, 78, 81, 114 Griswold, C.L. 12n36, 14n42, 166n42
Guthrie, W.K.C. 8n19, 14n40, 55n19, 71n67,
Ferrari, G.R.F. 4n10, 12, 13n39, 33n16, 36n24, 82, 83n16, 84n17, 116n27, 128n9, 130
38, 87n27, 113n21, 127n7, 127n8, 128, 148
Fiction 2n4, 2n6, 10, 13, 22–23, 34n20, 44, Halliwell, S. 1n2, 2n3–4, 14n39, 28n2, 99n55,
148, 164n35, 167 149n3, 150n4, 151n6, 156n15
Fictional 115 Halperin, D. 23n68, 78n2, 94n47, 96, 97n52,
narratee 18, 78, 93 99n56
narrative act 83, 164–65 Heterodiegetic See narrator
186 General Index

Homer 1, 4, 9, 20–22, 72, 147, 167 in the Phaedo 89–92, 145, 147
Iliad 3, 12, 130n17, 147, 167 in the Protagoras 82–83
Odyssey 20, 37, 82, 113 in the Republic 37, 95
Homodiegetic See narrator in the Symposium 96
Hunter, R. 2n3, 23n68, 30n9, 94n47, 96n50, initiated by the narratee 92
115 ontological 23, 83n15, 86, 92, 113, 132
rhetorical 23, 34, 37, 83, 86
Ignorance 59, 73, 95 See also narrative levels
initial 21, 65, 85, 88 Metanarrative comments 36n26, 59n27, 122,
See also reader; zooming-in 125, 132–35, 138–40, 142, 145, 147
Intertextual connections 29–30, 44, 62, strategically placed 133, 135
71, 73 See also plot-building tools
Intradiegetic 16, 19, 77–78 Michelini, A.N. 14n42, 129n13
domain 34, 78, 114 Mime 2n4, 9n21, 49–50, 55, 121
level(s) 6, 17, 19, 21–22, 31, 37, 39–41, 43, Mimesis, mimēsis 1–6, 9–10, 49–50n7, 77,
77–78, 81, 85, 88–89, 94–95, 98, 100–101, 119, 149–50, 153n7, 156n15, 158, 159–62,
105, 114 164, 166
metalepsis 114, 165n41 See also diegesis; paidia; painting
narratee 18, 101 Morgan, K. 2n3, 3n7, 6n15, 8n20, 13, 33n16,
narrative act 39 51n13, 84n17, 92n42, 99n55, 107, 116n27
narrator 17–19, 78, 101, 113n21 Multi-level 15, 43–44, 106, 115, 122
See also pseudo-diegetic dialogues 12n35, 23, 31, 44, 77, 97, 114–16,
Irwin, T.H. 108n7, 117n28, 118n32 121–22, 131, 163, 165n41
Iser, W. 11n27 narratives 18, 21, 80, 105–106, 112–13,
115–16, 121
Jowett, B. 4n8, 58n25, 119, 134 See also narrative levels
Murley, C. 12n35, 14n43, 33n17, 35n23,
Kahn, Ch. 14n42, 28n4, 29n6, 64n40, 65, 89n36, 127n8
71n66, 87n25, 87n26, 88n30, 97n53,
117–18, 137 Nails, D. 60n29, 62n35, 93n46, and passim
Kaklamanou, E. 14n39, 55, 57–58, 99n54, Narrated form 2, 8, 43, 76, 86, 105
99n56 abandonment of 8n19, 43, 106, 116, 118,
Kosman, L.A. 2, 10n26 121–23
dissatisfaction with 120
Laborderie, J. 10n24, 14n40, 14n42, 50n9, transformed into dramatic form 6–8, 13,
129n13, 148, 158n25 39, 97, 99, 162–63
Laughter 33, 35–37, 67n49, 82, 86–87, 90, Narratee 17–18, 34, 39, 76–78, 80–82, 92, 97,
95–96, 126, 127, 139, 144 101, 114–15, 165
personalized 78, 81, 86–88, 91, 93–94,
Macro-focalization See focalization by the 101–102, 146
