By Being, It Is

This page has been intentionally left blank.

By Being, It Is
THE THESIS OF PARMENIDES

Nestor-Luis Cordero

PARMENIDES
PUBLISHING

PARMENIDES PUBLISHING
Las Vegas 89109
2004 by Parmenides Publishing
All rights reserved
Published 2004
Printed in the United States of America
ISBN: 1-930972-03-2

Publishers Cataloging-in-Publication Data

Cordero, Nestor-Luis.
By being, it is : the thesis of Parmenides /
Nestor-Luis Cordero.
p. ; cm.
Includes bibliographical references and index.
ISBN: 1-930972-03-2
1. Parmenides. 2. Ontology. 3. Eleatics. II. Title.
B187.05 C67 2004
182/.3
English translation by Dinah Livingstone
for Translate-A-Book, Oxford, England

1-888-PARMENIDES
www.parmenides.com

Contents

Prologue
Acknowledgments

ix
xiii

Chapter I: Introduction to Parmenides

(a) The Region
(b) Chronology
(c) Life
(d) Works
(e) The Poem
(1) The Reconstruction
(2) The Form
(3) The Content

3
3
5
8
11
12
12
14
15

Chapter II: Prolegomena to Parmenides Thesis

(a) Parmenides theora
(b) The Allegorical Presentation of the Content
of the Poem
(c) You Must Inquire About Everything (1.28)

19
19

Chapter III: Parmenides Thesis and Its Negation

(a) The Alternative in Fragment 2
(b) The Only Two Ways of Leading Thought
(c) Lines 3 and 5 of Fragment 2
(1) Estin on Its Own and Its Negation
(2) The Modal Complement of estin on Its Own
and Its Negation

37
37
39
42
44

Chapter IV: The Meaning of Parmenides Thesis

(and of Its Negation)
(a) The Grammar of To Be
(b) The Meaning of Being and Returning to the
Question of the Subject of estin in 2.3a

21
30

54

59
59
60

vi

Contents

(c) The Absolutization of the Fact of Being,

the Negation of the Thesis, and the Ways
of Investigation
(d) The Opposition Between the Thesis
and Its Negation
(e) Structural Difference Between Statement
and Negation
(f) Why Is the Negation of the Thesis Impossible?
Chapter V: Parmenides Thesis, Thinking,
and Speaking
(a) Thinking Is Expressed Thanks to Being
(b) It Has to be Said and Thought That
That Which Is Being, Is
(c) Impossibility of Thinking and Saying That
Which Is Not Being
Chapter VI: Presentation of the Thesis and
Its Negation in Fragments 6 and 7
(a) 6.1b2a Reintroduces the First Way
of Investigation
(b) Relation Between 6.12 and Fragment 2
(c) Truth, Persuasion, and Deception
(d) The Exhortation to Proclaim That It Is Possible
to Be and That Nothing[ness] Does Not Exist
(e) Parmenides Does Not Recommend
Withdrawing from the Thesis Expounded
in 6.1b2a
(f) The Origin of the Notion of Withdrawing
as a Conjecture in 6.3
(g) Rejection of the Conjecture I Withdraw You
(h) The Thesis Expounded in Fragment 7
(i) A Possible Solution for the Gap in Line 6.3
(j) Discovering the Foundation of the Two Ways
in Fragment 6

64
69
72
79
83
84
90
92
97
98
101
103
105
108
112
116
117
119
122

Chapter VII: The Negation of the Thesis, Opinions,

and the Nonexistent Third Way
125
(a) The Senses and the Wandering Intellect Do Not
Distinguish Between Being and Not Being
129

Contents

(b) Logos as the Criterion by Which to Judge

the Critique of the Way Made by Men
(c) The Meaning of logos in Parmenides
(d) The Hypothetical Third Way
(e) Conrmation of the Existence of Only Two
Ways of Investigation
Chapter VIII: The Meaning of the Opinions
of Mortals
(a) Doxa Is Not Appearance
(b) The Object of Opinions
(c) Doxa and Names
(d) The Opinion-makers
(e) The Content of Opinions

vii

134
136
138
143
151
152
154
156
158
160

Chapter IX: The Foundation of the Thesis:

The Way of Truth
(a) The Only Way That Remains
(b) The semata of estin
(c) The Field in Which the semata Operate
(d) The First sema: That Which Is Being
Is Everlasting
(e) Immobility
(f) Homogeneity
(g) Oneness
(h) Truth

170
173
174
175
178

Epilogue

181

Appendix 1: Parmenides Poem

(a) Text
(b) Translation

185
185
190

Appendix 2: Note on the Transliteration

of the Greek Alphabet

197

Bibliography
List of Ancient Authors Cited
List of Modern Authors Cited

199
211
213

165
165
168
170

This page has been intentionally left blank.

Prologue

Historians of philosophy usually refer to Plato to conrm the importance

that Parmenides philosophy had acquired, even in his own time. They cite
not only the celebrated passage from Platos Sophist, in which the Eleatic
philosopher is described as the Athenians (obviously spiritual) father
(241d), but also the text of the Theaetetus, which calls him venerable and
fearsome (183e), according to the Homeric formula applied to the revered
Priam (Il. 3.172). Generally speaking, at this point, curiously, quotations
from Platos text peter out. But Plato continues to concern himself with
Parmenides, and in the following sentence we nd a true confession, proof
of the lucidity and sincerity with which the philosopher approaches his
ancestors thought: Parmenides, says Plato, seemed to me to have a power
that denotes a depth absolutely full of nobility. Even so, I am afraid we
may not understand his words, and I am even more afraid that what he
was thinking of when he said them goes quite beyond us (184a).
For us these words of Platos have always been an invitation, indeed,
an incitement, to take an interest in Parmenides philosophy. Less than a
century after his death, Plato is already confessing that he is afraid he cannot understand the meaning of the Eleateans philosophy, but that does not
prevent him recognizing its immense value or, especially, from criticizing
and even refuting it. This means that whatever the real meaning of Parmenides
ideas, they were taken by Plato in a certain way, and that is the Parmenides
whom Plato combats, or, if you prefer, revises and even improves. Today,
almost twenty-ve centuries later, we see that the Parmenideanism that
Plato criticizes is a combination of the Eleateans own ideas with Zenonian
and Melissian ingredients, and that this explosive mixture was very probably represented by Antisthenes at the time the Theaetetus and Sophist were
being written1 (cf. Epilogue). But all this is secondary: it is the image Plato
has of Parmenides that leads him to take an interest in him. And this is still
going on today. Other philosophers of antiquity (Aristotle, Plutarch, Sextus
Empiricus, Simplicius) offer us other aspects of Parmenides, and we might
even say they present us with other Parmenides. So did the numerous
doxographers, who often gave pride of place to a cosmological Parmenides.

Approximately 369367

B.C.

Prologue

This diversity of viewpoints has encouraged us to try to decipher those

words that were already enigmatic for Plato. Do we then aspire, from our
postmodernist stand, to grasp Parmenides thought from a more privileged
perspective than that from which his successors studied him? Not at all.
But we must not fall into the opposite extreme and blindly accept the classical authors viewpoints, without any critical spirit. Let us not forget that in
commenting on their predecessors thought they were not claiming to be
historians of philosophy. They were conversing with ideas, not real people. And these real people may have expressed themselves orally, but most
of themincluding Parmenideswrote texts. Almost miraculously, passages from these texts have come down to us.2 And in our case, there is
the assumption, or if you like, the prejudice that any interpretation of the
philosopher must be based on these. When ancient authors comment on
these passages, they must be listened to and respected as rsthand witnesses. This is the case with Plato when he cites and comments on the
current rst two lines of fragment 7 of Parmenides Poem in the Sophist
(237a, 258d), or when Sextus Empiricus transcribes nearly the whole of fragment 1 and presents his allegorical interpretation (Adv. Math. VII.111). On
the other hand, there are cases in which certain passages have not earned
the attention in antiquity that they merit for us today; in this case, we can
exercise our right to interpret them. This is the case with fragment 2, fragment 6, and the rst line of fragment 8, which today appear to expound
the nucleus of Parmenides thought, but which no one in antiquity commented on or cited (unless they did in texts now lost to us) for eleven
centuries, until the sixth century of our era.3
* * * * *
All these theoretical conditions (or prejudices, if you like) have made us
focus for years on the state of the Poems text. Any new interpretation of
Parmenides philosophy, or any criticism of previous interpretations, must
be based on a text that is as close as possible to the lost original. The titanic
task carried out over centuries by philologists and codicologists offered us
a rm starting point, but much still remained to be done. Passages of the
Poem remained inexplicably obscure. (For example, why does the Goddess
order withdrawal from a true way in line 6.3? How can it be said that
thought is expressed in being, as line 8.35 appears to say?) For this reason,
since my presence in Europe made it possible, I decided to check the manuscript tradition of citations (wrongly called fragments) of the Poem, in
order to propose a new version of it, puried of certain errors that had
2
3

In the case of Parmenides, cf. Chapter I (e).

In fact, the only citations of these texts occur in Proclus and in Simplicius.

Prologue

xi

accumulated over the centuries. A rst result of my search was presented

in 19714 as a doctoral thesis. Some years later, my book, Les deux chemins de
Parmenide,5 completed my work. New research on the manuscript sources
of the rst editions of the Poem, as well as a change of view in my assessment of the two ways, allow me to present this new version of Parmenides thesis today. In this work, I also take into account comments and
criticism that my previous studies on Parmenides have raised, and when
appropriate, (a) I defend myself, or (b) I accept and make certain corrections.
It is impossible to go into Parmenides philosophy without being bitten by the bug. I hope that readers of this book will feel the same.
Nestor-Luis Cordero
University of Rennes I
France

4
5

Paris IV, Sorbonne. Supervisor, Pierre-Maxime Schuhl.

Cordero, N. L., Les deux chemins de Parmenide (Paris/Brussels: Vrin/Ousia, 1984; second
edition, augmented and corrected, 1997).

This page has been intentionally left blank.

Acknowledgments
If Greek philosophy is still alive today it is due to two main factors. The
rst is the depth of Greek philosophers thought in questioning the foundation of reality, a depth that continues to be the base of all actual reection
today. That we are aware of this thought, however, and that we are able to
appreciate the Greek philosophers in all their magnicence, is due to quite
another, usually forgotten, factor, one to which I want to pay tribute in
this acknowledgment: the titanic work of those thousands of anonymous
intermediates who handed down the ancient texts. It is thanks to these
copyists and transcribers, true laborers of the intellect who were at rst
entirely unknown, that the texts we have today were able to survive. As
more experts joined the ranks of dedicated workers intent on perfecting the
quality of the texts, little by little the crafts of papyrology, codicology, and
philology were born. Our indebtedness and eternal gratitude to these preservers and transmitters of the ideas of the past are fundamental.
In our own time, the task of conserving old texts is close to completion
due to the promotion of ancient books and works dedicated to great philosophers of the past. A few publishers continue the tremendous task of the
old copyists. In this sense, the work of Parmenides Publishing is exemplary,
because it dedicates its efforts to defend Parmenides place as the true father of Philosophy. To be among its selected authors and to partake in this
essential promotion of Parmenidean philosophy makes me feel proud. I
sincerely thank them.

This page has been intentionally left blank.

By Being, It Is

This page has been intentionally left blank.

Chapter I: Introduction to Parmenides

(a) The Region
All historians of ancient Greece admit that in the middle of the eighth century B.C., an unprecedented phenomenon occurred within the Hellenic
world which, as we know, was made up of a group of autonomous cities,
gradually evolving toward what would form the basis of the democratic
polis a century later. This was the expansionwrongly named colonization6 toward the West.7 For various reasons, none of which had to do
with an idealistic thirst for adventureeither the growing shortage of
arable land (a phenomenon called stenokhora in Greek), or the need to seek
new markets for production surpluses, or simply for individual reasons
(invasions, natural disasters),8 large human groups left what were usually
the more prosperous cities in search of new horizons. This happened, above
all, in Corinth, Megara, Miletos, and Phocaea. Phocaea is the one that concerns us here.
According to Strabo (XIV, 1, 3 = 633), the Phocians, who came from
areas near Parnassus, in Phocis, settled in Ionia at a date that we can set
today at around the eleventh to tenth centuries B.C.9 They not only took
part in this later expansion toward the West but, according to Herodotus,
they were the rst among the Greeks to accomplish distant voyages. They
were the ones who discovered the Adriatic Gulf, Tyrrenia, Iberia, and Tartessos (Cadiz) (I.163). It is known that in about the year 600 B.C. they
founded Massalia (Marseilles) and that in 565 B.C. they also settled in Alalia
on the island of Corsica.10
While this settlement of the Phocians was taking place, a serious event
hastened the departure of a new wave of migrants: in 545 B.C., Harpagon,
6

7
8

9
10

Finley says that during this period of crisis a safety-valve was provided by the movement wrongly called colonization. (Finley, M. I., The Ancient Greeks [London: Viking
Press, 1963], 25).
A detailed study of the question can be found in Boardman, J., The Greeks Overseas (Baltimore: Penguin Books, 1964), especially 175231.
Claude Mosses chapter titled Lextension du monde grec a partir du VIIIe siecle examines the possible causes of this expansion (Mosse, C., La Gre`ce archaque dHome`re a` Eschyle
[Paris: Seuil, 1997], 7995).
Cf. Huxley, G. L., The Early Ionians (New York: Faber, 1966), 25.
Beraud, J., La colonisation grecque de lItalie meridionale et de la Sicile dans lAntiquite (Paris:
P.U.F., 1957), 267.

(a) The Region

one of Cyrus the Persians generals, invaded the mother city, Phocaea,
whose inhabitants mostly went into exile.11 Almost at the same time, an
alliance between the Etruscans and the Carthaginians attacked the Corsican
Phocians. The Greeks won, but they suffered such heavy losses that they
had to abandon the island. That is how the exiled Phocians from the east
(Alalia) and the west (Phocaea) joined together and disembarked at present-day Lucania, a few miles south of Naples. There they established the
settlement of Elea.12 A few years later, in the Phocian city in southern Italy,
Parmenides was born.
Without coming to any agreement, historians debate about a certain
detail, though it is not a very important one for us: in the place chosen to
found the new colony, was there already a local population, or did the
Phocians occupy an uninhabited area? The reader who is interested in this
question can refer to the documented works of E. Ciaceri,13 J. Beraud,14 G.
Vallet and F. Villard,15 M. Napoli,16 E. Lepore,17 and especially, J. P. Morel.18
Herodotus (I.167) says that, in fact, the Phocians took over the city in southern Italy that the Enotrians called Hyele. There must be some truth in
this viewpoint, because Strabo (VI, 1, 1 = 252) conrms that the later Greeks
gave the name of Elea to the place that the founders knew as Hyele, and
that this name was of pre-Greek origin.19 However that may be, we can
state that Parmenides was born and lived among a Phocian community,
that is, Ionian. Consequently, we can say at once that the arbitrary system
usually to be found in certain history of philosophy manuals (whose origin
goes back to Diogenes Laertius: I.13), which separates an Ionian school
from an Italic school, does not make sense. Not only was the founder of
the so-called Italic school, Pythagoras, born in Samos (an island from which
Ionia can be seen), but Eleas most important philosopher, Parmenides, was
of pure Ionian stock.
11
12
13
14
15
16
17
18

19

Cf. Huxley, Early Ionians, 117.

Cf. Pugliese Carratelli, G., Nascita di Velia, La parola del passato 25 (1970) 60.
Ciaceri, E., Storia della Magna Grecia (Milan: Albrighi, Segati & Co., 1927).
Beraud, La colonisation grecque.
Vallet, V. and Villard, F., Les Phoceens en Mediterranee Occidentale a` lepoque archaque et la fondation de Hyele`, La parola del passato 21 (1966) 16690.
Napoli, M., Civilita` della Magna Grecia (Rome: Eurodes, 1978) and La ricerca archeologica
di Velia, La parola del passato 21 (1966) 191226.
Lepore, E., Strutture della colonizzazione focea in Occident, La parola del passato 25 (1970)
1954.
Morel, J. P., Les Phoceens en Occident: certitudes et hypothe`ses, La parola del passato 21
(1966) 379420, and Sondages sur lacropole de Velia (Contribution aux premiers temps
de la cite), La parola del passato 25 (1970) 13145.
Cf. Capizzi, A., Introduzzione a Parmenide (Bari: Laterza, 1975), 16. According to Untersteiner (1958), 41, the root of the name is Tyrrenian.

Introduction to Parmenides

To complete this geographical sketch, we may add that Elea was a harbor city known as Velia by the Romans and in the Middle Ages called
Castellamare di Veglia or Della Bruca. Nowadays it has lost its access to
the sea because the coast has withdrawn by half a mile, and it is known as
Ascea.

(b) Chronology
All the ancient witnesses assert that Parmenides was born in Elea.20 However, nothing certain is known about his possible date of birth. As there
can be no doubt that Parmenides is an Eleatean, a post quem date applies:
Parmenides could not have been born before the foundation of Elea, an
event that, as we have seen, took place shortly after 545 B.C. From then on,
researchers have a free eld in which to propose all sorts of hypotheses.
Nevertheless, this freedom is not total: we may say that there are two possibilities to take into account, and as usually happens, there is a happy medium between the two. Hence we nd that three probable chronologies
have been proposed.
As a last resort, everything depends on the source chosen. In our case,
as we said, there are two different dates proposed by two different sources,
although in both cases, as we shall see, we have to resort to sophisticated
deductions. One of the sources is a historian; the other, a philosopher.
Which witness should we choose? The question is important because nearly
thirty years separate the dates proposed by these two sources. However,
we might say that the question could be resolved a priori: as it is a question
of establishing a date, that is, a historical fact, it would appear that the last
word should lie with the historian. In the present case, things are not that
simple, because the historian uses data that enable us to deduce a date, but
do not establish it for certain. The philosopher in question is Plato, whose
authority specialists usually reverently accept regarding everything he said,
including statements about historical events.
Let us begin with the historian. He is Diogenes Laertius, whose work,
Lives of Eminent Philosophers, is an inexhaustible source of facts. Nevertheless, we know that this work should be referenced with caution because its
reliability depends on the sources used by Diogenes himself, and when no
source is cited, then doubts arise. However, with his chronologies there is
a certain consensus among commentators to consider them fairly reliable,
because they depend on Apollodorus heavily documented Chronicles (sec20

Cf., among others, Diogenes Laertius, IX.21; Proclus, In Parm., I, 619, 4; Strabo, VI, I, 252.

(b) Chronology

ond century B.C.),21 which use the dates of the Olympiads as benchmarks.22
With respect to Parmenides, we read in Diogenes Laertius (who, in this
case, probably also takes the information from Apollodorus, although he
does not say so) that the philosopher reached his akme (that is, the height
of his activity, which usually coincides with his forties) during the sixtyninth Olympiad23 (IX.23). If this is true, Parmenides must have been born
between 544 and 541 B.C., that is, just about when the Phocians arrived in
Elea. His parents may have belonged to the party that founded the city.
Let us now look at what Plato has to say. As we know, the dialogue
Parmenides describes a visit to Athens by Zeno of Elea for the purpose of
making known orally (through conferences, as we would say today) the
contents of his book. Still according to Plato, Zeno arrived accompanied by
his master, Parmenides, and one of those present at the conference was
Socrates. Unexpectedly, the introduction of these characters gives us material of priceless value for the Parmenidean chronology, because for reasons
unknown to us, Plato is unstinting with details referring to the precise ages
of the protagonists: Zeno is approaching his forties; Parmenides, despite
his noble presence, has white hair and is fairly old, about sixty-ve. For his
part, Socrates is a mere stripling (Parm. 127bc).
In order to be able to deduce the date of Parmenides birth, we need to
know what year this philosophical encounter in Athens took place. Plato
only says that the meeting took place during the Pan-Athenian Festival,
and that at that time Socrates was very young. Historians have established
that this festival, celebrated every four years, took place in 454 B.C., 450 B.C.,
and 446 B.C. during Socrates youth. Hence, as Socrates was not only present
at the conference but appeared as a young philosopher already propounding the real existences of Forms or Ideas, the rst date must be discounted.
A fteen-year-old adolescent (Socrates was born, as we know, in 469 B.C.)
would have found it difcult to assume that role. We must also discount
the year 446 B.C., because a twenty-three-year-old is no longer a stripling
(sphodra neos, 127c). That leaves only 450 B.C., the year in which Socrates
would have been nineteen. And if Parmenides, as Plato states, was then
sixty-ve, he must have been born in the year 515 B.C., that is, almost thirty

21
22

23

This lost work has partly been reconstructed by Jacoby, F., Apollodoros Chronik, Eine Sammlung der Fragmente (Berlin: Weidmann, 1902).
On the importance of this work, cf. Untersteiner, M., in Problemi di lologia losoca, ed.
Sichirollo, L., and Venturi Ferriolo, M. (Milan: Cislapino-Goliardica, 1980), 244. Cf. also
Diels, H., Chronologische Untersuchungen uber Apollodors Chronik, Rheinisches Museum 31 (1876) 154.
Years 504501 B.C., because the rst Olympiad is thought to have taken place in 740737
B.C.

Introduction to Parmenides

years later than the date proposed by Diogenes Laertius and Apollodorus.
The difference is very important, especially when we want to position a
philosophers thought in relation to someone elses ideas, in this case, Heraclitus (as we shall see).
As usually happens, attempts to reconcile both dates were not slow in
coming, but most of them were based on modications of the original texts,
a mortal sin that any serious philologist must avoid. I mention, as a curiosity, that a desperate but ingenious solution was proposed in 1924 by Heinrich Gomperz.24 He retained Platos authority, but saw a contradiction between the description of Parmenides as someone very old (mala presbuten,
127b) and the fact that he was barely sixty-ve years old. His interpretation was that Plato meant that the philosopher appeared to be that age,
whereas he was really much older (he might have been eighty, Gomperz
supposed). Thence he proposed the year 530 B.C. as Parmenides date of
birth.25
Which testimony is the most reliable? Given the characteristics of the
Platonic text, I am inclined to opt for the Diogenes Laertius/Apollodorus
chronology. Let me say why. Plato is a philosopher, not a historian.26 His
interest in the rst part of Parmenides is to criticize certain aspects of his
theory of Forms. So imagine the scene: only a philosopher with great prestige, especially if he is a venerable person (as Parmenides had already
been described in the Theaetetus, 183e), would have the necessary authority
to admonish a stripling claiming to have already found a denitive truth,
as is the case with the character interpreted by Socrates. I say interpreted
because, although there may be doubts in other dialogues about the philosophical opinions expressed by Socrates (which might belong, according to
some scholars, to the historical Socrates), in the case of the Parmenides, Plato
puts into Socrates mouth a rigorous and orthodox presentation of his own
theory of Forms. This turns the great philosopher into an almost ctional
character, a sort of ventriloquist through whom Platos own voice speaks.
And just as he had to rejuvenate Socrates in order to attribute the difculties he found in defending his ideas to his inexperience, Plato had to resort
to the great Parmenides to make a criticism, which is in fact a self-criticism.
Everything indicates that Plato does not set real characters on his stage
but symbolic ones: the young philosopher, enthusiastic but dogmatic; the

24
25
26

Gomperz, H., Psychologische Beobachtungen an griechischen Philosophen, Imago 10

(1924).
Other possibilities are found in Bicknell, P. J., Dating the Eleatics, in For Service of Classical Studies, Essays in Honor of F. Letters, ed. Kelly, M. (Melbourne: Cheshire, 1966), 114.
On Plato as a historian, cf. Chiereghin, F., Implicazioni etiche della storiograa losoca di
Platone (Padua: Liviana, 1976), Chapter III.

(c) Life

old master, experienced and didactic. R. E. Allen admits no doubt in maintaining that the encounter is just a ction,27 which only conrms that lack
of condence already shown in antiquity about the reality of this fact.
Athenaeus had stated that its historicity was highly unlikely, and to back
this judgment he cited an epigram by Timon, who alluded to Platos ctions (or simulations: peplasmena: Deip., XI, 505f). For his part, Macrobius
quoted the case of the Protagoras, in which Plato presented two characters
who had already died of the plague some time before and, ironically, stated
that he did not claim to count his characters ages on his ngers (Saturn.
I, 1.5).
From the beginning of the dialogue, Plato does everything he can to
make the reader de-realize the story: the encounter is narrated by Cephalus, who heard it from Antiphon, who in his turn heard it from Pythodorus
(126a127b). After this series of Russian dolls (or Chinese boxes, as
Allen calls them),28 any similarity to real events, as certain movies declare,
is pure coincidence.29 Lastly, we should not forget that Plato is accustomed,
doubtless on purpose, to dropping into anachronisms. As M. Untersteiner30
remarks, in Timaeus 20d, Solon becomes younger by twenty and even by
thirty years. For all these reasons, I am inclined to accept the date proposed
by Diogenes Laertius/Apollodorus, which, incidentally, has an unexpected
secondary consequence. It puts an end to the sterile question of Parmenides supposed criticism of Heraclitus, because if Parmenides was born
between 544 and 541 B.C., he was practically contemporary with the
Chiaroscuro of Ephesus. Both philosophers reached their intellectual zenith
during the sixty-ninth Olympiad and, apparently, they were totally unknown to one another.

(c) Life
Little or nothing is known about Parmenides life, except for the name of
his father, Pyres. A few years ago, the discovery in Elea of a statue pedestal
inscribed Parmenides, son of Pyres, doctor philosopher encouraged the
idea of the existence of a school of medicine in the region, to which our
philosopher may have belonged or of which he may have even been the
27
28
29

30

In Plato, Parmenides, trans. Allen, R. E. (Oxford: Blackwell, 1983), 63.

Plato, Parmenides, 61.
Although Gomez-Lobo (1985), 19, believes, as I do, that the encounter is imaginary, he
thinks that Plato wanted to convince the reader of its reality, hence his precision in giving
the ages.
Untersteiner (1958), 19.

Introduction to Parmenides

head. Much has been written about this.31 The only certainty we can draw
for our knowledge of Parmenides is the conrmation of his social importance, which was already substantiated in antiquity by authors like Plutarch
(Adv. Col. 32, p. 1126A) and Strabo (VI, 1), who related that the philosopher
compiled the laws of Elea and that even many years later the citizens of that
city still swore obedience to those laws. Other details about Parmenides life
are the fruit of interpreters imagination. One interpreter, Karl Popper, goes
beyond what should be expected from a sensible researcher; referring to
Parmenidean terminology on light, he does not hesitate to state that our
philosopher was brought up by and with a beloved blind sister, three
years older than himself.32 The classical authors do not fall into these excesses of imagination. Diogenes Laertius, for example, bases himself on Sotion and says that Parmenides belonged to an illustrious and wealthy family (IX.21), which enabled him to put up a mausoleum in memory of his
friend Ameinias, a Pythagorean philosopher. This detail brings us to a secondary question, much debated not only in antiquity but also in our own
days: the problem of Parmenides masters.
Diogenes Laertius is very explicit in this respect: It was by Ameinias
and not by Xenophanes that he [Parmenides] was led to dedicate himself
to the contemplative life (IX.21). Despite Diogenes Laertius opinion, the
belief that Parmenides was a faithful disciple of Xenophanes (the origin of
which, as we shall see, goes back to Plato) is usually held.
Let us begin, like Diogenes Laertius, with Ameinias. Nothing is known
of this person except that he was a Pythagorean.33 In fact, Pythagorism had
spread over the south of Italy since the year 530 B.C., when Pythagoras had
settled in Croton (Calabria), eeing from Policrates, the tyrant of Samos,
his country.34 It cannot be denied that Parmenides knew about Pythagorean
philosophy,35 and it was his friendship with Ameinias that awoke a sort of

31

32
33
34
35

Cf. the works of Nutton, V., The Medical School of Velia, La parola del passato 25 (1970)
21125; Gigante, M., Velina gens, La parola del passato 19 (1964) 13537; and the three
articles by Pugliese Carratelli, G., Pholarkhos, La parola del passato 18 (1963) 38586, Parmendes phusikos, La parola del passato 20 (1965) 306, and Ancora su pholarkhos, La parola
del passato 25 (1970) 13438.
Popper, K., The World of Parmenides (London/New York: Routledge, 1998), 78.
The only mention of the name Ameinias (or Aminias) is to be found in Diogenes Laertius
IX.21. Cf. Dictionnaire des philosophes antiques, ed. Goulet, R. (Paris: CNRS, 1989), 159.
On Pythagorean inuences on southern Italy, cf. the classic work of Raven, J. E., Pythagoreans and Eleatics (Cambridge, England: Cambridge University Press, 1948).
The principles that Aristotle attributed to the Pythagoreans included phos and skotos (Met.
A,5,986a22), which are picked up by Parmenides in 8.5659 (pur . . . nuktos) and in 14.3
(phao`s ka` nuktos). And in 8.41, Parmenides speaks of khros, which, according to Aetios,
was the term used by Pythagoreans to refer to the surface of a body (I,15.2).

10

(c) Life

philosophical vocation in him. But we cannot be sure, despite certain rather

hasty judgments (for example, Proclus, In Parm. I, p. 619, 4; Strabo, VI, 1,
p. 252) that he himself was a Pythagorean.
The case of Xenophanes is very different. No ancient source was able
to establish the slightest personal connection between Parmenides and Xenophanes. Apparently, the latter, a peripatetic poet-philosopher, wrote an
ode in homage to the founders of Elea,36 but this does not imply his physical
presence in the city, especially as the Phocians were almost neighbors of
Colophon, Xenophanes city, where he may well have written the ode in
question. Moreover, no evidence exists for the presence of Xenophanes in
Elea.37 Diogenes Laertius relates that the philosopher passed through Sicily
(IX, 18), but the weight of prejudices insisting that Xenophanes was Parmenides master was so heavy that a researcher of the H. Diels stamp did
not hesitate to state that there was a gap in the Diogenes text and that the
missing text should read: he was also in Elea, where he taught. In other
words, although no ancient source had said so, the presence of Xenophanes
in Elea was indubitable, and the modication of a classic text now made it
possible. Luckily, the editors of Diogenes Laertius text did not adopt this
hypothesis, which is only to be found in Die Fragmente der Vorsokratiker
(1903) by H. Diels and in bad translations that rely blindly on this text. A
brisk summing-up of the status quaestionis can be found in G. Cerri: In
reality, Xenophanes of Colophon never had anything to do with Elea.38
As I have already mentioned, it was Plato who rst established a relationship between Xenophanes and Parmenides39 in an allusive, indirect
way. The relationship is a doctrinal one. In a passage in the Sophist, Plato
wants to refute the philosophers who approach being in a quantitative
way, whether they are monists or pluralists. When he comes to the monists,
Plato, who nds this conception explicitly in Melissus of Samos (cf. Theaetetus, 180e, 183e), confronts Parmenides in the Sophist as the presumed master
of this philosopher to whom he attributes the same idea, an idea belonging to the Eleatics. And when Plato gives the genealogy of this group

36
37
38

39

Cf. Diogenes Laertius, IX.20.

Cf. Corbato, C., Studi Senofanei, Annali Triestini XXII (1952), 200 and Woodbury, L.,
Apollodorus, Xenophanes and the foundation of Massilia, The Phoenix 15 (1961) 13455.
Cerri, G., Elea, Senofane e Leucothea, in Forme di religiosita` e tradizioni sapienziali in
Magna Grecia, ed. Casio, A. C., and Poccetti, P. (Pisa/Rome: Istituti Editoriali e Poligraci
Internazionali, 1993), 138.
On this Platonic ction, cf. Cordero, N. L., Simplicius et lecole eleate, in Simplicius, sa
vie, son oeuvre, sa survie, ed. Hadot, I. (Berlin/New York: Walter de Gruyter, 1987) 16682;
and Linvention de lecole eleatique (Platon, Sophiste 242d), in Etudes sur le Sophiste de
Platon, ed. Aubenque, P. (Naples: Bibliopolis, 1991), 91124.

Introduction to Parmenides

11

(ethnos, 242d), he says that it began with Xenophanes or even before.40 That
is all.41
Firstly, I should say that even Xenophanes monism is problematic,
because his writings only speak about one single god (fr. 23), whereas
his physics is eminently pluralist (cf. frs. 29, 32, and especially 33, where
he says we come from earth and water). As for Parmenides monism, as
I shall try to show, this was merely a didactic generalization proposed by
Plato,42 considered as a sort of revealed truth by later tradition, which associated his name with Xenophanes. In short, we can state that Parmenides
followed his own way and his philosophy is eminently personal and original.

(d) Works
Tradition unanimously recognizes that Parmenides wrote one single work:
a poem (cf. D.L. I.16). There is nothing exceptional about his writing just
one text; Diogenes Laertius says that the same was true of Melissus and
Anaxagoras. It is difcult in our postmodernity to imagine such unprolic
philosophers. But we should not forget that these sages had multiple
occupations (the contemplative life was an ideal, for some, only after
Aristotle). Very probably they wrote to leave a statement of their work,
which, perhaps, they expounded in oral lectures and which they would
certainly have used as a rule of life.43 So there is nothing surprising about
a single work, as in the case of Parmenides, or just a few small treatises, as
was the case with most of the Presocratics, being enough to establish the
fame of a thinker and become an important stage in the establishment of
this new type of knowledge directed toward action, which was later called
philosophy.
40

41
42

43

It is curious that the defenders of Eleatic monism have not paid any heed to the ironic
character of Platos expression even before. Plato had already used it in the Theaetetus
179e when, in speaking about Heraclitism, he says that these ideas come from Homer and
even from an earlier time. Strictly speaking, it is difcult to imagine witnesses earlier than
Homer. With respect to the doctrine attributed to the Eleatics, an ironical spirit might
recall that the only philosopher who wrote that everything [is] one was Heraclitus (fr.
50).
Cf. Cordero, Simplicius et lecole.
This is not the case with Aristotle, who cautiously states: it is said (legetai) that Parmenides came to be his disciple (Met. I, V, 986b22). On the other hand, the lexicon Suidas
states it clearly.
On ancient philosophy as a way of life, cf. Hadot, P., Quest-ce que la philosophie antique?
(Paris: Gallimard, 1995).

12

(e) The Poem

(e) The Poem

(1) The Reconstruction
Parmenides expressed his ideas in a poem, but his work has been irreparably lost for at least fteen centuries. Nothing remains of Parmenides
original Poem. The work was probably written at the end of the sixth or
beginning of the fth century B.C.44 Without any doubt, it was copied and
recopied (always by hand) over the course of many years, but all traces of
it were lost in the sixth century of our era, that is, practically a millennium
after it was written by Parmenides. The last concrete reference to the book
appears in the neo-Platonic philosopher Simplicius (who is known to have
left Athens in 526 A.D. because the Platonic Academy was closed down).45
After quoting some lines from the Poem, Simplicius explains that he is
taking that liberty because of the rarity (dia` ten spanin) of Parmenides
book (Commentary on Aristotles Physics, p. 144). From then on, nothing is
known about Parmenides work.46
So how do we explain the fact that today we can read, quote, and
comment on Parmenides Poem? It is thanks to the work of a large number
of scholars who managed to reconstruct the lost text, though unfortunately
only in part. Thus Parmenides fate is like that of most philosophers known
as Presocratic.47 Their writings were lost, but before that happened, their
texts were abundantly quoted by authors from all sorts of disciplines (not
only philosophers but also historians, doctors, mathematicians, tragic playwrights, etc.) whose works lasted for very different reasons. Thanks to
these passages of Parmenides Poem, sometimes cited to support authors
ideas or sometimes out of simple erudition by a long series of witnesses, it
became possible to reconstruct a good part of the original work.48 As may
be deduced from this brief overview, the word fragment applied to these

44

45

46
47

48

Cornford (1939), 1, suggests that the poem was written in about the year 485 B.C. Bowra
says that the poem was contemporary with Pindars Pythica X, written in 498 B.C., and
Aeschylus Suplicantes, written before 490 B.C. (Bowra, C. M., The Poem of Parmenides,
Classical Quarterly 32 [1937]).
Cf. the works of Hadot, I., La vie et loeuvre de Simplicius dapre`s des sources grecques
et arabes, 339; and Tardieu, M., Les calendriers a` usage a` Harran dapre`s les sources
arabes et le Commentaire de Simplicius a` la Physique dAristote, 4057, both in Hadot,
Simplicius, sa vie.
With hindsight we nd, at most, indirect references to Parmenides system. Cf. Baldwin,
B., Parmenides in Byzantium, Liverpool Classical Monthly (1990) 11516.
I say most because there are some very rare exceptions: we have received in direct form
(that is, not through quotations collected by sources) a fragmentary text by Antiphon, a
few discourses by Gorgias, and, recently, a previously unknown fragment of Empedocles.
Cf. Cordero, N. L., Lhistoire du texte de Parmenide in Etudes sur Parmenide, Volume II:
Proble`mes dinterpretation, ed. Aubenque, P. (Paris: Vrin, 1987) 324.

(1) The Reconstruction

13

passages does not make sense; they are quotationsliteral, we may presume
(the philologists do not have the last word on that, but they do have the
tools to enable them to estimate the reliability of the text)from lost texts.
Quotations from Parmenides Poem begin at an early date: the rst
source is Plato, less than a century after the philosophers death. Plato transcribes a few lines in the Sophist, the Theaetetus, and the Symposium.49 Then
it is the turn of Aristotle, Plutarch, Sextus Empiricus, and so on, until we
reach Simplicius. The quotations are sometimes very short (two or three
words) but sometimes, especially in the cases of Sextus and Simplicius, they
may be more than thirty lines long. On more than one occasion the same
passage is quoted a number of times (this is the ideal situation, because
when there are divergences, we can choose the version that appears to coincide most closely with the original). In other cases, there exists a single
source and our whole interpretation depends on it.
In order to try to reconstruct the lost text, it was necessary to assemble
all the quotations from it that could be found. The success of the task depends to a large extent on the state of the sources, that is, the book that
assembles every quotation from the Poem. Some of these texts have come
down to us in complete form and it has been possible to reconstruct them
because various manuscript copies of them have withstood the passage of
time.50 Other works have been ill-treated by history, and their reconstruction requires Homeric efforts. Attempts to reconstruct Parmenides Poem
began shortly after the Renaissance, but although they were very praiseworthy, there were classical texts still unknown at that time, and the quotations from Parmenides contained in them were not discovered until several
centuries later. These attempts at reconstruction go from Henri Estienne
(Poesis philosophica, 1573) to Hermann Diels (Die Fragmente der Vorsokratiker,
1903).51 Thanks to their work, which went on over many centuries, today
we can read a good part of Parmenides Poem. Nineteen different quotations were found (one of them translated into Latin!). These were unfortunately labeled fragments, which is why, for the sake of convenience,
works on Parmenides speak about fragment 3 or fragment 5. As each
fragment includes a number of lines, it is customary to write fr. 8.34, for
example, when quoting line 34 of fragment 8.
From what I have said, it can be seen that the version of Parmenides
Poem we possess is not complete. Passages that werent quoted by anybody will
49
50
51

On the decisive importance of Parmenides philosophy to Plato, cf. Palmer, J. H., Platos
Reception of Parmenides (Oxford: Oxford University Press, 1999), passim.
This is the case with Simplicius indispensable Commentary on Aristotles Physics. A new
edition of it, edited by Leonardo Taran, is eagerly awaited by all Hellenists.
One of the rst attempts, undertaken in about 1600 by Joseph J. Scaliger, remained unpublished until 1982, when I discovered this version in the archives of Leyden University

14

(e) The Poem

remain unknown forever. Of course, the authors we use today as sources

(perhaps abusively, because these authors were writing to express their
own ideas, rather than to leave testimonies of other thinkers, except in the
case of historians of thought such as Theophrastus) quoted only those
passages that interested them. There is nothing more subjective than a
scholars interest. A paradigmatic case is the vital Parmenides text, our
present fragment 2, which postulates the existence of being, quoted for the
rst time by Proclus (In Tim. I.345) a thousand years after it was written.
Probably the discovery of the fact of being by Parmenides seemed so obvious that nobody thought to quote it. Perhaps the same thing happened
with other passages of the Poem; we will never know. Even so, today we
possess nearly 152 lines of Parmenides, and these are an inexhaustible
source of reection. So let us take advantage of them.