narrator See also intradiegetic; extradiegetic;
McCabe, M.M. 14n42, 71n67, 115n26, fictional
142n48 Narrative act 17, 18–19, 28, 34, 37, 39, 76–77,
Metalepsis 13, 15, 16, 22–23, 33n16, 80–81, 101, 83, 114, 163–65
105, 113–14, 167 suppressed 97, 101, 105, 167
and parabasis 87n26, 114 See also extradiegetic; intradiegetic;
as a distancing strategy 43, 96, 115 fictional
in the Charmides 34 Narrative levels 6–7, 16–19, 21–22, 33n18, 77,
in the Euthyphro 86–87 82–83, 86, 90–91, 101–102, 105, 113,
in the Parmenides 43 131–32
General Index 187

and metalepsis 22–23, 113–14 Narrator’s text 15–16, 34, 60, 75, 122, 125, 129,
economizing on 6, 78, 101 132–39, 142, 145, 147, 167
explicit 6, 105, 116 and character speech 132, 136–37
multiple 13, 15, 43, 112 strategically placed 36, 45, 49, 67, 75, 83,
pseudo-extradiegetic 78, 101, 114n24 90–91, 45, 124–27, 133
suppressed 43, 105 See also plot-building tools
See also extradiegetic; frame; intradiegetic; Niederhoff, B. 19n54, 20n58
multi-level Nightingale, A.W. 14n41, 14n42, 140n38,
Narrative situation See narrative act 162n31, 164n37
Narratology, narrative theory 13–14, 16, 19, Novel, modern 1, 12n36, 21, 34
21–24, 44, 58, 114 individual works 12, 18, 21n60, 34, 38n32,
Narrator 4–5, 7, 14, 24, 33, 38–39, 43, 52–55, 41, 68, 79n4, 89n36, 92n43, 114
59–60, 74–76, 80–81, 89, 93, 96–97, 99, Nünning, A. 20, 24n69
108, 115, 119, 123, 124, 129, 130, 147–49,
162, 165–67 Openings 51, 54, 56, 60–61, 66, 72–75, 84, 94,
and omniscience 19, 33n16, 43, 90, 95 100, 118, 122, 125, 129, 135, 138, 142
as editor 37–38, 45, 94, 99, 107, 124, 148 in the dramatic dialogues 47–49
asks questions 21, 33, 51, 53, 56–57, 62, in the mixed dialogues 76–79, 81–82
88, 98 in the narrated dialogues 27–35
as spectator 73–74, 120, 163, 165
autodiegetic See narrator-hero Paidia 4, 41, 133, 157–59, 162, 166
explicit 8, 15–16, 27, 51–52, 62, 67, 74–76, See also mimesis; writing
81, 85, 88–89, 94–95, 106, 118, 123, 125, Painting 1n1, 149–54, 160–65
163–64, 166–67 See also dialogues; sculpture; writing
first-person 15, 18, 19–20, 27, 31, 35, Pavlou, M. 14n39, 55, 57–58, 99n54, 99n56
37–39, 42, 44–45, 58, 71, 75, 80n7, 82, 112 Perception 31, 35, 38, 42, 50, 54, 57–58, 61,
heterodiegetic 17, 39, 44, 97 65, 69, 74, 81, 85–86, 89, 95, 107–109, 119,
homodiegetic 17–19, 20, 31, 41, 44, 97, 124, 129
113n21 and knowledge 15, 45, 59, 119
implicit 7–8, 15–16, 47, 50, 52–53, 56, focus of 15–16, 33, 35, 38, 40, 45, 51–53,
58–60, 62–63, 68, 70–72, 74–78, 80–81, 57–59, 61, 64, 65, 68–70, 72, 75, 106–108,
85, 88, 93–94, 97–98, 100, 106–107, 120, 119, 121, 123, 165