(2) The Form

At a time when sages, who would later be called philosophers, expressed
themselves in prose, albeit in brief treatises or epigrams, Parmenides decided to compose a Poem. Xenophanes had done it before him (although
this writer was a poet sporadically introducing philosophical reections
into his compositions, rather than a philosopher expressing himself through
poetry). But most of Xenophanes works are semi-satirical poems called
Silloi (literally, jokes), along with two elegies dedicated to the foundation
of Colophon and Elea. In contrast, Parmenides poetry takes its cue from
Homer and from Hesiods epic hexameter.52 Although Parmenides is of Ionian stock and most of the inhabitants of the Elea region are Dorians, the
poem is written in the pan-Hellenic dialect of the Homeric poems.53 All
these details combine to fulll a common aim: Parmenides wants to interest
(and be understood by) the widest possible public.54
The epic hexameter is easy to remember; the educated and even semieducated public of his time knew long passages, or even the whole works,
of Homer and Hesiod by heart. W. Jaeger rightly shows no doubt in stating

52

53
54

Library and made this known in La version de Joseph Scaliger du Poe`me de Parmenide,
Hermes 110 (1982).
An exhaustive study of Parmenides metrics can be found in Mourelatos (1970), 2 and
26468, and on Parmenides poetry in general (especially the formulas inherited from tradition) in Bohme, R., Die verkannte Muse. Dichtersprache und geistige Tradition des Parmenides
(Berne: Francke, 1986), especially 3385.
Cf. Jaeger (1947), 92.
Floyd nds a deeper relationship between Parmenides and Homer, as the philosopher is
inspired by the poet when he too presents his teaching by means of the opposing notions
of Truth and Opinion (Floyd, E. D., Why Parmenides wrote in verse Ancient Philosophy
12 [1992] 251, 263).

(3) The Content

15

that Parmenides uses a didactic epic.55 Indeed, Parmenides, convinced

he has discovered an essential, basic, and fundamental truth,56 wants to
communicate his discovery, and in order to do so he presents his Poem as
a real course in philosophy, in which a professor (in the text, an anonymous
goddess) explains to a pupil (in the text, an enthusiastic but inexperienced
youth) how to go about seeking truth.57 Parmenides deploys all the didactic
resources at his disposal: allegorical images, persuasion, demonstration
through the absurd. It is highly suggestive that, despite the Masters (that
is, the Goddesss) intellectual superiority, the criterion of authority, as well
as any claim to a revelation, are absent from the Poem. The Goddess
orders that anything she says is to be judged by reasoning (logos, fr. 7.5).
We said that Parmenides is claiming to be the bearer of a truth. This
is another way in which his Poem stands in the Homeric and Hesiodic
tradition. Both Homeric poems, like Hesiods Theogony and Works and Days,
begin with an invocation to the Muses or to a goddess to relate a truth.
Parmenides allegorically replaces the invocation with a journey toward the
Goddesss home.58 And as in the case of epic poetry, a truth will be proclaimed. On this occasion, it is not about the consequences of Achilles rage,
the vicissitudes suffered by Odysseus on his journey home, the genealogy
of the gods, the invention of women, the ages of humanity, or the propitious days for carrying out particular activities, all these realities expounded by the muses who when they wish, know how to proclaim true
things (Theogony 28). Parmenides truth will be different. The discourses
of the muses, daughters of Memory, would be incapable of presenting it,
because from Parmenides onward, truth will not come from the past; it
will arise from well-directed, methodological reection that is respectful of
certain principles and aware of the abyss that opens with the possibility
of error.

(3) The Content

I have said that, up until now, nineteen quotations from Parmenides lost
Poem have been found, and that with these, part of the text has been recon55
56

57
58

Cf. Jaeger (1947), 92.

According to Arrighetti, Parmenides decided to write in verse to differentiate himself from
the physicists, because his Poem follows another way (Arrighetti, G., Lheredita` dellepica in Parmenide, in Festschrift fur Robert Muth zum 65. Geburtstag, ed. Haendel, P.,
and Meid, W. [Innsbruck: Institut fur Sprachwissenschaft der Universitat Innsbruck, 1983],
11).
Cf. Cordero, N. L., La Deesse de Parmenide, matresse de philosophie, in La naissance
de la raison en Gre`ce, ed. Mattei, J. F. (Paris: P.U.F., 1990) 20714.
As Montaner points out, Empedocles, a disciple of Parmenides, reverts to attributing the
teaching he presents to the Muse (cf. frs. 131.1, 3.3, and 4.2) (Montaner, A., La struttura

16

(e) The Poem

structed. It is impossible to know (except in one case) what place in the

Poem each of these quotations should occupy. Since the rst attempts at
reconstruction, they have been arranged in accordance with the conceptual
content of each fragment. However, it must be recognized that Parmenides represents a special case among the philosophers called Presocratic.
Quotations from the other Presocratics have mostly been arranged in accordance with the arrangers criterion. This is the case, for example, with Heraclitus: every scholar arranges the fragments in the order that he or she
thinks is right, and an index of concordances is usually added as an appendix to the works to help the reader nd the text in the classic version
of Diels and Kranzs Die Fragmente der Vorsokratiker. Parmenides is an exception because, given the systematic character of his philosophy, there are
passages of his text that have to precede or come after others. Brief texts of
a general nature are difcult to place, and it can be said that in the present
state of the reconstruction,59 Fragment 4 is the most disputed of all fragments. Placed logically between fragment 3 and fragment 5 by most interpreters, it is considered as a sort of enclave within fragment 8 by certain
scholarsZeller60 among themor is thought to belong directly between
fragments 6 and 7.61
The exception I referred to in the placing of the fragments is the text
considered today to be fragment 1. The sole source of the rst twentyeight lines of this quotation is Sextus Empiricus, and in the passage preceding his transcription of Parmenides lines, the skeptic philosopher explains
that at the beginning of his [text] on nature [Parmenides] writes . . . (Adv.
Math. VII.111). So nobody can object to considering the extensive quotation
presented by Sextus to be fragment 1.
This quotation ends with a sort of project: the Goddess invites the
youth who wants to know to consider two ways: that of truth and that
of opinions. A text that begins precisely with the presentation of these two
possibilities seems to be the logical continuation of fragment 1, and for
that reason it is considered to be fragment 2. A single line that speaks of
the identity between being and thinking was placed next as fragment 3,
because toward the end of fragment 2, Parmenides alludes to the impossi-

59

60
61

del Proemio di Parmenide, Annali dellIstituto Italiano per gli Studi Filosoci V [197678]
138).
Which, nevertheless, is always capable of being improved: cf. Bicknell, P. J., A New
Arrangement of Some Parmenidean Verses, Symbolae Osloensis 42 (1968) 4450; and Kent
Sprague, R., Parmenides: A Suggested Rearrangement of Fragments in the Way of
Truth, Classical Philology 50 (1955) 12426.
Zeller, E., Die Philosophie der Griechen in ihrer geschichtlichen Entwicklung (Leipzig: G. R.
Reisland, 1892), 693.
Reinhardt (1916) 4849.

(3) The Content

17

bility of saying and thinking what is not, which would establish a relationship between what is and what is thought. Next comes the enigmatic fragment 4, which can be placed where the interpreter sees t, as well as the
quotation called fragment 5, which is eminently methodological or programmatic. In contrast, the text considered to be fragment 6 seems to pick
up on the nal lines of fragment 2, and the quotation called fragment 7
naturally continues the content of fragment 6. Fragment 7 is prolonged directly without any break into the extensive fragment 8. This fragment 8
ends with a negative allusion regarding any possible cosmology, which,
according to Parmenides, would form part of the opinions of mortals.
As the eleven remaining quotations (frs. 919) also refer to cosmological
questions, they have always been placed, fairly coherently, after fragment
8. The current fragment 19 is a sort of conclusion to the Poem.
According to this reconstruction of the Poem, we can state the following: (a) Parmenides begins with an allegorical presentation of his philosophy
in which, by means of a series of images easily interpretable by the public
of his time, he describes two ways of seeking offered to thought (fr. 1); (b)
he presents both possibilities philosophically and shows that one is viable
and the other is a blind alley (fr. 2); (c) he shows why anyone who tries to
pursue both ways will reach the conclusion that only one possibility can be
admitted (frs. 6 and 7); (d) a long list of properties are deduced from
the single possibility that accompanies truth (fr. 8.152); and (e) despite
everythingeven though they are deceptiveit is necessary to be aware
of the opinions of mortals, illustrated, for example, by illusory cosmologies (fr. 8.53fr. 19).
Let us now see how this project takes shape in terms of Parmenides
theora, that is, his way of confronting the reality of things.

This page has been intentionally left blank.

Chapter II: Prolegomena to

Parmenides Thesis

(a) Parmenides theora

All philosophical systems have aspired to understand reality. With the
Greek philosophers, at least in the early stages, the object of this interest
was everything that is as a whole, because when a Greek thinks of the
notion of whatfrom a Latin rootwe call reality, he thinks of everything that possesses the fact of being, that is, the totality of beings. With
this outlook they tried to establish a certain familiarity between human
beings and things, which would make possible a way of life in accordance
with this vision of reality. The Greek word for vision, contemplation, or
viewpoint is theora. This noun, related to the verbs theoreo and theaomai,
originally meant observing with precise interest, a kind of scrutiny that
goes beyond mere looking (blepo). In the title of this section I have put
the word in italics because this is a direct transliteration of the Greek. The
moment has not yet come to present Parmenides theory (not in quotation marks, this time), which would be synonymous with his philosophy.
That will be the consequence of a certain way of contemplating the reality
of things, and for the moment I present this rather special way of looking
at the Greek word theora, alluding, I repeat, to a preliminary but fundamental stage: anyone who simply looks at things will never reach amazement, that state of mind (pathos) which both Plato (Theaet. 155d) and Aristotle (Met. A, 2, 982b 12) make the starting point for doing philosophy. In
contrast, anyone who observes reality as if it were a theatrical representation
(theater derives from theaomai) will surely be interested in the plays plot
and will ask questions about how the action happens; instead of just looking, they will ad-mire (that is, they will be amazed), and thus will become a lover of knowledge, a philosopher.
Parmenides theora is different from that of his predecessors. For more
than a century, this new activity (which gradually became a kind of
knowledge) had been acquiring a certain specicity. Although we may
not blindly trust the didactic outlines drawn up later by Aristotle (which
reached posterity through Theophrastus), there is an air of kinship between
the thinkers of Miletos, the rst Pythagoreans, and another isolated thinker
who did philosophy before Parmenides. And this atmosphere will reappear

20

(a) Parmenides theora

later among the last Presocratics, the atomists. To a greater or lesser extent, these philosophers observed reality (that is, put into practice a theora)
with the intention of explaining it through principles or elements, from
which, as the case might be, either internal or external forces could be derived that justied the developmental stages leading from the rst germ to
reality as it was now constituted.
The task of these lovers of wisdom was titanic. If we had preserved
their original writings, we would have to invent an even more grandiose
adjective to describe their discoveries. Butand there is nothing pejorative
about this but; I am simply trying to point out the difference between
these theoras (which are fairly similar to each other) and that of Parmenidesthis way of confronting reality assumed a basic fact, without which
it would have been impossible to reect on the principles or elements of
things: the existence of things. As there are things (we will look at this
notion of thing later), it was reasonable to investigate them and question
their origins, their components, their connections, and their disconnections.
We can say that everything was questioned, except this apparently obvious
fact: the existence of these things.
Parmenides theora is set on a different plane. He is interested in this
basic fact assumed by his predecessors: there are things. What does it
mean to assume that there are things? It would appear that Thales stated
that the rst principle of reality was water (or rather moisture), that Anaximander proposed an indenite rst principle (to` apeiron), and that Anaximenes inclined toward air and the Pythagoreans toward mathematical
entities. But no one asked what it means that this rst principle should
exist, and even less, why it exists. The rst principle exists; things exist.
Isnt that amazing? Why is it amazing? Because there could have been nothing, yet there is something. Why? This is not asking about a possible creation, because, as we know, that notion has no place in Greek thought.
The establishment of the fact that there is a reality (which later becomes
Parmenides thesis) opens up unexpected horizons for the philosopher,
contains undreamt-of riches, and displays an inexhaustible fruitfulness, to
the point that it would appear that for Parmenides the inescapable, basic
task of the philosopher must consist in grasping the ultimate consequence
and total scope of the formula there are things. And Parmenides gives
the example, because his Poem undertakes to do this.
There can be no doubt that Parmenides was aware of the vital importance of the message he proposed to transmit to his contemporaries. That
is why he used poetry as his medium of expression and why, in his own
way, he methodically demonstrated his intuitions. As we saw in the previous chapter, his use of the epic hexameter shows a clear desire to address
listeners didactically, or possibly also readers (since few people could read),

Prolegomena to Parmenides Thesis

21

who were not necessarily already attracted to the new philosophical adventure. Doubtless, this is the reason that the Poem begins with a sort of introduction abounding in easily recognizable Hesiodic images (day and night
roads, ethereal gates, winged chariots, gracious daughters of the Sun, and
so forth). Through this ploy, once the listener/reader nds himself on familiar ground, the dryness of the philosophical discourse that begins at the
end of this introduction seems almost familiar.
Parmenides presents his ideas as stages on a way or road to be traveled,
along which a future philosopher will have the privilege of being guided
by an anonymous Goddess, who acts as a real professor expounding a kind
of masters course. This forces him to follow a method (remember that
this term comes from the Greek hodos, way), with axioms, stages, conclusions, demonstrations through the absurd, principles, and so on. Parmenides wants to suggest the unsuspected universe hiding under the apparently banal assertion there are things. And this assertion does appear
banal, because in Greek to assert that there are things is (or would be) a
tautology. Indeed, the word things does not exist in Greek (there are
terms to refer to certain types of things, for example, utensils, khremata; productions or matters, pragmata; but not for things in general). So, with the
same meaning that the word things has in English (or cosas, or
Dinge, or choses, or cose), a Greek uses the present participle of the
verb to be ta` onta (as a plural noun), that is, literally, beings, that
which is or that which is being (in the plural), the [things] that are.
As E. Benveniste wrote, the linguistic structure of Greek created the predisposition for the notion to be to have a philosophical vocation.62 Or, as
I said, to assert solemnly, as Parmenides does, that there are things means
admitting that there is what is being or, more generally, by being, it is.

(b) The Allegorical Presentation of the Content of the Poem

Sextus Empiricus, the single source of the rst twenty-eight lines of Parmenides fragment 1, tells us that this passage corresponded to the beginning of the Poem.63 This fact is of priceless value, since the beginning of a
work is usually where the keys to understanding the rest of it are to be
found. A century after Parmenides, Plato uses the same procedure in his
dialogues. Let us look at just two examples (because this is not our main
subject). The Timaeus, a dialogue that uses numbers and mathematical cal62
63

Benveniste, E., Proble`mes de linguistique generale (Paris: Gallimard, reprinted 1966), 73.
Sextus Empiricus, Adv. Math. VII.111: In fact, [Parmenides] wrote in this way at the
beginning (enarkhomenos) of his Per` Phuseos . . .

22

(b) The Allegorical Presentation of the Content of the Poem

culations to propound a structure of the universe, begins thus: One, two,

three, who is the fourth? (17a). And in the Sophist, where Plato presents
his philosophys dernier cri, which will consist in assimilating non-being
with the different, the dialogue begins by saying that the Stranger of Elea
is different (heteron, and not hetaro)64 from the Parmenideans.
Given the eminently didactic character of Parmenides Poem, it is not
surprising that the text should begin with poetic images and direct allusions to traditional epic, as if trying to grab the attention of the reader/
listener (as I already said, few people, even the educated, could read at that
period: ideas were transmitted, as M. Detienne graphically expressed it,
by mouth and by ear).65 As we shall see, the origin of certain cliches are
easily detectable by the reader/listener of the period, and once the author
has established communication, he can transmit the nucleus of his thought.
Once the introduction had grabbed the readers/listeners attention, the
philosophy course could begin.
The Poem begins with the story of a journey. In every journey there is
a way to travel (a way that assumes a starting point and a goal), there is a
traveler, there is a means of transport, and when the traveler is inexperienced, there are guides or leaders to help him in his task. All these elements
are present in the rst fragment of the Poem, and all have a symbolic value.
First of all, we may note that Parmenides does not use suspense: the rst
lines conrm the successful outcome of the journey, because the traveler
refers to his experience as something that has already happened, and
speaks of himself as someone who already possesses the knowledge he
sought.66 This is the reason why he decides to convey his experience, which
amounts to relating the events of his journey along a way by which he
arrived at the goal: complete successperfection, if we use this term in its
etymological sense, the achievement or nishing (per-) of an undertaking.
It is precisely the image of the journey and the way (that is, route),
which is both physical and intellectual, that will be central in Parmenides
philosophy. Indeed, this will become the presentation of the single way for
thought to travel, and the demonstration of the foundations establishing that only

64
65
66

Cf. my translations of Plato, Sosta, Dialogos, Vol. V (Madrid: 1988), 332, and Plato, Le
Sophiste (Paris: GF-Flammarion, 1993), 73.
Cf. Detienne, M., Par la bouche et par loreille, in Linvention de la mythologie (Paris:
Gallimard, 1981).
According to Ruggiu (1975), 21, the rst lines describe a return journey. Other authors,
like Taran (1965), 30, prefer to speak of a repetitive experience, which also presupposes
that he has already arrived at the end of the journey and is starting out again. Finally,
Gomez-Lobo (1985), 30, sees two stages of the journey: one in which Parmenides is led
towards the way, and another in which he travels along the way.

Prolegomena to Parmenides Thesis

23

this way exists. The notion of way, represented by various different

terms,67 but mainly by hodos, appears fteen times in the Poem (hodos, nine;
keleuthos, three; patos, one; amaxitos, one; atarpos one). This fact, which is not
accidental, shows that for Parmenides, knowledge is gained by a route,
a journey, a conceptual course, that is, through a method. Indeed, as we
have already seen, method comes from meta` hodon, making for, being
on the way. Of course, the ways Parmenides proposes are conceptual ones,
that is, methods to follow, but their presentation constantly swings between
a rigorous description and allegorical images, especially in this rst fragment. Finally, we may say that with Parmenides Poem, the image of the
way, or more broadly, that of a journey as a method of access to the
truth, makes its entry in denitive form into the domain of philosophy.68
As I have said, fragment 1 tells a story. It is not simply a literary
artice, as L. Taran69 asserted, but a summary of the Poem in general. A.
Marsoner, who stresses the rst fragments architectonic character, does
not hesitate to state that this prologue is a sort of pediment to the temple
that harbors the secret of truth.70 To gain access to this secret, the traveler has to progress in a particular direction (method?). But the undertaking
will be difcult, because obstacles will appear that will have to be overcome; however, nally, the quest will be crowned with success. Looking
back, the traveler will see that he has passed from a dark road to a light
one, but he could never have reached the light road without going along
the dark one rst. The rest of the Poem develops these elements.
Various unknowns have to be resolved in order to understand the
scope of this fragment 1. The rst concerns the character we have hitherto
called the traveler. He is a character speaking in the rst person who
says he has decided to go on a journey as far as his thumos will take him
(fr. 1.1). This notion (which Plato will resume again in his division of the
67

68

69
70

These terms are synonymous, and Parmenides chooses one or another for metric reasons.
For example, in fragment 2, after saying that there are only two hodo (ways) of investigation, Parmenides points out that one of them is the keleuthos (way) of persuasion. There
are scholars who have found differences between the terms. For example, Wyatt says that
Parmenides wants to exploit the wealth of the Greek roots of each term, and that the
different words he uses for way have different meanings (Wyatt, W. F., The Root of
Parmenides, Harvard Studies in Classical Philology 94 [1992] 114).
We need only remember the journey of Platos prisoner, who escapes the cave and goes
in search of the Sun (Rep. VII, 514a517a). On the image of the way in general, there are
two classic works: Misch, G., Der Weg in die Philosophie (Berlin/Leipzig: 1926); and Becker,
O., Das Bild des Weges und verwandte Vorstellungen im fruhgriechischen Denken,
Hermes Einzelschriften 4 (1937) 1223. With respect to this image in Parmenides, the most
complete study continues to be Chapter 1 of Mourelatos (1970), especially 1624.
Taran (1965), 31.
Marsoner, A., La struttura del Proemio di Parmenide, Annali dellIstituto Italiano per gli
Studi Storici 5 (197678) 181.

24

(b) The Allegorical Presentation of the Content of the Poem

soul into three parts in the Republic)71 refers in Homer not just to courage
and impulse72 (which for us would be equivalent to will),73 but also includes a certain capacity to discern. When Circe presents Odysseus with
the two routes (hodo: ways, routes) available for him to navigate (Scylla
and Carybdis) she tells him he will have to decide in his thumos which one
to take (Od. 12.58). So here thumos is also connected with deliberation. It has
to do with a reexive will that meditates, a kind of sensible impulse that
decides upon an action with a clear and precise aim that has to be reached,
even at great risk. Or rather, Parmenides, as a master of philosophy, demands
a voluntary and conscious impulse on the part of anyone who wants to
learn.74 This way of confronting the approach to knowledge contrasts with
the more passive attitude of the listener to the Muses and other traditional
masters of truth, who can deceive as well as teach.75 In these cases, the
listener has to trust the master who announces a truth when he likes (or if
he likes; cf. Hesiod, Theog. 28). Parmenides traveler sets out in search of
truth impelled by his thumos. Once he has taken in the Goddesss message,
he will judge by reasoning the arguments he has just heard (fr. 7.5).
Nothing is known about this traveler. The Goddess describes him as
kouros (young) when she receives him, once the introduction to the
journey is over (fr. 1.24), but this adjective can be interpreted in very different ways.76 Despite the fact that nothing could be more natural than that it
should be a young person wanting to be educated, this epithet heightened
the imagination of various scholars, so that the account in fragment 1 was
even interpreted as autobiographical. According to this point of view, Parmenides did not reach his akme during his forties (the usual arbitrary reck-

71
72
73

74

75
76

Cf. Plato, Republic, 439e441c.

It is thumos that impels Demodocos to sing before the gods (Od. VIII, 45).
Ardor, Riaux (1840), 207, and Cassin (1998), 71; will, Untersteiner (1958), 121; heart,
Taran (1965), 8; boldness, Gomez-Lobo (1985), 27, and vital impulse, will, 30; desire, Conche (1996), 42, and Collobert (1993), 10; spirit, Coxon (1986), 44. For Casertano
(1978), 13, it is an intellectual passion, and he therefore translates it impulse of my
mind.
As Arrighetti says, the man who knows [ . . . ] has already reached total awareness of the
strength of his intellect and its capacities (Arrighetti, G., Leredita` dellepica in Parmenide, in Festschrift fur Robert Muth zum 65. Geburtstag, ed. Haendel, P., and Meid, W. [Innsbruck: Institut fur Sprachwissenschaft der Universitat Innsbruck, 1983], 13).
Cf. Detienne, M., Les matres de verite dans la Gre`ce archaque (Paris: Maspero, 1967), 77:
Un matre de Verite est aussi un matre de tromperie.
Cosgrove analyses the term kouros, young, in a context of initiation and deduces that it
means without experience, which does not necessarily imply that this is a young person (Cosgrove, M. R., The kouros Motif in Parmenides, Phronesis 19 [1974] 94). As here
it is a goddess addressing a disciple, Marsoner proposes the translation oh, son (Marsoner, La struttura del Proemio, 155).

Prolegomena to Parmenides Thesis

25

oning) but much earlier.77 There is nothing to authorize this interpretation.

As occurred in Platos dialogues, Parmenides speaks through his mouthpiece: the Goddess, in his case; Socrates (and sporadically, the Stranger of
Elea, Timaeus, the Athenian, and so forth) in Platos. If the Goddess addresses a young listener, this listener can only be a possible pupil of
Parmenides.
The traveler is carried in a chariot drawn by mares. It is normal that it
should be a chariot:78 gods and heroes (including the Sun) travel in chariots
when their route is a journey by land . . . or in the sky.79 In contrast, sea
voyages are made using all kinds of ships. For its part, the reference to
mares has been the subject of very different interpretations. For GomezLobo, the fact is almost banal, because it is the custom of Homeric heroes.80 But this is not always the case. Sometimes (for example, Il. 24.326,
11.615) heroes are drawn by stallions. As the Orphics usually use mares,
the suggestion has been made of an Orphic inuence on Parmenides.81 A. H.
Coxon believes that Parmenides preferred female horses because everything concerning the feminine is warmer than the masculine, and therefore nearer to re and light.82 Lastly, there has also been the suggestion of
Parmenides feminism avant la lettre, because most or nearly all of the
mythical gures in the Poem are feminine.83
Much more interesting than the problem of the horses sex is the description of the maiden guides on his journey, because both the traveler
and the mares are practically dragged by the Daughters of the Sun, the
Heliades. The choice of these divinities by Parmenides is not accidental but
decisive. I said earlier that an educated reader/listener of the period would
be capable of recognizing the poetic images of fragment 1. So we should
remember that the Heliades, whether they are just two (Homer, Od. 12.127),

77
78

79
80
81
82
83

On the autobiographical nature of the reference, cf. Reinhardt (1916), 111, and Kranz, W.
Vorsokratisches, Hermes 69 (1934) 118.
Even so, it is interesting to point out that Pindar, a contemporary of Parmenides (Bowra,
The Poem of Parmenides, 38), states that the Poems composition coincides with that of
the Pythic Ode X; he associates the image of the chariot with the Muses, that is, with poetry:
cf. Ol. 9, 80; Isth. 2, 1 and 8, 62. On the relationship between Parmenides and Pindar, see
Martinelli, F., Fra Omero e Pindaro: Parmenide poeta, in Forme del sapere nei Presocratici,
ed. Capizzi, A., and Casertano, G. (Athens/Rome: Edizione dellAteneo, 1987) 16986.
Both Zeus (Il. VIII, 41, 438; XIII, 23) and Hera (Il. V, 748, 380) usually cross the sky by
chariot, which they themselves drive.
Gomez-Lobo (1985), 30.
Cf. Taran (1965), 9.
Coxon (1986), 157.
Cf. Merlan, P., Neues Licht auf Parmenides, Archiv fur Geschichte der Philosophie 48 (1966)
26776.

26

(b) The Allegorical Presentation of the Content of the Poem

ve, or seven,84 have a brother, Phaeton, whose story is well known. When
he wanted to take his fathers place and drive the Suns chariot, his inexperience produced such great catastrophes that Zeus struck him with his
thunderbolt.85 What caused Phaetons disaster? The cause was double: (1)
the rebel son did not have the right to do what he did, and (2) he undertook
a journey without knowing the way to go; he drove forward at random,
without any guide, without obeying any kind of parameter. Thus, Phaeton
becomes a negative image of the Parmenidean traveler, whose journey does
have those elements that were absent from the unfortunate child of the
Suns feckless dash: (1) the guarantee of right and justice (as we shall see)
and (2) maiden guides who know the right direction. So the Goddess will
be on his side. The choice of the Heliades as guiding divinities spells a
message which Parmenides contemporary reader/listener would certainly
pick up.
Having presented the characters, now let us look at the journey. The
structure of this fragment 1 is very complex, because Parmenides swings
continually between the present (when the traveler has already arrived at
the Goddesss realm) and the past, when he followed the ways that led to
his goal. A very detailed analysis of Parmenides method in annello can be
found in A. Marsoners excellent work, to which I refer the reader,86 and
also in C. M. Bowras classic article, The Proem of Parmenides.87 The
three rst lines conrm that the traveler was taken toward the way of the
Goddess, a way that is abundant with signs (poluphemon). When we analyze
the content of this way which is found in the extensive fragment 8, we will
study the meaning of these signs. For the moment, we may say that,
faithful to his method, Parmenides will present arguments in support of
his theory and that each argument will be a sort of proof (in the legal sense)
of it. Each proof will be a sort of pheme, meaning a sign or oracular word,
an announcement.88
Along this way, the Goddess leads the man who knows.89 Anyone
familiar with Parmenides text will be surprised not to nd in my summary
the formula according to which the Goddess (or the way, for those who
believe way is the subject of lead90) leads the man who knows through
84
85
86
87
88
89
90

Cf. Falcon Martnez, C., Fernandez-Galiano, E., and Lopez Melero, R., Diccionario de mitologa clasica, Vol. I (Madrid: Alianza, 1980), 292.
Cf. Euripides, fragments of the tragedy Phaeton, 77186.
Marsoner, La struttura del Proemio.
Bowra, The Poem of Parmenides, 97112.
Cf. Sophocles, Oedipus Rex, 86: What message (phemen) from the god do you bring us?
An analysis of these three lines can be found in Cordero (1997) 17678.
In line 1.3 there is a feminine relative pronoun, and its antecedent can be either the
Goddess or the way (feminine, in Greek). In the rst case, it is the Goddess who leads,
and we adopt this possibility, given that it is she who acts pedagogically toward the trav-

Prolegomena to Parmenides Thesis

27

all cities (aste). In fact, classical texts evolve, and the word cities, which
was introduced into a gap in line 1.3 of the Poem in 1909 due to an editors
mistaken reading of a manuscript by Sextus,91 disappeared in 1969 when
another scholar92 demonstrated that the codex in question presented the
same state of incompleteness as the whole of the manuscript tradition. So
for more than thirty years, it has been known again that fragment 1.3 has
a gap that must be lled if we want to complete the text with conjectures.93
Even so, regrettable though it is that recent translations ignore the inexistence of cities, a really pathetic case is that of G. Reale, who asserts that
this nonexistent word is la meglio attestata.94
So the journey has an arrival point, which is the realm of the Goddess
(and which is, according to our interpretation, the enigmatic there), but
what was the starting point? There can be no doubt: the answer is darkness.
Parmenides does not say so clearly, but the traveler and his guides and
companions make toward the light (1.10),95 and the Heliades, who accom-

91

92

93

94

95

eler and, as is well-known, in the root of the term pedagogy we nd the idea of leading (agein). Our choice is also supported by the role Parmenides attributes to the Goddess,
who rules all (fr. 12.3). In contrast, most interpreters opt for the way, and in that case,
the man who knows would be led by (and not along) a way.
This was Mutschmann, who thought, erroneously, he had seen this term in the manuscript
berlieferung der Schriften des Sextus EmLaurentianus 85.19 (cf. Mutschmann, H., Die U
piricus, Rheinisches Museum 69 [1909]).
Coxon, A. H., The Text of Parmenides Fr. 1.3, Classical Quarterly 18 (1968) 69. Direct
consultation of the Laurentianus manuscript 85.19 in the Laurenziana library in Florence
proved to us that the text of folio 124 conrms that Coxons thesis is correct.
I proposed the word there, and interpreted the preceding phrase as in every respect,
which does not contaminate the text in question too much (the fact that the Goddess leads
in every respect will be conrmed by fr. 1.28, and the enigmatic there appears various
times in fragment 1: cf. Cordero, N. L., Le vers 1.3 de Parmenide, La revue philosophique
107(2) (1982) 15979, where I also examine all the conjectures proposed to date). For other
interpretations, cf. Cerri, G., Il v. 1.3 di Parmenide: la ricognizione dellesperienza, in
Mousa, Scritti in onore di Giuseppe Morelli (Bologna: Patron, 1997), 5763. Fairly recent
works, which commit the mortal sin of disinformation and continue translating a nonexistent text, include Les Presocratiques, trans. Dumont, J. P., and others (Paris: Gallimard,
1988), 255 (translation re-used in the compendium Les ecoles Presocratiques [Paris: Gallimard, 1991], 345); Parmenides, I frammenti, trans. Trabattoni, F. (Milan: Marcos y Marcos,
1985), 15; De Tales a Democrito. Fragmentos Presocraticos, trans. Bernabe, A. (Madrid: Alianza, 1988), 159; Tzavaras, G. To` Poema tou Parmende (Athens: Domos, 1980), 20; etc.
Parmenides, Poema sulla natura, trans. Reale, G. (Milan: Rusconi, 1991), 85. On this publication, cf. Cordero (1997), 1314. More coherent is the position of Lesher, who admits the
terms nonexistence, but proposes it as an acceptable conjecture (Lesher, J. H., The
Signicance of kata` panta<s>te in Parmenides Fr. 1.3, Ancient Philosophy 14 [1994] 15), and
of Gunther, who speaks of a Minimalkonjektur (Gunther, H. C., Aletheia und Doxa, Das
Proomium des Gedichtes des Parmenides [Berlin: Dunker & Humblot,1998] 15).
Most interpreters are of this opinion (among them, Kahn [1968/69], 704; and Vlastos, G.,
Parmenides Theory of Knowledge, Transactions and Proceedings of the Amer. Philol. Assoc.
77 [1946] 73, note 43). A contrary viewpoint can be found in Burkert, W., Das Proomion
des Parmenides und die Katabasis des Pythagoras, Phronesis 14 (1961) 130.

28

(b) The Allegorical Presentation of the Content of the Poem

pany the traveler (1.9, 1.24) and show him the way (1.5), abandon the
realm of night (1.9).96
Consequently, we may suppose that the young traveler also leaves the
domain of darkness, otherwise it cannot be explained how the maidens can
guide him or drive him (1.8) until he reaches some gates (1.11), which appear to close off one region and open the way to a different sphere. So there
can be no doubt that the beginning of the journey takes place along a road
belonging to the realm of night, and he must continue until he comes to
the home of the Goddess, represented by an opposite way. It is interesting
to point out the assimilation Parmenides makes between way and domain, an analogy that becomes plain when the Goddess congratulates the
traveler for having taken this way (tendhodon, fr. 1.27), which is none
other, she says, than my home, (hemeteron do, fr. 1.25), whereas nights
home (domata nuktos, fr. 1.9) corresponds to the way of night (fr. 1.11).
The gates the traveler nds in line 1.11 are the gates of the ways of
night and day. Parmenides makes use of an image already used by Homer
and Hesiod to refer allegorically to two incompatible, contradictory, exclusive spheres. Both ways are close (eggus) to each otheras we read in Od.
X.86but one comes after the other; they are successive.97 The same shepherd cannot look after his ocks by day and by night; to do so he would
have to do without sleep, and in that case, the author of the Odyssey says
ironically, he would earn a double wage. When Hesiod takes up the image
of night and day he conrms that the same home (do) does not shelter both
at once: when one enters, the other goes out (Theog. 74851).
The journey along the way of night ends, as I said, when the traveler
and his companions come to two heavy, closed gates, which prevent them
from going any further (1.11).98 They have encountered an obstacle that
prevents them from entering the way of day (symbolized by light), that is,
access to truth. The possible opening of these gates depends on Dike, a sort
of porter who, together with her sisters, the Hours, keeps the keys of the
sky, which enable them to regulate the seasons. In Parmenides she holds
the keys that alternate (1.14) and that will open the gates of the realm of
truth. Here, too, the philosopher resorts to classical images, because Dike

96

97
98

Kern, O., Zu Parmenides, Archiv fur Geschichte der Philosophie 3 (1890) 17376, analyzed
the Orphic origin of the image of the realm of night, as well as the decisive role of Eros
(in Parmenides, cf. fr. 19), father of the night.
According to Becker, eggus has a temporal value in this passage (Becker [1937], 12, note
7).
On the position and number of these gates, cf. Cordero, N. L., Acerca de tres pasajes del
Poema de Parmenides, Revista latinoamericana de losofa 1 (1975) 23743, and Cordero
(1997), 17981; for a polemical viewpoint, cf. Gomez-Lobo, A., Parmenides. Las puertas
de la noche y del da, Revista latinoamericana de losofa 3 (1977) 18588.

Prolegomena to Parmenides Thesis

29

is traditionally associated with truth: between Justice and Truth there is

no distance at all.99 Dike, together with Themis, who will appear in line
1.28, represents justice and right.100 Already in Parmenides time her sentences, like lawyers submissions, are supported by arguments (logoi). In
the case we are concerned with, these logoi are presented by the Heliades:
they are persuasive or enveloping (1.15)101 arguments that show, we presume, that the traveler, unlike the unfortunate Phaeton, has the right
to continue on his way. Indeed, Dike, who abundantly chastises (she is
polupoinos, 1. 14), allows the traveler to enter the realm of truth. These arguments persuade, convince102 (pesan, 1.16) the severe Dike. The heavy
gates are opened and a new great way (amaxitos, 1.21), belonging (hemeteron, 1.25) to the anonymous Goddess, awaits the travelers. The philosophy
course can begin.
Up to this point in his Poems prologue, how do we assess our understanding of Parmenides philosophy? Two spheres monopolize the scene of
the journey: darkness, night on the one hand; light, day on the other. Both
are represented by ways or roads, each having their own gates, which open
and shut at a certain point (There stand the gates of the ways of night and
day [1.11]) through which the traveler can pass. According to A. Pieris
expression, the gates divide two regions, two ways, one towards darkness
and the other towards light.103 The analogy between darkness and ignorance is perfectly plain.104 Someone wanting to know is ignorant of the truth,
and his mind is dark, fogged. Even so, this total absence of knowledge
potentially possesses all knowing. Much has been written about the Hesiod
passage that inspired Parmenides image.105 This comes in lines 744 onward
in the Theogony, which describe this dark, underground realm in which
99
100

101

102
103
104
105

Detienne, Les matres de verite, 34, note 14. In support of his statement, Detienne cites
texts by Mimnermos, Solon, and Plutarch, as well as Hesychius Lexicon.
Cf. Deichgraber, K., Parmenides Auffahrt zur Gottin des Rechts (Wiesbaden: Akademie der
Wissenschaften und der Literatur in Mainz, 1959). According to Burkert, Dike controls
the legal exchange between Day and Night (Burkert, Das Proomion des Parmenides, 10,
note 34).
Parmenides takes the expression enveloping arguments from Homer and Hesiod: Patroclus tries to calm Eurypilos sufferings with these types of arguments (Il. XV. 390),
Calypso wants to detain Odysseus (Od. I.56), Zeus deceives Metis so that he can swallow
her (Theog. 890), and Apollo sets them in Pandoras breast (Works 78).
When we look at line 2.4, we will analyze the notion of persuasion (peitho) in Parmenides.
Pieri, A., Parmenide e la lingua della tradizione epica greca, Studi Italiani di lologia
classica 49 (1977) 80.
Platos cave allegory is the most widely-known example, but even in everyday language
obscurantism is synonymous with ignorance.
Cf., for example, Pellikaan-Engel, M. E., Hesiod and Parmenides: A New View on Their
Cosmologies and on Parmenides Poem (Amsterdam: Adolf M. Hakkert, 1974), and Schwabl,
H., Hesiod und Parmenides, Rheinisches Museum 106 (1963) 13442.

30

(c) You Must Inquire About Everything (1.28)

the sources (pega) of everything are found (809). As in Anaximanders

indenite (apeiron), everything is programmed in that kind of alreadyfertilized nothing that is ignorance. When they go through the gates, darkness vanishes and truth begins to show. At that moment the Heliades, who
hid their brightness (inherited from their father the Sun), drop the veils that
covered their heads (1.10).106 Anachronistically, we might say that from now
on, at lux.

(c) You Must Inquire About Everything (1.28)

The travelers efforts have not been in vain: the Goddess receives him
readily and tells him what his task will be if he wants to become a man
who knows. So it is necessary for you to inquire about everything; on
the one hand, the unshakable heart of well-rounded truth, and, on the
other, the opinions of mortals, in which there is no true conviction (1.28b
30). As the Goddess will guide him, she begins by proclaiming what the
content of her philosophy course will be, and the reference to the totality
of things being the object of her teaching encourages me to propose the
conjecture about everything (kata` pan) to make sense of the corrupt passage in 1.3.107 A priori, the project looks excessive, because the Goddess invites him to inquire about everything (panta), but the scope of this everything is dened in the following lines: it consists of two contents that are
complementary but, apparently, necessary: (1) the heart of truth, and (2)
the opinions of mortals.
Firstly, let us say that after the division of knowledge illustrated by the
rst twenty-eight lines of fragment 1 into two possibilities (ignorance,
truth), this new pair (which will also turn out to be necessary [khreo`, 1.28],
106

107

In this unveiling Somville nds a reference to the wedding ceremony of anakalupteria,

literally, the removal of the veil (Somville, P., Parmenide dElee: Son temps et le notre
[Paris: Vrin, 1976], 37). When I read this Parmenides text, I cannot help thinking of the
characteristic of un-veiling, dis-covery, the notion that truth has for Heiddeger, M.
(cf. especially Sein und Zeit, 9th ed. [Tubingen: Max Niemeyer, 1960], 44). So it is strange
that Heidegger did not use this passage to support his thesis (indeed, Heideggers Parmenides [Gesamtausgabe II: Abteilung, Vorlesungen 19231944, Vol. 54 (Frankfurt am Main:
Klostermann, 1982)] contains no reference to fragment 1.10). This absence accords with
Heideggers habit of not seeing things where they are and seeking them where they are
not. The most signicant examples of this tendency are his analysis of truth in a work
entitled Aletheia, which studies this idea in a text of Heraclitus (fr. 16), from which this
notion is absent, and his writing on truth in Plato, based on the cave allegory, where the
idea only appears in allusive form, whereas it appears abusively in the Euthydemus, the
Cratylus, the Theaetetus, and especially in the Sophist, where it is stated that the topos of
truth is the logos, a phrase which Heidegger repeats ad nauseam in Sein und Zeit.
Cf. Cordero, Le vers 1.3.