123, 125, 163–67 limitations of 19–20, 59, 72, 74, 89, 95, 165
suppressed 6–8, 15, 48n3, 50, 52, 97–98, See also focalization; ignorance
106 Perspective(s), narrative 16, 19n56, 20, 91,
See also extradiegetic; frame narrator(s); 113, 148, 163–67
intradiegetic; narrator-hero; narratorial complex 101
authority; narrator-observer; narrator’s contrasting 92, 115–16
text; Socrates double 84
Narrator-hero 17, 31, 34, 41, 44, 58, 63, 74–75, multifocal 88
81, 85, 100, 107–10 overarching 56, 64, 102
Narratorial authority 15–16, 41, 44–45, single 56, 109, 112, 165
54–55, 58, 60, 62, 85, 89, 95, 99, 106 See also bifocality; focalization
limits of 124, 147–48, 166 Pfister, M. 56n21, 64
of frame narrators 39–41, 44–45 Pier, J. 6n14, 17, 22–24, 34n20, 132n21
signals of 16, 38, 45, 50, 52, 72, 74–75, Plato, transmitted works of
80–82, 107, 123, 125 Alcibiades I 47
Narrator-observer 17, 41, 44–46, 63, 66, 68, Alcibiades ii 10, 47
70, 72, 74–75, 89–90, 95, 105, 108–10, Apology 29n8, 66n46, 96n51
117, 120 Axiochus 27n1, 30, 86n21
188 General Index

Plato, transmitted works of (cont.) Minos 47

Charmides 2, 11n29, 22n65, 23, 27–28, Parmenides 2, 7–8, 13, 21, 23, 27–28, 31,
30–31, 31–34, 34–35, 37–39, 41, 44, 47n1, 35n21, 38–44, 59, 63, 72, 74, 77, 93n46,
51, 60, 62, 65, 66n46, 71n64, 74, 82, 86, 94–99, 101–102, 105, 108–109, 112, 115,
100, 107, 112–14, 121, 126, 127, 130, 131 116–17, 118, 119, 122, 126, 132–33, 158,
Cleitophon 35n22, 47, 108 163, 166
Cratylus 49, 54n17, 58–59, 60, 62, 72, Phaedo 2, 12–14, 16, 21n61, 23, 27n1, 33n16,
74–75, 90n38, 95, 107n2, 126, 130, 133 37, 41, 68, 76, 79–80, 83, 87, 88–92, 94,
Critias 7, 32n13, 48, 70–74, 75, 108n8, 116, 101–102, 105, 108n8, 109, 112–17, 120–21,
118, 120, 122, 132, 150, 153, 162, 163, 165 124, 126–28, 130, 131, 135, 138, 142–46,
Crito 15, 47, 49, 53, 54, 55–58, 64, 74, 147, 163, 165n41
90n39, 105, 115, 121, 126 Phaedrus 4n9, 11–12, 35n22, 48, 49, 54n17,
Eryxias 27n1, 29, 130 58, 60–61, 62, 74–75, 106, 107n2, 110–11,
Euthydemus 2, 11, 14n39, 21, 23, 27n1, 117–18, 119, 122, 126–27, 128–29, 134,
31n12, 36–37, 54n17, 67, 76, 78–80, 83, 137n32, 138, 150, 153–55, 157, 162, 165
84–88, 89–92, 94, 101–102, 105, 107n2, Philebus 49, 63, 68–70, 71, 74–75, 108n8,
111–15, 121, 126, 127, 130, 131, 165n41 116–19, 122, 125–27, 131, 133–34, 150, 155,
Euthyphro 11n29, 15, 47, 49, 53–54, 54–58, 162
64, 74, 105, 115, 121, 127, 132 Protagoras 2, 11, 21n61, 23, 27n1, 29n8,
Gorgias 11–12, 16, 48, 49, 63, 66–68, 32n13, 36–37, 53n15, 54, 60, 67, 69n53,
74–75, 106, 108n8, 109, 120, 122, 126, 127, 71n64, 76, 78–80, 80–84, 86, 88–89,
129, 130, 131, 135, 138–42, 146, 147 92, 94, 101–102, 105, 107, 112, 113, 115, 121,