Prolegomena to Parmenides Thesis

31

since just as there was no knowledge without previous ignorance, there

will be no truth without opinion) seems to resume the preceding lines
dichotomous presentation, in which, a priori, only two possibilities could
be taken into account.
So the task is to inquire about two possibilities. The Goddess does
not say know or learn, because, according to Parmenides assessment
of certain concepts, one of the possibilities will prove to be completely unknowable (panapeuthea, fr. 2.6). Later on he will say that there are two possibilities offered to thought, but not that both are thinkable (fr. 2.2). It is
only a matter of being abreast of everything, even though the scope of
this everything has now been delimited. On the one hand there is the
heart of truth. The image was exhaustively analyzed by E. Martineau,108 and
its content is plain: the heart is not only the vital core but also the central
nucleus of the individual. Truth also has its heart, and for truth to be
truth, that nucleus cannot change from moment to moment; it must be
unalterable, rm, constant, or literally intrepid (atreme`s, 1.29). The rest of
the Poem will show that this nucleus will be proof of the fact of being, and
that the way that proclaims this accompanies the truth (fr. 2.4). But as
the manner of approaching truth will be circular (cf. fr. 5: It is common
for me that where I begin, there I shall return again), truth itself is considered here as well-rounded (eukukleos, 1.29).109 And on the other hand,
there are the opinions110 of mortals. It is worth pointing out the fact that in
Parmenides, opinions are always a prerogative belonging to mortals. The
genitive opinions of mortals is subjective: the mortals have the opinions;
it does not mean opinions about mortals. This point is important because,
in Parmenides, doxa is always a point of view, an opinion, and never an
image or an appearance. Parmenides is not Plato, who distinguishes
between being and appearing. Unfortunately, a retroactive application
of Platonic schemes to Parmenides is one of the hindrances unnecessarily
obscuring our understanding of the philosopher of Elea.111
So the content of the apprenticeship of the future philosopher includes
truth (especially its central nucleus) and opinions. The formula used in
Greek by Parmenides leaves no doubt: he says clearly eme`n (on the one

108
109

110
111

Martineau, M., Le coeur de lAletheia, Revue de philosophie ancienne 4 (1986) 3386.

There is an interminable dispute between those who accept Simplicius version at this
point, eukukleos, and the partisans of the version given by Plutarch, Diogenes, and Clement, eupeitheos (well persuasive). Among the latter recently to be found are Coxon
(1986), 51; Conche (1996), 61; and Cassin (1998), 73. My choice is based on the greater
reliability of Simplicius testimony.
On the notion of doxa, cf. the excellent work by Lafrance, Y., La theorie platonicienne de la
Doxa (Paris/Montreal: BellarminLes Belles Lettres, 1981).
Cf. Cordero, N. L., Parmenide platonise, Revue de philosophie ancienne 18(1) (2000) 1524.

32

(c) You Must Inquire About Everything (1.28)

hand) and ede` (on the other): it is necessary to inquire about both. So it is
hard to understand why most scholars of Parmenides thought are surprised when the Goddess completes her project and also expounds a possible model for opinions. Truth is absent from opinions, but knowing that
opinions are not true, is true. As mistress of philosophy, the Goddess must
didactically show the disciple what the error consists in. In M. Detiennes
text, which I cited in note 14, the author stated that a genuine master of
truth is also a master of deception.
Parmenides is not the only one to present an erroneous doctrine in
order to show its essential aw. Mathematicians of Euclids school presented students with false reasonings called pseudaria to accustom them to
recognizing formal aws and thus be on their guard against error. Proclus
says that Euclid presented a method for detecting paralogisms and wrote
a treatise on this called Fallacies, a work that is both cathartic and gymnastic (In primun Euclidis elementorum librum commentarii, 69). Opinions fulll
this same function in Parmenides. Indeed, opinions must also be the object
of study, but even from line 30 of fragment 1, Parmenides indicates unambiguously that opinions are not reliable, and one cannot have any real condence (pstis alethes) in them. It is not exaggerated to deduce from this
expression that opinions are not true, given the dichotomous way in
which Parmenides thought is presented. In another passage of the Poem,
the Goddess says: Henceforward learn the opinions of mortals, listening
to the deceitful order of my words (8.5152), which directly suggests that
opinions are wrong.
Why learn something that is not true, which may also be a source of
errors, deception, as is the case with opinions? Parmenides is aware of the
unusualness of his proposal, so he explains. After having stated that opinions are not true, he uses a strongly concessive formula: allempes (1.31),
yet, nevertheless, you will also learn these things.112 This formula had already been used in Homer to allude to a restriction in relation to what has
just been written.113 Opinions are not true, but nevertheless, it is necessary
to learn them. Why? Lines 1.31b32 explicitly answer this question, and,
given the precision of Parmenides text, I cant help being surprised by the
sterile debate which, as we shall see, this passage has aroused.

112

113

An excellent and subtle analysis of this passage can be found in Dehon, P. J., Les recommendations de la deesse. Parmenide, fr. 1.2832, Revue de philosophie ancienne 6(2) (1988)
27189. However, I do not agree with the conclusion that the author derives from his
analysis.
Dehon, Les recommendations de la deesse, 273. In Il. 2.297, Odysseus recognizes that
the Achaens are uneasy because even if they do not ght, nevertheless, they would be
ashamed to return empty-handed. In Il. 8.33, Athena knows that her father has decided
on the destruction of the Danaos, but nevertheless, her heart is sad.

Prolegomena to Parmenides Thesis

33

The term tauta (these things, this) in 1.31 resumes the notion of
opinion,114 which will appear in the following line, once again in the plural
in the expression ta` dokounta (what appears in opinions). As W.
Wiersma says, this expression does not refer to ta` phainomena (appearances) but to ha` doke (the things that seem, the things that are
thought)115 among mortals. Heraclitus uses the term in the same sense in
the ironical fragment 28, which may refer both to Homer and Hesiod, both
victims116 of his sharp comments: The most renowned only knows and
stores up dokeonta (opinions).117 Mortals see the world in a certain way
and ta` dokounta is the world as they see it.118 But Parmenides Poem is
didactic: that way of responding to the question about the reality of things
(the being of beings, if you prefer) makes no sense if the truth is known.
Nevertheless, the future philosopher must be alert: if truth were inaccessible, then only opinions would exist. Happily this is not so, and therefore that
possibility is presented in a hypothetical manner, but the temptation to be
carried along by daily inertia (cf. the reference to long habit in fr. 7.3) is
great, and Parmenides also has to expound a probable cosmic order so
that no viewpoint of mortals will prevail over you (fr. 8.6061).
This shows that line 1.31 resumes the content of the preceding line and
that, consequently, it does not introduce a new element, beyond truth and
opinions, as some interpreters have believed.119
The text of 1.31b32 expounds this impossible possibility: opinions are
not true; yet, nevertheless, you will also learn this: how it might have been

114
115

116
117

118
119

Garca Calvo, A., translates, And, all the same, you will have to learn even those (Lecturas Presocraticas [Madrid: Lucina, 1981], 188).
Wiersma, W., Notes on Greek Philosophy (Parm. 1.17, 2.4, 8.61), Mnemosyne 20 (1967)
405. My point of view is that an opionion is a way to think that takes one reality (even
Being) in a subjective way. Appearances has no ontological status in Parmenides philosophy: Parmenides is not Plato . . . The appearance is in thought, not in reality. Appearances are things as they appear in opinions.
Homer is criticized in fragment 56; Hesiod, in fragments 40 and 57.
Guthrie, W. K. C., translates this as opinion (Guthrie [1965], 413); Marcovich, M., translates it as fantasie (or: false opinion) (Eraclito. Frammenti [Florence: La Nuova Italia,
1978], 53); Kahn, C. H., translates it as imagining things (The Art and Thought of Heraclitus [Cambridge, England: Cambridge University Press, 1978], 69); Eggers Lan, C., translates it as las cosas que le parece (Los losofos Presocraticos, I [Madrid: Gredos, 1978],
384); Bernabe, A., translates it as meras creencias (De Tales a Democrito: Fragmentos
presocraticos [Madrid: Alianza, 1988], 135); Lami, A., translates it as credibile sono le
conoscenze . . . (I Presocratici [Milan: Rizzoli, 1991], 209).
Schwabl, H., Sein und Doxa bei Parmenides, Wiener Studien 66 (1953) 401.
For example, Mourelatos (1970), 209. We think, like Taran, that the meaning of the phrase
is . . . the opinions of mortals, despite (allempes) the fact that they are false (Taran
[1965], 211). Cf. also De Rijk, L. M., Did Parmenides Reject the Sensible World? in
Graceful Reason, ed. Gerson, L. P. (Toronto: Pontical Institute of Mediaeval Studies, 1983),
31.

34

(c) You Must Inquire About Everything (1.28)

necessary that the things that appear in opinions really existed, ranging
over everything incessantly. The decisive terms in the passage are the verb
khren120 (would have been necessary) and the adverb dokmos121 (really,
truly). Most scholars believe that dokmos in this context means acceptable,122 plausible, but this is only possible if we go back to the words
etymology, which derives from the verb dokeo (seem, believe). In the
few examples of this term that have been preserved despite the passage of
time (collected in the Liddell, Scott, and Jones Lexicon) it has the meaning
really, truly. It is true that, as P. J. Dehon says,123 few translators have
adopted the meaning proposed by the dictionary, but this is due to a misunderstanding of the passage. These writers have used the following logic:
(1) the passage refers to appearances; (2) Parmenides cannot assert that
appearances really exist; therefore, (3) they do exist, but only apparently,
and only acceptably, and that is the meaning of dokmos. These three stages
are wrong: (1) Parmenides is not referring to appearances but to opinions;
(2) Parmenides does not say that these are real; but holds that (3) they
might have really existed (that is, they might have occupied the place of true,
real knowledge) if truth did not exist. We should not forget that the imperfect khren is a casus irrealis, as W. Kranz124 and R. Falus125 said, because it
alludes to something that might have happened if you do not take the true
thesis into account.126
The hypothetical phrase (starting with hos, namely, that, as, qualifying the notion of learn) containing the imperfect khren, is completed in
the following phrase, in which ta` dokounta (things as they appear in opinions) is the subject, enai (existed) is the verb, and dokmos (really) is
120

121

122

123
124
125
126

The correction khre, proposed by Peyron, A., Empedoclis et Parmenidis fragmenta ex codice
Taurinensis Bibliothecae restituta et illustrata (Leipzig: I. A. G. Weigel, 1810), 55, and accepted by Stahl, J. M. (Kritisch-historischer Syntax des griechischen Verbums der klassischen
Zeit [Heidelberg: C. Winter, 1907], 536), had scant echo.
The possibility dokimos(ai), proposed by Diels (1897), 57 ss., and accepted, among others,
by Untersteiner (1958), clxvii, note 7, is based on an elision (of ai) that is difcult to accept
in a hexameter, as Wilamowitz, U. v., pointed out ( Lesefruchte, Hermes 34 [1899] 204).
For example, Dehon, Les recommendations de la deesse, 286; Taran (1965), 213, note 27;
Bormann (1971), 33; Verdenius (1942), 49; doivent etre en leur apparatre, Cassin (1998),
73.
Dehon, Les recommendations de la deesse, 283.
Kranz (1916), 1170: ta` dokounta . . . Bestand haben mussten.
Falus (1960), 286.
Examples of this imperfect khren are found in Herodotus VII, 9, 25; in Euripides, Hyppolitus, 297; and again in Herodotus, II, 20, 8, in a passage in which the historian eliminates
the unreal hypothesis, according to which the Ethesian winds might be the cause of the
rising of the Nile; if this was the case, then this cause might have been (khren) valid for
other rivers too. Cf. contra Brague, for whom the verb is in the past, that is, in a tense
that expresses that the illusion has been dispelled (Brague, R., La vraisemblance du
faux, in Etudes sur Parmenide, Vol. II, ed. Aubenque, P. [Paris: Vrin, 1987] 59).

Prolegomena to Parmenides Thesis

35

the predicate. That is, with the previous apprenticeship of the rst thesis,
which is the only true one, this second possibility (according to which opinions really exist) makes no sense. However, the previous history of philosophy has in fact fallen into this error. The unreal imperfect khren shows that
this hypothesis is impossible if we accept Parmenides thesis, which takes
it for granted that there is no truth in opinions (fr. 1.30). If we do not know
this truth, opinions would range over everything incessantly (as in fact,
they do: cf. fr. 19: thus these things arose according to opinion [kata` doxan],
and thus they are present now . . .). The participle peronta127 (ranging
over) refers to opinions, and the formula dia` pantos should, I believe, be
interpreted in the temporal sense of alwaysas is the case in Plato128
incessantly, throughout the length of all [times].129 All these elements
have led me to translate lines 3132 as I have: but nevertheless, you will
also learn this: how it might have been necessary that things that appear in
opinions really existed, ranging over everything incessantly.
Some interpreters have seen in these lines the appearance of a third
type of teaching content, apart from truth and opinions. This would be
appearances, which they believe Parmenides also wanted to include. But
we have already seen that ta` dokounta does not mean appearance, but
harks back to the notion of opinions. Moreover, a new content of learning
could not have been presented in a phrase of the adversative-concessive
type, beginning with yet, nevertheless. It is true that the phrase contains
the word also (ka), but this refers to the preceding phrase and not to the
following passage: despite the non-truth of opinions they also have to be
learned. P. J. Dehon showed that the meaning of ka here is adverbial,130
even, and the reason for learning a thought-content that is wrong is explained by Parmenides himself toward the end of fragment 8, where, after
having expounded on the opinions of mortals with reference to a sort of
cosmology, he says that he states (phatzo) this totally plausible (panta eoikota) cosmic order (diakosmon) so that no viewpoint (gnome) of mortals will
prevail over you (parelassei). The plausibility of the discourse can convince
anyone who does not know the truth. Only after traveling along the way
of truth will you have the necessary elements to grasp the falseness of opinions. That is why anyone claiming to know must be abreast of them.
127

128
129

130

Some Simplicius manuscripts offer the reading per onta. Gomez-Lobo adopts this possibility and translates [siendo la totalidad] de las cosas (Gomez-Lobo [1985], 29). A passionate defense of per onta can be found in Brague, La vraisemblance du faux, 4468.
Cf. Statesman 294c8: to` dia` panto`s gignomenon, that which remains always.
This is the meaning of dia` pantos in Sophocles, Ajax, 105; Xenophon, Anabasis, VII, 8, 11;
Herodotus, I, 12, and Thucydides, V.105: we know that the gods reign always (dia` pantos) through the necessity of their nature. The temporal meaning arises particularly from
the combining of panta with dia` pantos, as in Hippocrates, De victu 1,1.
Dehon, Les recommendations de la deesse, 273.

This page has been intentionally left blank.

Chapter III: Parmenides Thesis and

Its Negation
(a) The Alternative in Fragment 2
After allegorically presenting the content of his philosophy and, in particular, after didactically showing that truth can only be reached if thought is
directed correctly (that is, if it pursues a way, a hodos, which means adopting
a met-hodos, that is, a method), Parmenides, through the intermediary of
the Goddess, offers the future philosopher a rigorous presentation of his
thesis. Curiously, this fundamental text was only quoted for the rst time
more than a millennium after it was written by Parmenides, as its oldest
sources are Proclus and Simplicius, two authors from the sixth century
131
A.D. Proclus quotes seven lines of it in his commentary on Platos Timaeus
(I.345, 1827) and repeats lines 5b and 6 in his commentary on the Parmenides (1078, 45). Simplicius only transcribes lines 38. Conventionally, this
text has been considered to be fragment 2, because its content seems to
follow immediately on from the ideas expressed in the last lines of what is
called fragment 1.132 This text begins in the following manner: Well then,
I will tell.133 Let us now see what the Goddess will say.
Just as in fragment 1 Parmenides had shown that anyone who wants
to know cannot advance blindly, chaotically, with no xed direction, the
131
132
133

On this curious fact, cf. my work Lhistoire du texte de Parmenide, in Etudes sur Parmenide, Vol. II, ed. Aubenque, P. (Paris: Vrin, 1987), 324.
However, in the rst editions of the Poem, this text was considered to be fragment 4.
Cf. the work quoted in the preceding note.
The rst words of the text, which I have translated somewhat freely as Well, I . . . (ei
dagego`n), are habitually used by Homer to indicate the continuity of an account (cf. Il.
1.302, 19.108; Od. 1.169, 1.271). With respect to this text, I should point out that the manuscript tradition unanimously offers the version ei dage ton, which does not appear to
make any sense (and which may be based on the existing confusion between t and g
with lower-case transliterating of classical texts, which were written, as we know, in capitals). The conjecture accepted today almost unanimously (with the exception of Vitali
[1977], 33, who proposes ageton ereo and translates Orsu`, io amo le cose vere) belongs
to Karsten (1835), 70, who proposes the resumption of the Homeric formula. If Parmenides really did use the pronoun egon, that would conrm his wish to give his text the
easily memorizable style of the classical epic, since ego`n ereo is a habitual formula in
Homer (cf. Il. 1.76, 1.103, 13.735). If he is not using an archaic cliche, then Parmenides is
using the current form ego: cf. 8.60. However, although Parmenides uses the pan-Hellenic
Ionian of Homer, he sometimes departs from it. Cf. examples in Pieri, A., Parmenide e
la lingua della tradizione epica greca, Studi Italiani di lologia classica 49 (1977) 69, 73.

38

(a) The Alternative in Fragment 2

rigorous presentation of the thesis we shall nd in this fragment 2 will take

the form of a way, along which anyone wanting to reach the heart of truth
must direct his or her thought. As I have said, Parmenides is the rst philosopher who argues, who gives reasons in support of his thesis. Later philosophy will coin the formula logon didonai to express this idea of giving the
reason for something, and Parmenides himself will use the term logos in
fragment 7.5 to mean reasoning. A reasoning is not a thought; it is a
series of thoughts related in a certain way. But this series of thoughts requires an ordered, methodical succession. That is why Parmenides proposes to lead thoughts, that is, create a reasoning from them. The term
logos alludes precisely to this ordered and coherent series of thoughts, expressed in a discourse.
Let us return to the speech the Goddess presents at the beginning of
fragment 2. It will be presenting a thesis, but the Goddesss presentation of
this thesis will be double. This procedure is habitual in Parmenides: the
truth of a notion is reinforced when the falsehood (or impossibility) of the
contrary notion is shown. In fragment 2 we nd the repetition of the general scheme presented in lines 2830 of fragment 1, which stated the necessity of inquiring both about the heart of truth and the opinions of mortals,
which were not true (because there is no true conviction in them). This
alternative between truth and non-truth reappears when, after allegorically
presenting his teaching, Parmenides puts into the mouth of the Goddess
the rigorous exposition of a true course of philosophy. The Goddess will
present a method (or, as we saw, a way), that is persuasive and convincing (because it goes with the truth). At the same time she will expound a
completely unknowable way, which will be described as not true in line
18 of fragment 8. The future philosopher must inquire about the content of
both ways, and once he is in possession of the elements that will make
possible his decision (fr. 8.15), he will choose the single way that will enable
him to reason, that is, direct his thought. This way is Parmenides thesis.
Just as in line 28 of fragment 1 the Goddess began her speech with an
exhortation about the necessity of inquiring into everything (it is necessary), in the rst line of fragment 2 she addresses her listener with an
imperative accent: you, who listens (akousas), receive (komisai) my word
(muthon).134 The Goddess invites the listener to make the muthos that she
134

The accusative muthon can relate both to komisai and akousas. The expression muthon
akousas is common in Homer (Il. 17.694, Od. 3.94, 4.597), but in the passage we are concerned with muthon appears to be the object of komisai. Collobert shares this viewpoint:
et toi ecoutant, acueille ma parole (Collobert [1993], 13); as does Couloubaritsis: et toi,
mayant ecoute, prend soin de ma parole (Couloubaritsis [1990], 370); Vitali links listen with muthos (ascoltando il discorso, Vitali [1977], 32), and the great majority of
interpreters, without taking into account the difference between the conjugated verb kom-

Parmenides Thesis and Its Negation

39

proclaims his own. The verb I have translated as receive (komisai) has the
meaning of look after, make your own, preserve something by taking
it with you.135 That is to say, the teaching that will be offered must form
part of the disciples intellectual baggage. Henceforth he will not be able to
judge without using the Goddesss muthos as a reference point.
What does muthos mean in line 1 of fragment 2?136 Whatever the verb
on which this term depends (listen or receive; my preference inclines
toward the latter), muthos simply means account, word, or even discourse.137 It is interesting to point out that all the wealth implicit in this
single word (is) will be expounded in what follows in an extensive . . .
logos. That is to say, there is no fundamental distinction between muthos
and logos in Parmenides either.

(b) The Only Two Ways of Leading Thought

The Goddesss account expounds the only two ways of investigation there
are to think (fr. 2.2). This phrase deserves an explanation. The term only
(mounai) cannot be relativized. Nevertheless, most scholars of Parmenides
thought have not taken it literally. The Prologue itself of the Poem presented two ways in allegorical form (ways of day and night), and toward
the end, the Goddess spoke of truth as a way far from that of mortals (fr.
1.27). Be that as it may, only two possibilities emerged from this presentation. When Parmenides resumes the subject in fragment 2, before describing
the content of each way, he says that these are the only (mounai) ways
that exist for thought. Parmenides could not have written only by chance;
the numeral adjective indicates with precision that it would be illusory to
try to seek other ways apart from the two that will be presented in what
follows in lines 2.3 and 2.5.
Even so, there are scholars who underrate the force of this term and
interpret it in a relative sense. A symptomatic case is that of M. Conche,
who rejects my literal reading of the text:138 certainly, these two ways are
the only legitimate ones, a priori, the only ones that rightly exist; but that

135
136
137

138

isai and the participle akousas, make muthos the object of both terms (tu preserva el
relato, despues de escucharlo, Gomez-Lobo [1985], 57; pay attention to the account
when you have heard it, Taran [1965], 32; mais toi, charge-toi du recit que tu auras
entendu, Cassin [1998], 77).
Cf. Il. 1.594, 8.284, 6.278.
An exhaustive analysis of this question can be found in Couloubaritsis (1990), passim.
As is well known, the distinction between muthos and logos in early Greek thought is an
invention of historians of philosophy: both terms means exactly the same. Cf. Vernant,
Raisons du mythe, in Mythe et societe en Gre`ce ancienne (Paris: Maspero, 1974).
Conche (1996), 76.

40

(b) The Only Two Ways of Leading Thought

does not prevent the fact that others can exist, and do exist.139 If that is so,
these will be ways invented by interpreters: Parmenides is innocent.
Various authors before Conche had maintained that really Parmenides
was speaking here of (1) the only two ways of investigation, or (2) the
only two ways that are thinkable (or that it is possible to think, eisi is
given potential value). Let us look at both possibilities. (1) If it is a matter
of the only ways of investigation, say the partisans of this position, there
is nothing to stop another way appearing later that is not suitable for investigation. This would refer to the way formulated in fragments 6 and 7. It is
true that in fragment 6 it only says that this way was created by men, who
know nothing (fr. 6.4), but the description of this way continues in fragment 7, and there Parmenides says that this way of investigation (fr. 7.2)
is to be avoided. So this way of investigation has to be one of the only
two presented in fragment 2.
Case (2) is even more debatable. In Parmenides philosophy, a way of
investigation that, at least a priori, is not thinkable (if we base ourselves
on those who give the verb noesai a passive sense), would not even have
been presented. Even the way that is condemned in line 8.17 as unthinkable was one of the ways offered to thought in fragment 2, as the term
reappears at the beginning of fragment 8, once reasoning has suppressed
one of the ways of investigation, and on that occasion the Goddess declared
that so there remains one single muthos of the way: is. Moreover, those
who see a new way appearing in fragment 6 (a way that is not thinkable)
admit that the two thinkable ways were already presented in fragment
2, but as it happens, one of these two ways is already described in fragment
2 as completely unknowable,140 as it tries to assert that it is necessary not
to be.
Regarding the relation between the two ways and the notion of think,
we generally witness a wrong interpretation of the term noesai in fragment
2.2. The verb think (noesai) is evidently a nal or consecutive innitive,141
but it has always been read as if it had passive value, either directly or as
if eis was interpreted as having potential value. Consequently, it has
139
140
141

Conche (1996), 76.

On the equivalence in Parmenides between think (noen) and know (gignoskein), cf.
Mansfeld (1964), 57, note 1.
Cf. Kahn (1968/69), 703: for knowing. Aalto maintains that this nuance of the innitive,
assimilated by him to the dative, has its origin in certain forms of Indo-European: ji-se,
for example, is equivalent to zu siegen (Aalto, Studien zur Geschichte des Innitivs im
Griechischen [Helsinki: P. Katara, 1953], 11). Constantineau rejects the nal value of noesai
because the aorist has a temporal and not an absolute meaning, and, according to him,
one of the ways only has an ephemeral existence (Constantineau, La verite chez Parmenide, Phoenix 41 [1987] 220). The same argument is valid for the ephemeral character of
a way that can only direct thought a priori.

Parmenides Thesis and Its Negation

41

generally been translated as a passive participle. This is what happened

with translations of this type: welche Wege der Forschung allein denkbar
sind,142 the only ways of inquiry that can be conceived,143 quali sole vie
di ricerca siano logicamente pensabili,144 pensables,145 conceivable,146
a` penser,147 da pensare, pensabili,148 zu bedenken.149 Nevertheless, H.
Gomperz already related the innitivewith active valueto an indenite
subject: (Strassen) die man erkennend betritt,150 whereas Mourelatos proposed a long list of potential subjects: logically speaking, the subject of the
innitive is something like you or men or mortals [ . . . ] and an object
such as ti or khema or even eon is implicit,151 which was interpreted by J.
Jantzen: welche Wege es fur dich allein sind, etwas zu denken.152 I prefer
to leave both the subject and the object of the innitive indenite, because
this is not a verb inserted into a sentence but an isolated innitive, and I
reject any passive nuance, which, as we saw, would be incompatible with
one of the ways.
Therefore, the ways proposed in line 2.2 are ways to think, that is,
ways along which, a priori, you can direct thought,153 just as you use a knife
to cut (without it being cuttable itself) or you use a chair to sit on
(without it being sittable itself). An almost literal equivalent of the expression existing ways to think (eisi noesai) can be found in Empedocles,
poros esti noesai (fr. 3). The text asserts the role of the senses as organs of
knowledge, since in each of them there is a pore that enables us to know.
Clearly it is not the pore itself that is knowable. The pore is the way
through which we know something.
Moreover, the nal character of the innitive noesai is the only way of
explaining the four declarative conjunctions (hopos once and hos three
times), which introduce lines 3 and 5 of fragment 2. In fact, in these lines,
142
143
144
145
146
147
148
149
150
151

152
153

Diels (1897), 33.

Taran (1965), 32.
Untersteiner (1958), 129.
Gomez-Lobo (1985), 57.
Coxon (1986), 52.
Conche (1996), 75; Cassin (1998), 77.
Cerri, G., Parmenide di Elea: Poema sulla natura (Milan: Rizzoli, 1999), 188.
Wiesner (1996), 251.
Gomperz, H., Psychologische Beobachtungen an griechischen Philosophen, Imago 10
(1924) 7, note 19.
Mourelatos (1970), 55, note 26. In fact, only you deserves to be kept, because the Goddess is not speaking for men or mortals, but only for the youth who has made a
journey in order to listen to her.
Jantzen, J., Parmenides zum Verhaltnis von Sprache und Wirklichkeit (Zetemata H. 63) (Munich: C. H. Bechsche Verlagsbuchhandlung, 1976), 117.
So thinking is an activity to carry out in the future. Cf. Chantraine: the aorist innitive
often refers to the future (Chantraine, Grammaire homerique, II: Syntaxe [Paris: Kliencksieck, 1953], 189).

42

(c) Lines 3 and 5 of Fragment 2

the content of the two ways of thinking is expounded (one being valid and
the other wrong). As both contents depend on the term way, interpreters
who give noesai a passive value (thinkable) have to include an understood verb on which to make the declarative conjunction depend. Thus we
nd translations such as: the one <says> . . .;154 lun (dit) . . .;155 Weg,
welcher besagt . . ..156 The way does not speak. The way is a way of thinking,157 and when you think you think that . . . The declarative conjunctions
correspond to the activity of thinking proposed by each way: on one hand
there is a (way to think) that . . .; and on the other hand, there is another
(way to think) that . . .158 This is normal, since there are only two single ways
to think, and each corresponds to the two contents of thought expounded
toward the end of the rst fragment: the heart of truth, and the opinions of
mortals. One of the two ways of thinking is valid and fruitful, whereas the
other will be revealed as inadequate, sterile, closed, and, nally, unthinkable (fr. 8.17), because it is based on . . . nothing. Nevertheless, a priori, an
investigation can be based on either way. Both are ways of investigation
(dizesios) (fr. 2.2). Of investigation is subjective genitive: investigation has
two ways available to it. Once it has been demonstrated that one of the
ways is not viable, since the conclusion has been reached, a posteriori, that
it was not the true way, then the Goddess orders: Withdraw thought from
this way of investigation (fr. 7.2).

(c) Lines 3 and 5 of Fragment 2

The only two ways of investigation are stated in lines 3 and 5 of fragment
2 (line 4 is a commentary on the rst way and lines 68 show why the
second way is impracticable). Each formulation is preceded by the particles
he me`n (on the one hand, in the rst place) for the rst and he de
(on the other hand, and then) for the second. Both formulas, customary
in the case of an enumeration, resume fragment 1s double program of
study: on the one hand (emen), truth; on the other hand (ede`), opinions
(fr. 1.2930). Therefore the reader/listener presumes that the two formulas
in fragment 2 coincide with those in fragment 1, but doubtless expects that
154
155
156
157
158

Taran (1965), 32.

Couloubaritsis (1990), 370.
Wiesner (1996), 251.
The English term way, which means both way as road and way as manner, is a
literal translation of hodos.
Gunther translates thus: der eine Weg ist (besteht darin), zu denken, dass . . . (Gunther,
H. C., Der Satz des Parmenides von der Identitat von Denken und Sein, Studi Italiani
di lologia classica 15(2) (1997) 166).

Parmenides Thesis and Its Negation

43

now that the philosophy course has begun, the Goddess will go more
deeply into both possibilities; indeed she will say what is truth and what
are opinions, or if you prefer, what is the heart (the nucleus, the perhaps
hidden core) of both. The Goddess will amply fulll these expectations.
Despite the parallelism that I have tried to demonstrate between lines
1.2930 and 2.3 and 2.5, there is a difference. In the rst fragment the expressions eme`n and ede` are conjunctions, that is, neutral terms with
respect to the content that follows; anything at all can be stated on the one
hand and on the other, and it is the context that gives the passage its
meaning. In contrast, in fragment 2, the formulas he me`n and he de are
made up of a relative (he, feminine singular) followed by the particles
me`n and de`. This means that in fragment 2, lines 3 and 5 presuppose a
feminine subject, a subject about which something is said on the one hand
and something else on the other. The only feminine subject provided by
the passage is way (hodos is feminine in Greek), but this term cannot be
a candidate to be taken up by the relatives (cf. infra, note 31). Both lines 2.3
and 2.5 present ways. But as I have said, both lines begin with declarative
conjunctions: hopos159 (2.3) and hos (2.5). This means that each way is a
way to think that . . .
The content of each thought is expressed through a double phrase,
made up of two coordinated formulas which, from now on, we shall call
hemistiches. The rst way, expounded in line 2.3, is a way to think that
A and that B; the second way, expounded in line 2.5 is, as we said, the
negation of this rst way: it is a way to think that not-A and that not-B.
In Greek, A is represented by estin160 and not-A by ouk estin; B is represented
by ouk esti me` enai and not-B (as I shall show, because it is not clear) by
khreon esti me` enai. The rst way, which thinks that estin te ka ouk esti
me einai is a way that is accompanied by truth, it is (we could say) true.
It is the only way possible, which, once it has been explained in the following fragments, will be Parmenides thesis. The second way that thinks exactly the opposite, viz. ouk estin te ka` khreon esti me` enai, will be considered by the Goddess as the negative aspect of the thesis, and for this reason
it cannot even be approached as a viable way, because thought cannot direct itself along it.
In the didactic scheme I set out above, the rst way of investigation
thinks, rst, that estin (A in our scheme) and then also that ouk esti me`
159
160

A different interpretation, according to which this hopos has a modal meaning, has been
maintained by Untersteiner (1958), lxxxv.
The original texts have esti. These were changed to estin by Mullach in order to
respect the meter (Mullach, F. G. A., Fragmenta Philosophorum Graecorum [Paris: Didot,
1860], 113). Although this modication is not important, Diels wrongly attributes it to H.
Stein (Dielss edition of Simplicius, In Phys., 116).

44

(c) Lines 3 and 5 of Fragment 2

enai (= B). As we know, estin is the third person singular of the present
of the verb to be, enai. This third person appears either on its own or
negated, in the two rst hemistiches of each way. Let us begin by analyzing
these rst hemistiches.

(1) Estin on Its Own and Its Negation

First of all, we should say that A can be translated literally as is, and
not-A as is not. When we analyze the value of the verb to be in
Parmenides philosophy, which will be the central point of our quest, we
shall see what Parmenides meant by saying is or is not. However, at
once we come to a difculty: a simple reading of the text upon which we
wish to comment shows us that Parmenides says is in one line and is
not in the other (and we shall see with what intention), but he does not
tell us who or what is or is not.
This means the verb is (in the afrmative or negative) has no subject.
It is true that Greek, like various modern languages, does not require an
explicit subject in the third person (and other languages require it as a
purely formal subject, but to say it is is not saying anything; indeed, to
understand the meaning of the sentence, we have to know what or
who is the it that is), but, in parallel passages, the preceding context
usually indicates what is the subject in question. Parmenides truncated
phrase (hos/hopos estin) is also found in other authors, but in all the parallel passages the subject is made explicit by the context. This is the case with
Protagoras famous statement man is the measure of all things, which
continues in the following way: of the things that are (ton onton), that are
(hos estin) . . . The phrase that are, identical to the one in Parmenides
(since it is in the singular in Greek) has no subject, but the subject has
already been mentioned and we know that it is the things that are. The
same occurs in the famous passage in the Sophist (263b) in which Plato says
that true speech tells the things (ta` onta) that are [or as they are] (hos estin)
about you. The subject of hos estin (once again, the phrase is identical to
that of Parmenides) is undoubtedly the things.
In the case of Parmenides, nothing indicates or suggests what the possible subject is.161 Various explanations can be proposed to give a reason for
this unusual fact. But it occurs to us that there is a previous question to
answer that is generally overlooked, and which is essential: if a subject is
proposed (or found) for the esti that occurs in both 2.3a and 2.5a, this subject
161

Untersteiner asserts that the subject is the way (Untersteiner [1958], lxxxvi), but this
position is incompatible with the properties (semata) that fragment 8 will deduce from
the so-called subject.

(1) Estin on Its Own and Its Negation

45

must be the same for both occurrences of the verb. This is obvious even if the
two ways that are stated in lines 2.3 and 2.5 are opposite ways. The only
way to respect the value of the opposites that the Goddess propounds (one
way is accompanied by truth, whereas the other is completely unknowable)
is to suggest or keep the same subject for both of them. This is the position
adopted by, among others, A. Finkelberg, for whom line 2.3 states that to
be is . . . , whereas line 2.5 says that to be is not . . . 162 Thus, when something appropriate is said about this subject, the way is true; when something incorrect is said about the same subject, the way is wrong. In contrast,
if opposing subjects are proposed, the value judgments about each way
would have to be identical. On this point, J. Mansfeld says that to` eon could
be the subject of esti, but it is mistaken to assert that to` me` eon could be
the subject of line 2.5.. If that were so, 2.5a would say [to` me` eon] ouk
esti, that is, that which is not, is not, a thesis that cannot be described as
wrong, as we read in the following line (2.6), where it repeats almost word
for word the second hemistich of line 2.3: it is not possible not to be,
there is no non-being. In contrast, if the subject of 2.5a were the same as
that of 2.3a, 2.5a would state: [to` eon] ouk esti, that is, that which is, is
not, an absurd and aberrant thesis that Parmenides criticizes. Montero
Moliner fell into the same error as Mansfeld when he wrote that the rst
way states that the being [beings] are and the second that the non-being
[non-beings] are not (or not-being is not).163 The same happens with Coxon,
who states that the only possible subject of is not and must needs not be
is Nothing.164 J. Wiesner also commits a similar error when he says that
the second way states that Nichts gibt es nicht.165 In that case Parmenides would be saying that nothing or non-being is not, and that they necessarily are not. If he were saying that, how could it be said that that thesis
is not true?
For his part, L. Taran criticizes those who assume, as I do, that both
ways must have the same subject (if there is one), because, according to
him, in that case, Parmenides would have had to demonstrate why one
way is valid and the other not, whereas the absence of a subject makes that
demonstration unnecessary.166 To this it can be answered that within the
limits of his logic Parmenides at least demonstrates the impossibility of
the wrong way, and that this impossibility assumes the possibility (and
even the necessity) of the true way.
162
163
164
165
166

Finkelberg, A., Parmenides Foundation of the Way of Truth, Oxford Studies in Ancient
Philosophy 6 (1988) 47.
Montero Moliner, F., Parmenides (Madrid: Gredos, 1960), 68.
Coxon (1986), 182.
Wiesner (1996), 177.
Taran (1965), 38. Cf. contra Mansfeld, J., review of Taran, Mnemosyne 20 (1967) 317.

46

(c) Lines 3 and 5 of Fragment 2

Let us return to the explanations proposed to resolve the problem of

the subject of estin.167 I think there are four possibilities: it could be suggested (1) that there is an error in the transmission of the text (if this is so,
it must be corrected to introduce the absent subject); (2) that there is an
implicit conceptual subject that has to be sought in the rest of the Poem; (3)
that there is no possible subject; and (4) that the subject must be extracted
from the isolated verb: estin produces its own subject. All four of these
possibilities have had their champions.168
(1) F. M. Cornford held that now at this stage of the text the subject is
to` eon, and stated that the absence of a subject for estin suggests that Parmenides wrote he me`n hopos eon estin.169 His thesis is based on the similarity there is between 2.3 and line 6.1, which, effectively, reads eon emmenai.170 However, this analogy does not authorize the modication of a
passage situated in a different context and, especially, at a previous stage of
the argument. J. Loenen also proposed modifying the Parmenides text and
adding the subject ti (something).171 According to his correction, hemistich 2.3a would say estin ti. According to this scholar, the proposed correction is conrmed by an analogous expression that appears in the reading
some manuscripts proposed for 8.1: de ti, instead of the generally accepted text, deti. Moreover, Melissus (fr. 1) and Gorgias (fr. 3) also spoke
about ti, but J. Mansfeld correctly observed that in the case of those scholars, the presence of ti is also conjectural.172 To sum up, I consider it very
risky to introduce modications into a text (2.3a) that has come down to us
in identical form, both in Proclus and Simplicius versions.
(2) Most scholars adopted this second possibility and asked the following question: about what could Parmenides say estin in line 2.3a and ouk
estin in 2.5a? The favorite candidates were eon and enai, generally and
literally translated as that which is being, or that which is, or being,
considered as synonymous.173 Thus, for example, Diels translates: dass
(das Seiende) ist;174 O. Becker, dass es ein Sein gibt;175 F. M. Riaux, que

167

168
169
170
171
172
173
174
175

In tackling the thorny subject of the subject of Parmenides thesis, we have to differentiate between the problem of the grammatical subject, absent from lines 2.3 and 2.5 (about
whose absence various hypotheses can be proposed) and the subject around which the
whole Poem revolves, which is obsessively present: [to`] eon, meaning that which is
being, the fact of being. Cf. infra.
More lavishly, Meijer proposes twelve possibilities (Meijer [1997], 114).
Cornford (1939), 30, note 2.
Emmenai is the epic form of the innitive enai, and the formula means that which is,
is, or that which is being, exists, or by being, it is.
Loenen (1959), 12.
Mansfeld (1964), 52, note 2.
Cf. Taran (1965), 37.
Diels (1897), 33.
Becker (1937), 141.