Halcyon 66n46 126–29, 139, 141n43
Hipparchus 47 Republic 1–5, 6–8, 12–16, 22–23, 27–31,
Hippias Major 11n29, 47, 49, 53, 57–58, 33n16, 34–38, 39, 41, 44–45, 50, 57, 62,
62, 63–64, 74–75, 105, 107n2, 111, 121, 127 70, 71, 74, 87, 93–95, 99, 106–10, 113–14,
Hippias Minor 47, 49, 63–64, 66, 74–75, 117–19, 121, 122, 124, 126–31, 134, 138,
106, 108–109, 121–22, 166 146, 148, 149–50, 151–52, 154–56,
Ion 11n29, 47, 49, 53, 56–57, 57–58, 62, 159–62, 165
74–75, 107, 121 Rival Lovers 27n1, 30
Laches 11n29, 40, 48, 49, 63, 64–66, 68, Sophist 7, 11–12, 48, 49, 51n11, 51n12,
72, 74–75, 95, 96–97, 108n8, 109, 120, 70–74, 75, 95, 108, 111, 116–20, 122, 127,
122, 166 132–33, 138n35, 150n4, 158, 159, 160–62,
Laws 48–49, 58, 62–63, 72, 74, 107n2, 165
116, 118, 120, 126, 132, 134, 150, 152–53, Statesman 7, 11–12, 48, 49, 51n11, 51n12,
158–59, 162 70–74, 75, 108, 116–20, 122, 127, 132, 134,
Letter 7 154n12 150, 152, 154–55, 158, 162, 165
Letter 11 71n63 Symposium 2, 7n18, 10–13, 21n61, 23, 27n1,
Lysis 2, 11n29, 16, 27–28, 30–31, 31–34, 35n21, 38, 41, 43–44, 52n15, 54n17, 67,
34–35, 38–41, 44, 47n1, 48n2, 51, 54n18, 68, 76–80, 84, 90, 92, 93–97, 98–102,
62, 65, 74, 100, 107n2, 111, 112, 121, 126, 127, 105, 108–10, 112, 113, 115–17, 120–21, 124,
130, 131, 135–38, 142, 147 126, 127, 131, 137, 158, 163
Menexenus 15, 33n15, 47, 49, 53, 54–55, Theaetetus 2, 5–8, 11–15, 21n61, 38, 40–41,
55–58, 64, 74, 121 44, 48–51, 51–52, 53–56, 58, 62, 65,
Meno 11, 49, 58, 59–60, 62, 64n42, 74–75, 70–71, 73–75, 77–80, 84, 92, 97–100,
105–106, 107n2, 121, 125, 127–28, 131, 133, 101–102, 105–108, 111, 112, 115–118, 120n35,
144 122, 124, 126, 127, 130, 134, 162, 163
General Index 189

Theages 47 Schmid, W. 17n46, 18n49, 19, 75n72

Timaeus 7, 32n13, 48, 49, 70–74, 75, Schofield, M. 108n4, 110, 128n10
93n45, 108n8, 116, 118, 120, 122, 132, 150, Schur, D. 2n6, 10n26, 12n36, 14n42, 28n2,
152–53, 155–56, 158, 162, 165 110n14, 128n10, 129n13, 145n52
Play, playful See paidia Scope of narrative 44, 99, 124
Plot 10, 16, 23–24, 31n10, 36, 45–46, 83, 88, limited by the narrator’s experience 19,
92, 113, 117, 124–25, 128, 133, 138, 147, 45, 59, 74, 82, 95, 107
155n14, 162 See also focalization; narrator; perception
and argument 10–12, 124–25, 166 Sculpture 1n1, 134n28, 150n4, 161, 165
and story 10, 124–25, 147 See also mimesis; painting
building of 24, 118 Sedley, D. 8n19, 72n69, 88n30, 88n31, 90n39,
delivers a message 12, 166 112n20, 116n27, 130n16, 130n17, 145n52
See also digressions; plot-building tools; Sermamoglou-Soulmaidi, G. 85n20, 87n28
structural breaks; turning points Setting 10, 29, 31–32, 34, 38, 45, 47–49,
Plot-building tools 125, 129, 132, 135, 138–39, 53–54, 59–61, 63, 66, 74–75, 81, 85, 89,
141, 146–47 106–107, 124, 125–27, 129, 135, 146
See also change of interlocutor; clustering; See also audience; narrator’s text; openings
emplotment; metanarrative Socrates 3–8, 10–11, 21, 27–30, 38–40, 42,
comments; narrator’s text 47–49, 54–63, 76–77, 88, 102, 106,
Pseudo-diegetic narrative 6, 13, 79, 100 110–14, 116–17, 119–20, 163, 166
See also Genette; narrative levels as a character 33, 36–38, 44, 58–59,
63–70, 79, 84–85, 87, 96–97, 108–109,
Reader(s), “we” 10–12, 16–18, 24n70, 30–31, 126–29, 131–48, 152, 155
34, 37, 41, 43, 52, 55, 64, 69, 73, 78, 84, as the narrator 14, 27, 31–39, 41, 51–52,
85–89, 92, 96, 110, 115, 117, 119, 125, 129, 54, 58–60, 66–67, 70–75, 78–79, 81–83,
132, 148, 164, 166 85–86, 92, 94–95, 97–100, 107–108, 110,
absorbs information with the ­narrator 114, 117, 120, 124, 132, 137, 148, 163,
21, 28, 33, 45, 47, 51, 53–62, 66, 81–82, 165–66
88–89, 93–95, 98 being focalized 42–44, 53, 64, 65, 68,
implied 18, 34, 39, 78, 101 89–97, 109, 117
See also apostrophic addresses; privileged 56, 110
­focalization; ignorance; zooming-in Socratic dialogue 2n4, 28, 50n10, 76, 88n30,
Relative chronology 8n19, 15, 52n16, 117–18, 97n53, 99n55, 121
122, 167 as a literary genre 9–11, 13–14, 125, 164
Reported speech, two-tier system of 7n18, See also Aeschines of Sphettus;
23, 43, 96 Xenophon
See also metalepsis Sophron See mime
Reversal See turning points Structural breaks 67, 83, 90
Ricoeur, P. 1–2, 24 See also narrator’s text; plot
Rimmon-Kenan, S. 19n56, 164n36
Rowe, C.J. 14n42, 70n60, 92n42, 110n15, Tarrant H. 28n3, 48n3, 50n10, 94n47, 99n55,
112n20, 145n51, 153n8, 157n16 100n59
Rutherford, R.B. 8, 9n21, 14n42, Taylor, A.E. 14n40, 91n40, 140n38, 142
70n60, 116n27, 118n30, 119n33, Thesleff, H. 8n19, 14n42, 28n2, 48n2,
120, 122n40, 130n17, 138, 140n38, 71n67, 80n5, 87n25, 91n40, 100n59,
145n52 102n60, 116n27, 117n28, 118n32, 119n33,
Ryan, M.-L. 22n63, 23, 86, 132n21 121n38
190 General Index

Tragedy 3, 74, 126, 145n51, 155n14 Xenophon 27n1, 29–30, 32n13, 58n26, 59n28,
See also Euripides 66n46, 80n7, 90
Turning points, in the plot 33–34, 36, 41, 75, Symposium 29, 32n13, 41, 86n21
91, 128–29, 133, 137n32, 138, 145n51, 147
See also narrator’s text; plot Zooming-in 20–21, 28–30, 33, 35, 38, 40, 45,
47–48, 51, 53–54, 59–60, 72, 74–75, 81,
Wilamowitz-Moellendorff, U. von 29n6, 85, 98, 107, 147, 167
50n10 See also focalization; ignorance; narrator;
Wolf, W. 79n4, 113n21, 114 reader
Writing
and painting 153–54, 162
as paidia 4n9, 12, 119, 126–27, 157–59
in the Theaetetus 5, 99