(1) Estin on Its Own and Its Negation

47

letre est.176 Some scholars postulate the pronoun it (which may be il,
el, or es) as the subject, but once more, its antecedent is being or the
being.
This possibility attracted most researchers because it seems evident that
the rst way of investigation states, as its central thesis, that being is or
there is being, that being exists, as K. Reinhardt stressed strongly.177 R.
Mondolfo nds an indirect proof of the existence of this subject in lines
2.78, where it is stated that me` eon (that which is not) is unknowable
and inexpressible. The knowable and expressible would then be the contrary of that negation, that is to say, eon, which is then conrmed in 6.1:
It is necessary to say and to think that that which is, is.178 Continuing
with this viewpoint, we could also cite the case of line 8.3, since that which
is (eon) is unbegotten and indestructible, where it is difcult to deny that
the subject is eon; and even more clearly in 8.19, how then could to` eon
perish?; and in 8.25, eon touches eon. In this latter case, if we consider
that eon, without article, is not the subject but a participle referring to
another subject, then this tacit subject would touch eon but what could
touch that which is, except that which is? Let us not forget that eon is
unique; only eon can be the subject of touches, since it is impossible
to force that which is (eon) not to be connected with that which is (eon)
(fr. 4.2).
One thing is clear: all these examples show something obvious, that is,
that Parmenides Poem, and especially the characteristics of fragment 8,
concern that which is being (eon). But the partisans of this possibility
do not explain why, in certain passages (especially at the beginning of his
exposition, in fragment 2, and when he returns to the single remaining way
again, in line 8.2), that subject does not appear. Given this certainty, we can
only share the opinions of R. Falus: the subject eon may complete the thesis
esti;179 and of G. E. L. Owen: no one will deny that, as the argument
goes, to` eon is a correct description of the subject.180 Thus we arrive at a
position that admits the existence of eon as subject, but only at later stages
than fragment 2. For example, J. Mansfeld maintains that in 2.3a and 2.5a
there is no subject because here we have the premises of a disjunctive syllogism that is valid for statements (Aussagen) but not for concepts (Begriffen),181 and that any possible subject would have foreseen the conclusion

176
177
178
179
180
181

Riaux (1840), 209; cf. also Robin, L., La pensee grecque et les origines de lesprit scientique,
3rd ed. (Paris: La Renaissance du Livre, 1963), 103.
Reinhardt (1916), 36.
Mondolfo, R. Discusioni su un testo parmenideo (fr. 8.56), Rivista critica di storia della
losoa 19 (1964) 311.
Falus (1960), 274, note 30.
Owen (1960), 90.
Mansfeld (1964), 58.

48

(c) Lines 3 and 5 of Fragment 2

of the syllogism in advance. Nevertheless, still according to Mansfeld, fragments 3 and 8.4 ff. authorize the introduction of the subject, so that es ist
means das absolut Seiende ist.182 On the same tack, we may include here
an apparently independentand clearly originalthesis to which I have
already alluded. This is the point of view of M. Untersteiner, for whom the
subject is the way, but this, in its turn, is an eon, since the reality of this
way will become the reality of eon, so that we reach a total confusion
between the hodo`s alethes (= the true way) and eon itself.183 However that
may be, there can be no doubt that it is difcult to do without the concept
of eon, even if we agree to leave the esti of fragment 2 standing on its own.
Other scholars prefer to propose a general subject, either abstract or
concrete. The former is the case with S. Tugwell, for example, for whom
the alternative in fragment 2 takes this form: that which can be known,
must exist, or not.184 This position is shared by Owen and a large number
of Anglo-Saxon researchers, for whom the subject is that whereof one may
speak or think.185 For his part, C. H. Kahn states that the rst way does
not have a grammatical subject, but does have a logical one: the object of
knowing, what is or can be known.186 Among those authors who have
proposed a general, but concrete, subject, we nd L. Woodbury, who maintained that the subject is the real world, which embraces everything
about which we can say estin, since Being is the name of the world.187
Relying on a reading by J. Burnet, for whom the subject is what we call
Body,188 and on the interest shown by Parmenides in cosmological questions,189 Y. Lafrance says that the subject is the material universe.190 T. M.

182
183
184
185

186
187
188
189

190

Ibid., 45.
Untersteiner (1958), lxxxix. Casertano shares this hypothesis. For him, Parmenides only
says that there exists one way (Casertano [1978], 63).
Tugwell, S., The Way of Truth, Classical Quarterly 14 (1964), 36.
Owen (1960), 95. Nevertheless, the way followed by Owen is curious, since he arrives at
his interpretation through an analysis of 2.7, where it says that to` me` eon is unthinkable
and inexpressible. If that is so, Owen notwithstanding, that which is thinkable and expressible (which would be the subject) is . . . eon.
Kahn (1968/69), 710.
Woodbury (1958), 152. Also Casertano (1978), 94, cio` che e`, e` il mondo. Wiesner (1996),
232, shares this position: the Poem is about the Welt.
Burnet, Early Greek Philosophy, 4a ed. (London: A. & C. Black, 1930), 178.
A cosmological Parmenides monopolizes the work of Popper, K., The World of Parmenides (London/New York: Routledge, 1998), 114: So, Parmenides was, in my opinion,
essentially a cosmologist. Cf. also Bollack, J., La cosmologie parmenidienne de Parmenide, in Hermeneutique et ontologie, Hommage a` Pierre Aubenque, ed. Brague, R., and Courtine, J. F. (Paris : P. U. F., 1990), 1753.
Lafrance, Y. Le sujet du Poe`me de Parmenide: Letre ou lunivers? Elenchos 20(2) (1999)
302. According to this scholar, the inscription discovered in 1962 in Elea stating that Parmenides was a doctor (ouliades) demonstrates, according to one possible etymology of
the term, that the philosopher was concerned with the whole, the oulon (268).

(1) Estin on Its Own and Its Negation

49

Robinson adopts the expression all that is collectively real,191 a position

perhaps inspired by W. J. Verdenius, for whom the subject of estin is so
obvious [that it] is not stated,192 and is none other than Reality, that is, the
totality of things. This position has the merit of bringing out the breadth of
Parmenides thesis, but it is more a commentary on the Poem than a decision on the absent subject. Literally, Verdeniuss position would be unsustainable, since it would assume that reality is different from being, since
being refers to reality, instead of being identical with it.193 A few years
later, Verdenius abandoned this opinion and wrote: if Truth is the subject
of the Goddesss discourse, it is by implication the subject of estin194 insofar
as truth is the real nature of things. With this new viewpoint, Verdenius
was perhaps pointing to a more primordial concept than reality; if that is
so, he could continue his research further, since there is an even more primary notion, the heart of truth, that the Goddess exhorts the seeker to
know rst and foremost (fr. 1.29). Moreover, in 2.4 Parmenides says that
truth is accompanied (or escorted: opede ) by the way announced in 2.3, but
that it is not identical with this way or with any part of it.195 Finally, we
may add the most general hypothesis concerning the subject of estin, which,
while acknowledging a certain debt to Owen and Kahn, is much more generous, since it holds that the verb refers to any object. This is the interpretation of A. Gomez-Lobo, for whom Parmenides is saying that there are
two and only two ways for the investigation of any object; one assumes
that it is, and the other assumes that it is not.196 J. E. Raven197 and W. K. C.
Guthrie198 had said something similar.
(3) There are scholars who have said that there is no subject in 2.3a and
2.5a. All the proposed interpretations of this kind follow G. Calogeros
view, who, from an analysis of line 8.34, wrote esti does not have a dened
subject, as the specic expression of a given reality, but keeps the indeterminacy of a purely logical and verbal element of afrmation.199 Moreover,
191
192
193
194
195

196
197

198
199

Robinson, T. M., Parmenides on the Ascertainment of Real, Canadian Journal of Philosophy 4 (1975) 56.
Verdenius (1942), 32.
Frankel (1930), 162.
Verdenius, W. J., Parmenides B 2.3, Mnemosyne 15 (1962) 237.
Even Wiersma, one of the very few scholars who do not adopt Bywaters conjecture
aletheei in line 2.4 (he conserves alethee) states that truth accompanies this way
(Wiersma, W., Notes on Greek Philosophy, Mnemosyne 20 [1967] 407).
Gomez-Lobo (1985), 68.
For Raven there is no dened subject, since one can say it is or it is not about everything
(Kirk, G. S., and Raven, J. E., The Presocratic Philosophers, 1st ed. [Cambridge, England:
Cambridge University Press, 1957], 270).
According to this scholar, estin is the belief in the existence of something, to which the
word estin fully refers (Guthrie [1965], 22).
Calogero (1932), 18.

50

(c) Lines 3 and 5 of Fragment 2

it would be inconsistent to state that eon is the tacit subject, since eon is
actually Parmenides invention, and so, it would have been difcult for a
reader of his own time to assume it.200 This is the reason why anyone who
does not want to make Parmenides an idiot (sciocco), and Plato another
idiot for considering Parmenides to be venerable and fearsome, must give
up stating that the subject of estin and of ouk estin is the implicit eon.201 The
way of truth says is; the way of error says is not; and the two formulas
translate the two ways belonging to the logical-verbal process. This
means that for Parmenides, being is being as in the verbal copula, and that only
a confusion between the predicative value and the existential value of the
verb could have led Parmenides to state, for example, that being exists.
A. P. D. Mourelatoss position is, if anything, even more extreme: estin
is just a certain way of linking any subject to any predicate, and therefore
the structure of Parmenides thesis is the following: . . . is . . . Mourelatos
calls the scheme speculative predication.202
Calogeros thesis was revolutionary in his time,203 but an interpretation
that, from my point of view, intolerably weakens Parmenides estin aroused
very violent criticisms. If we take into account the rigorous analysis of being that is developed in fragment 8, we may well ask whether, contrary to
Calogeros thesis, the predicative value that he maintains could not be a
usage derived from a deeper reality: the absolute and necessary value of
being. We shall return to this point, but we can say now that beings semata,
expounded in fragment 8, cannot belong to a mere formal and empty estin.
H. Frankel also held that estin has no subject, but his arguments differ
from Calogeros. For Frankel it is an impersonal verb, like rain or snow,
and if you try to add a possible subject (e.g., the rain or the snow) you
fall into a tautology: the rain rains,204 or even worse, you introduce a
factor of confusion by suggesting that anything else except rain could
rain. The idea is interesting, but it rests, I believe, on an erroneous concept of so-called impersonal205 verbs. Furthermore, Frankel appears to
200
201
202

203

204
205

In note 194 we saw that Verdenius had said exactly the opposite.
Calogero (1936), 155.
Mourelatos (1970), 56 ff. Later, Mourelatos made his position a little more exible: the
copula is really a conveyor toward the predicate, which is approached as a characterpower (Heraclitus, Parmenides, and the Naive Metaphysics of Things, Phronesis, Supp.
I, Exegesis and Argument, Studies in Greek Philosophy presented to G. Vlastos, ed. Lee, E.
N., Mourelatos, A. P. D., and Rorty, R. M. [Assen: 1973], 43).
His inuence is detectable in W. Kranzs translation, dass IST ist (in Diels, H., and
Kranz, W., Die Fragmente der Vorsokratiker, Vol. I [Zurich-Berlin: Weidmannsche Verlagsbuchhandlung, 1956], 231). However, Calogero criticized Kranzs version, stating that he
to disguise such eccentricities has to resort to typographical ploys (Calogero [1936],
155, note 1).
Frankel (1951), 403, note 13.
I will set out my position on this point infra.

(1) Estin on Its Own and Its Negation

51

apply his position to the whole of the Poem, and not just to the passages
in which the verb appears without a subject, since, according to him, any
identication of a subject would determine being in an inadmissible
way.206 This is not true. In 6.1 Parmenides says tautologically that that
which is being, is (eon emmenai), and in other passages the verb to be is
used in an undoubtedly predicative way, which requires a subject; this
would not be possible if, as Frankel says, impersonal verbs cannot have a
subject, even a conceptual or psychological one.207
Halfway between the positions of Calogero and Frankel, we nd L.
Taran. He states that 2.3a and 2.5a do not have any subject and that the
verb estin, in the third person, signies the notion of existence in these
passages.208 In other passages, the same notion is expressed by a participle
or by an innitive. That is, like Calogero and Frankel, Taran considers that
estin is used in an impersonal manner, but, contrary to Frankel, he limits
this use to fragment 2, and contrary to Calogero, he gives the verb an existential value and not just a predicative one. In contrast, the rest of his interpretation oats on a highly hypothetical plane, because he bases it solely
on the fact that Greek, unlike English, admits the existential as well as the
copulative impersonal without an expressed subject.
I adopt explanation (4). It cannot be denied that Parmenides estin has
a subject, because it appears explicitly in various passages of the Poem. I
have already cited 6.1, eon emmenai, and 8.19; we may add 8.3637: nothing estin or estai [is or will be] apart from to` eon. But the fact that there is
a subject does not mean that this must be already postulated in 2.3a and
2.5a. Parmenides starting point is estin because the philosopher wants to
give pride of place to an undeniable certainty (the malign genius of Descartes had not yet been born): now, in the present, at this very moment,
is. It does not matter who or what is, but no one, much less a future
philosopher, can be unaware that is being. Parmenides will draw a series
of consequences from this sort of intuition, but the best way of giving pride
of place to the imperious and present character of this fact consists in
presenting it on its own. In any predicative sentence, the predicate claries,
informs about, or characterizes the subject, and the subject is the central
nucleus of meaning. Write remains empty of meaning if we do not know
who is writing, and when we add the subject, for example, Borges, we
know something . . . about Borges. If Parmenides had made a subject explicit from the beginning of his philosophy course, it would have been said

206
207
208

Frankel (1951), 403.

For the difference between a conceptual or psychological subject and a grammatical subject, cf. Brugmann, K., Griechische Grammatik (Munich: C. H. Beck, 1913), 656, note 1.
Taran (1965), 37.

52

(c) Lines 3 and 5 of Fragment 2

about this subject that it is. Parmenides starts from is and shows that
that is hides a broader, richer notion within itself, which will be reached
only after having grasped the present and undeniable force of is. For
Kahn, partisan of the veridical meaning of enai in Parmenides, the estin
is just a point of departure, from which Parmenides develops other aspects
of the ontological claim entailed by this assertion, and among these aspects, fundamentally, we nd the existential nuance, according to which
what is must be something rather than nothing.209 That broader notion
will be the subject of is, the only possible subject, that is, a sort of
product extracted analytically from the predicate. As B. Cassin luminously
writes, the verb has no other possible subject than itself, which unfolds,
segregates itself as subject: that which is, is being.210
Indeed, what else can be unless it is the fact of being? The fact of
being (which, as I shall show, is expressed in Parmenides both by the participle eonvery rarely, to` eonand by the innitive enai) is the only notion
whose reality is dened by stating it through the conjugated form is. That
is also grammatically only denotes being in the present tense, just as
is writing denotes that someone is writing now, that at this moment, the
fact of writing is happening, even though when we say is writing we
do not yet know who embodies that fact. But the fact is undeniable from
the moment that we say is writing. Exactly the same occurs with the fact
of being: we are saying is being when we say is. And we are saying
is being even in a tautological way, in that which is (eon), which is that
which is being par excellence. No one can deny that that which is, is being: eon
estin.
The Spanish expression se estranslated either as plain is or it
isthat I have often used, may suggest that in our interpretation the estin
in 2.3a is considered to be an impersonal verb. Yes and no. Estin is not
impersonal, but it is used by Parmenides as if it were. To clarify this point,
we must look, very briey, at the question of verbs called impersonal.
As I have already said, I do not share the viewpoint of H. Frankel about
the decidedly impersonal character of estin in 2.3a,211 because, I believe, his
interpretation confuses two levels: on the one hand, there is the psychological aspect of impersonal verbs, which he brings up, and on the other, the
meaning content of his examples. For a Greek of the classical epoch, the
psychological subject of verbs called impersonal was probably the divinity.212 But it is obvious that the meaning content of is raining is not Zeus
209
210
211
212

Kahn, C. H., Being in Parmenides and Plato, La parola del passato 43 (1988) 247.
Cassin, B., Si Parmenide (Lille: Presses Universitaires de Lille, 1980), 55.
Cf. note 204.
This is the opinion of Schwyzer, E., Griechische Grammatik, Vol. II (Munich: C. H. Beck,
1950), 362; Kuhner, R., Ausfurliche Grammatik der griechischen Sprache, 2nd ed., Vol. II(1)

(1) Estin on Its Own and Its Negation

53

is raining;213 at most, we might think, Zeus is making it rain. In fact,

when we say in any kind of language that it is raining or it is snowing,
we are alluding to the effective presence of something (rain, snow) in the
corresponding verb tense (it rained if it happened in the past, and it will
rain if we are talking about the future). That is, semantically, it is raining
means at this moment the fact of raining is happening. We do not know
whether the origin of the notion raining preceded the formation of the
term rain, or vice versa (personally, I think, without being able to prove
it, that the formation of the verbs preceded the nouns). However that may
be, it cannot be denied that the third person singular of the present tense
of the verb to rain is it rains, and when we say it rains we are asserting the presence of rain now in the present. The verb in question is called
impersonal because there is no person to be the subject. It is raining
positively states a process taking place outside the I-you, the only pronouns referring to persons, E. Benveniste wrote.214 If we take this interpretation of impersonal verbs into account, I have no doubt in stating that in
2.3a and 2.5a the two instances of estin, as autonomous elements up until
now, are used in an impersonal way. Then, just as it rains means the
fact of raining is happening now, raining is present now, it is means
the fact of being is happening now, the fact of being is present now.
In summary, let us say that the two instances of estin do not have a
subject at rst because, for reasons of method or didactic reasons, Parmenides preferred to expound his thesis by means of an afrmation that implies, itself, the only possible conceptual subject (just as rain is the only
possible subject of the phrase it rains). For this reason, my position
with regard to Parmenides estin is a little different from the classical conceptions: I do not deny that there is a subject, but I do not believe that this
subject must be extrapolated from the passages in which it is found. The
subject must be analytically extracted from the meaning of estin as Parmenides fundamental thesis.
As a result of everything I have said, and provisionally (because I have
not yet adopted a position with respect to the value of the verb to be in
Parmenides), we may propose is as a translation of 2.3a and is not as
a translation of 2.5a, with the proviso that the subject of both expressions,

213

214

(Hannover: Hansche Buchhandlung, 1870), 33; Holscher (1969), 78, note 30; Brugmann,
K., Griechische Grammatik (Munich: C. H. Beck, 1913), 656, note 1.
Contra C. H. Kahn, who thinks this belief exists even in modern Greek, where the subject
god is added to is raining to say theos brekhei (Kahn, C. H., The Verb Be in Ancient
Greek [Dordrecht/Boston: D. Reidel Publishing Company, 1973], 174). Benveniste had already said that this expression was a rationalization backwards (Benveniste, E., Proble`mes
de linguistique generale, Vol. I [Paris: Gallimard, 1966], 230).
Benveniste, Proble`mes de linguistique.

54

(c) Lines 3 and 5 of Fragment 2

which will appear at the opportune moment, is eon or einai, that which
is, the fact of being, [that which is] being.

(2) The Modal Complement of estin on Its Own and Its Negation
The content that each way proposes to think is constituted, as I said, by
two coordinated phrases. Or, if you prefer, there is a double content. According to our symbolization, we have already looked at A and not-A.
Now we must decipher the meaning of the second hemistiches of each
formulation, that is, B (ouk esti me` enai, fr. 2.3b) and not-B (khreon esti
me` enai, fr. 2.5b). From the syntactic viewpoint, unlike what happened with
the rst hemistiches, both B and not-B are complete sentences: there
is an innitive (enai) negated (me), which acts as the subject of the two
impersonal expressions, ouk esti and khreon esti. If the scheme is valid, it
assumes that the value of the two esti is different from that in the rst
hemistiches of each formula. Is that possible? There can be no doubt about
it. Even in Homer, the verb enai (to be) has multiple215 values, and
among these, as well as a strong sense (exist), we nd copulative or
impersonal216 meanings. The same thing happens in Parmenides, who
draws his inspiration from Homer.
Nevertheless, there are scholars, who without being able to deny the
evidence that the esti in 2.5b is linked to khreon, and together they form a
cliche,217 deny that ouk esti in 2.3b has an impersonal character. This is the
case with O. Gigon, who translates 2.3b as non-being, is not;218 H. Frankel,
Nicht-Sein ist nicht;219 and Ruggiu, il non-essere non e`.220 From the
viewpoint of the passages content, all these translations are correct (in fr.
6.2 Parmenides says exactly the same thing: mede`n, douk estin, nothing
is not). I have adopted an impersonal version since I believe that Parmenides took care with all the details of his thesis presentational structure,
especially in this fragment 2, which introduces it. So it would be difcult

215
216
217

218
219
220

Ebeling nds nineteen different meanings of enai already in Homer (Ebeling, H., Lexicon
Homericum [Leipzig: B. G. Teubner, 1885]).
Il. 13.114: it is not possible [ouk estin] for us to abandon the ght; 21.191: it is not
possible [ouk estin] to ght against the son of Kronos.
In this cliche, each of the two terms has had to abandon its own specicity: esti does not
have existential value, and khreon has abandoned (if indeed, it ever had) its participial
aspect (since, according to some scholars, the term originates from the combination of
khreo and on: cf. Wackernagel, J., Vermischte Beitrage zur griechischen Sprachkunde [Basel:
1897], 62; and Redard, G., Recherches sur khre, khresthai [Pars: H. Champion, 1953], 73).
Gigon, O., Der Ursprung der Griechischen Philosophie von Hesiod bis Parmenides (Basel:
Schwabe & Co., 1945), 251.
Frankel (1951), 403.
Ruggiu (1975), 251.

(2) The Modal Complement of estin on Its Own and Its Negation

55

to admit that in this structure, one of the second hemistiches denitely has
an impersonal value (2.5b, khreon esti), whereas the other (2.3b, ouk estin)
does not. My hypothesis is the following: in the double presentation (te ka,
and) of each way, the second hemistiches have a modal value; the rst
hemistich alludes to possibility and the second to necessity. In the rst case
the possibility is denied (it is im-possible), as it concerns the negation of
the reality of is (indeed, it is impossible to deny that one is); in the second
case the necessity is asserted, as it also concerns the negation of is, but in
this case the negation of it would make this way a wrong one: to assert
necessarily that one is not being is an aberration.
Hence we adopt a modal version of the second hemistiches, but in
this case as well, there are two possibilities, since both ouk esti (2.3b) and
khreon esti (2.5b) are related to the negative innitive me` enai. Note that
it is the same idea occurring in both ways that are considered to be opposites. This conrms that those who assume a change of subject in the rst
hemistiches of the two ways (cf. supra) are mistaken. The two possibilities
are based on the fact that, from a syntactic viewpoint, impersonal expressions have a subject, and this may be (1) a verb in the innitive or (2) a
complete phrase. For example, (1) it is necessary to die, it is possible to
survive; or (2) it is necessary that human beings die, it is possible that we
may survive. The phrase called the subject complement (example [2]) has
a subject and a conjugated verb (in our examples, human beings/we
and die/may survive, respectively). Matters become complicated in
Greek because in the complement phrases, the verb (conjugated in Spanish)
is in the innitive. [Translators note: English may use either a conjugated
verb or an innitive.] It is worth pointing out that the innitive ouk esti
(not to be) is open to both possibilities: it could be the subject innitive
of the impersonal phrases, or it could be the conjugated verbfor which
Greek uses the innitivethat follows them. The difference in our case is
vital. In our examples, it is to die that is necessary, it is to survive that
is possible. There is no subject, but once the possibility or necessity of the
fact is admitted, then candidates may appear to fulll the action of dying
or surviving, and these candidates are plentiful. In contrast, in the second
case, where the verbs are conjugated, they are predicated of a subject, and
the necessity of dying is asserted about human beings, and the possibility of
surviving is asserted about us. As can be seen, the difference between one
case and the other is vast: in one case, the weight of modality falls upon an
action, denoted by a verb; in the other case, it falls upon a subject, whatever
that might be. As regards Parmenides, the choice between one or the other
of these brings back the thorny problem of the subject of estin.
Let us see what solutions have been proposed. Let us begin with the
second possibility. Its partisans assume that there is a tacit subject in the

56

(c) Lines 3 and 5 of Fragment 2

second hemistiches as well. They consider that the verb me` enai corresponds to this subject, and as these second hemistiches are linked to the
two rst ones, and a change of subject in the same line would be inadmissible, they assume that me` enai has the same subject that they assumed for
estin. Thus they arrive at translations corresponding to this scheme: [being] is, and it is not possible that [being] should not be (should not be
is me` enai transposed into conjugated form); and [being] is not, and it
is necessary that [being] should not be. This is how F. M Cornford translates: [that which is] is, and it is impossible for it not to be, it is not, and
must needs not be;221 D. Gallop, that [it] is, and that [it] cannot not be
and that [it] is not, and that [it] needs must not be;222 J. Beaufret and J. J.
Rinieri, comment il est et quil ne soit possible quil ne soit pas,223 among
others.
As I favor postulating a subject already in 2.3b and 2.5b, I adopt the
possibility that considers only the innitive me` enai to be the subject
of the modal expressions. If this is so, the second hemistiches have a pleasant surprise for us: retroactively, they provide a possible conceptual subject for the rst hemistiches. I say conceptual, since the notion of being,
around which the whole Poem revolves, is polysemic: it appears represented by esti on its own, the innitive enai (and synonyms), and the
participle [to`] eon. Indeed, if the two hemistiches are linked by te ka
(and), they must refer to the same notion, and that of which is is said
in 2.3a must logically reappear in 2.3b, although negated, since the verb is
also negated:224 it is not [possible]. So I propose this translation scheme
for the second hemistiches, which I will develop throughout this work:
and it is not possible (ouk esti) not to be (me` enai);225 and it is necessary
(khreon esti) not to be (me` enai).
221
222
223
224

225

Cornford (1939), 3. The sametranslation can be found in Guthrie (1965), 13; in Austin
(1986), 159; and in Kirk-Raven-Schoeld (1983), 245.
Gallop (1984), 55.
Beaufret- Rinieri (1955), 79.
Holscher, who does not heed this negation, criticizes those who hold that the subject of
2.3b is me` enai, since then, given that the phrases are linked, the same would have to
go for 2.3a. Parmenides presents two linked phrases, each with its own structure, one
in the afrmative and the other in the negative; it is logical that, if the phrases do not
contradict one another, the same subject should also appear, once in the afrmative and
once in the negative. (Holscher, U., Grammatisches zu Parmenides, Hermes 84 [1956]
393).
In the 1997 edition of my Deux Chemins de Parmenide, this hemistich was translated in the
following way: ne pas etre nest pas possible (p. 27), that is, in function of the potential
value of ouk esti. Although his article is extremely subtle, Constantineau is mistaken
when he says that my translation is heterodox. He must have misread me: I had translated the passage in the same way as him. Moreover, he accuses me (after paying homage
to my impressive erudition: thanks!) of confusing me` enai and to me` eon (op. cit. in
note 11, 227). This is not the case. Both formulas mean the same, but their syntactic

(2) The Modal Complement of estin on Its Own and Its Negation

57

Before leaving this passage, we may note that various scholars are
against considering me` enai to be the subject of 2.3b and 2.5b. For example, G. Calogero says that if this were so, there would be a nominalization
of the innitivemaking it into a nounand this would go against Greek
syntax, which requires an innitive and not a noun as the subject of khreon
esti.226 This criticism can be applied to scholars who introduce the notion of
the non-being here. But this is not the case with me: I always think of the
fact of being or not being.227 With respect to these passages, we may say
that Parmenides does not hesitate to turn an innitive into a noun (cf. to`
pelein, 6.8), but this does not mean a reication (thingifying) of the
notion: from Homer onward, the innitive, with or without the article,
means the development of an action.228

226
227
228

function is totally different, since me` enai can be the subject of a potential impersonal,
and to me` eon cannot.
Calogero (1936), 157.
I share Mansfelds opinion: In frs. 2 and 3 the innitives have their usual meaning, they
are not turned into nouns (Mansfeld [1964], 81).
Falus (1960) 279.

This page has been intentionally left blank.

Chapter IV: The Meaning of Parmenides

Thesis (and of Its Negation)
Now that we have looked at the structure of the presentation of the only
two ways for investigation presented by the Goddess, we must try to grasp
the meaning of the true thesis they transmit. As I have already said, Parmenides presents his thesis in positive form and then reinforces this position by negating it. In the rst way we nd the statement of the thesis; in
the second, its negation. The syntactic structure of the two ways (that is,
the body of terms that congure both formulations, since each way has the
form of a logos, i.e., a discursive phrase and not an isolated term) has displayed an abusive presence of the verb to be in both formulations, found
in lines 2.3 and 2.5. If we leave aside conjunctions, connecting terms, adverbs and the impersonal khreon, all the terms used are persons or modes
of the verb to be (and some scholars say that forms of this verb are also
assumed as the subject).

(a) The Grammar of To Be

There can be no doubt that Parmenides Goddesss philosophy course is
concerned with being. But saying this is not saying anything. In Greek, as
in Spanish [or English], to be is a verb and, like any verb it can be used
as a noun, and then we can speak of being (used as a noun). But this
verbal noun is essentially different in Greek than it is in other languages,
and so we cannot ignore the problem. This specicity is one of the results
of the exibility of the Greek language, which permits all kinds of juggling.
E. Benveniste wrote that the linguistic structure of Greek created the predisposition for the notion to be to have a philosophical vocation.229 Indeed, the use of the verb to be as a noun absolutely does not mean what
philosophers call being (the noun). To use an innitive as a noun in
Spanish it must be preceded by an article, in this case el [the]. Then
the innitive ser [to be] becomes el ser [the being] used as a noun,
in Greek to` enai. However, this formula never gured among the concerns of the Greek philosophers. No Greek philosopher who inquired into
229

Benveniste, Proble`mes de linguistique, 73.

60

(b) The Meaning of Being

what today we might call the being of things, or even certain types of
beings, including the supreme being, ever asked what is to` enai? literally what is being?230 As we know, especially since the Aristotelian systemization, the formula used by all Greek philosophers to ask the question
of being is t esti to` on (to eon in Parmenides), What is being? To` eon is
the present participle of the verb to be, used as a noun. The difculty of
grasping the scope of this neuter present participle (since there is also a
masculine and a feminine present participle) has always given rise to all
kinds of misunderstandings, since its use as a noun, represented by the
neuter article to, is deceptive, and so Parmenides avoids it whenever he
can. Indeed, just as verbal-noun innitives always have a dynamic character,231 something similar occurs with the participle to` on, which as a present
participle means that which is being, that which engages in the act of
being now. In all that I have said up till now, philosophy is absent: I have
only summarized, perhaps too supercially, what Benveniste calls un fait
de langue,232 a fact about Greek simply as a language.
It is upon this linguistic fact that Parmenides reects. In Greek the word
for things is onta. Even in current everyday language, things are beings,
something(s) that is (are), that which is being. Philosophy has not yet
come into it: thats the way the Greek language is. But why do we call
something that is a being? Because the fact of being manifests itself in
that which is; if there is that which is, then the fact of being is assumed.
Without the fact of being, there would not be things that are. This sort of
platitude will constitute the nucleus of Parmenides philosophy. And that
is the reason why his thinking starts from an analysis of the notion of the
fact of being, arrived at from the evidence that is is occurring. If there is
something undeniable for anyone who is, it is is. If Greek syntax allowed
the formula, we could say, with R. Regvald, that the basic question would
be t esti esti, What is is?233

(b) The Meaning of Being and Returning to

the Question of the Subject of estin in 2.3a
Let us return to the abusive presence of the verb to be in the presentation
of the two ways of investigation. At the beginning of this work, I said that
230

231
232
233

Only the overactive imagination of Heidegger could think that to` enai was an object of
reection in Greece. Cf. Heidegger, M., Einfuhrung in die Metaphysik, 4th ed. (Frankfurt am
Main: Max Niemayer, 1976), 73.
Cf. Falus (1960), 279: the innitive, whether with or without article, signies the development of an action.
Benveniste, Proble`mes de linguistique, 71, note 1.
Regvald, R., Parmenide: Le trajet de la non-concidence, La revue philosophique 176
(1986) 18.

The Meaning of Parmenides Thesis (and of Its Negation)

61

Parmenides takes his inspiration from Hesiodic and Homeric models,

which may aim to make his Poem sound like an archaic, classical text.
And I also said that the presentation of the two possibilities for investigation, called ways by Parmenides, revolve around forms of the verb to
be. It would not be exaggerating to think that Parmenides is also taking
his inspiration from classical models in his use of the verb to be. So we
may inquire what the value or meaning of to be is in the Homeric poems,
for example. It goes without saying that my analysis will be impertinently
brief, since it might well require several volumes to do justice to the question.234
Already in Homer, the verb to be has very different meanings, but it
is interesting to note that the Homeric poems still contain vestiges of the
original concrete meaning of the verb, synonymous with live, breathe,
possess the breath of life.235 However, in most cases, the verb has a more
abstract value, called strong by linguists. For E. Schwyzer, this is equivalent to exist (existieren), be present (vorhanden sein),236 and P. Chantraine
stresses that this means existing in the strong sense of the term.237
As he wants to give his philosophy an epic-didactic atmosphere, Parmenides retakes the original meaning of to be238 and gives it an absolute
and necessary character, to the point of making it the central concept of his
system. In the passages of the Poem where the verb appears on its own, or
at most accompanied by a subject, the meaning be present, exist, possess effective reality prevails. For convenience, in translations or paraphrases, I shall continue to speak of to be, but when this term is used, it
must be interpreted in this work according to the above-listed meanings.
From what I have said, we can now return to the question of the subject
of estin in 2.3a and 2.5a. The reply to this question must now be sought
in the meaning of estin as the basic thesis of Parmenides teaching. When
Parmenides simply says estin on its own, he is stating (or proposing) a
fact: that there is, that exists, that is, that is being, that there is [a]
presence. So, what is it that there is, what is it that exists, what is it that
is present? For the momentthat is to say, as the starting point of his systemParmenides does not tell us, and he does not tell us because he wants
234
235

236
237
238

We need only mention the excellentand much discussedwork of Kahn, The Verb
Be, 486.
Cf. Curtius, G., Grundzuge der griechischen Etymologie (Leipzig: B. G. Teubner, 185862),
350 ; and Muller, M., Nouvelles lecons sur la science du langage (Paris: 1868), 69. Cf. Il. 2.641:
the children of Aeneas were already dead (literally, no longer were), as well as the
epithet usually attributed to the gods: the ever-living (literally, those who always
are).
Schwyzer, Griechische Grammatik, 624.
Chantraine, P., Dictionnaire etymologique de la langue grecque (Paris: Klincksieck, 1968), 137.
On the verb to be in texts before Parmenides, cf. Cordero (1997), Appendix I, 21533,
La signication du verbe einai dans la litterature pre-parmenidienne.

62

(b) The Meaning of Being

to stress the fact of being present and not a possible subject to take on
this role. By leaving the verb without a subject, for the moment, with a hint
of provocation (since, as I said above, in Greek too the phrase is unusual),
Parmenides wants to draw attention to a fact: it is not his intention to present a traditional type of formula (at that point) in which such and such a
principle or element would deserve to be considered as the being of
things. In a traditional formula of the type S is P or S is, something is
predicated of a subject by means of the verb, or, if there is no predicate, the
verb gives information about the subject. In formulas of the type Socrates
is Athenian or Socrates is, the verb adds an attribute to the subject or
states the subjects existence. But in both cases, the subject, Socrates in
our examples, is the term to stress. The subject is what something is said
about: that he exists, or is Athenian.
If Parmenides had put forward a subject, his estin on its own would
have related to that subject. But Parmenides avoids this schema on purpose,
by presenting his verb at the beginning, without a subject. Thanks to this
syntactic anomaly, Parmenides proposes a genuine thesis (etymologically,
this is the term that corresponds best to what Parmenides wants to do:
establish, set forth, and maintain an assertion): the presence, the existence,
the effectiveness of the fact of being. Parmenides expresses this idea in
multiple ways, thanks to the exibility of the verb to be in Greek, through
innitives (enai, pelein), through the participle ([to] eon), and fundamentally
through the verb standing on its own in the third person singular, estin, as
in 2.3a and in 2.5a. Clearly Parmenides feels that this latter form of the verb
is the one that best expresses his thought, because not only does it appear
at the beginning of what he has to say, but it also reappears in the recapitulations (cf. 8.2) and in the key moments of the Poem (cf. 8.16).
This preference for the conjugated verb (in the present tense) is perhaps
because it avoids any kind of reication (thingifying) of the notion.239
There can be no doubt that the participle and the innitive, even when they
are used as nouns, always keep their verbal character, as Parmenides wants
to stress, but according to the original meaning of enai, it is clearly the
present240 tense that respects Parmenides thesis besthis postulation of
presence241 because, as B. Cassin also saw, that is when the term has its

239

240

241

This is the case with a term absent from Parmenides, ousa, which expresses the notion
of being as an already effective reality. It is interesting to point out that Kahn calls Parmenides use of the verb to be veridical; in it, the innitive and participle serve
merely as a convenient nominalization of the indicative esti (Kahn, The Verb Be, 191).
As we shall see, this certainty prevents Parmenides from referring the verb positively to
the past or the future. Homer, who was ostensibly not a philosopher, did not have this
problem: cf. Il. 1.70.
Cf. 8.5: it neither was nor will be, but is now . . .

The Meaning of Parmenides Thesis (and of Its Negation)

63

full force as a conjugated verb.242 Contrary to what is usually said, Parmenides estin, at least at the beginning of the Goddesss speech, is not nontemporal. Albeit a platitude, we must remember, with L. Taran, that is
is the present of the verb to be.243 We may add that the reference to the
present is reinforced by the adverb nun (now) of line 8.5, which assumes the punctual, that is, temporal meaning of the verb.244 This presence value of the verb enai, whose Homeric roots were strongly stressed
by E. Heitsch,245 is the key, for this scholar, to Parmenides thought: Sein
ist Gegenwartigsein.
Despite what I have said, there are passages in the Poem in which the
third person of the verb does not appear on its own, but is accompanied
by a subject. In the rst case, after the presentation of the two ways, in 6.1
we nd eo`n emmenai246 ([that which] is being, exists, or better still, by
being, it is.247 Another example can be found in 8.46, that which is not
being (ouk248 eon) does not exist (out . . . estin), where, in virtue of the double negative, the same thing is being said as in 6.1. Finally, to go on to a
rather complicated syntactic structure, we may say the same about 8.3637,
since what else exists (estin) or will exist (estai), except that which is being
(tou eontos)?
So I am not denying that Parmenides estin has a subject (cf. supra,
where I comment on line 2.3a, for my position on this), but we must respect
Parmenides wish not to have put it in where he did not think it was appropriate to do so. If estin appears without a subject in the decisive passages
of the Poem, it is because Parmenides wants to make clear that it is enough
to admit exists in order to deduce from that, automatically (even tautologically) that there is existence. Indeed, what else could be except that which is
being (eon)? And that which is being, is being because the fact of being is possible
and manifesting itself now (estin). In order to say this, it is enough to say estin
and give this verb its original meaning, now conrmed as a philosophical
thesis. Only an a posteriori analysis can distinguish a subject and a predicate in such a notion, and Parmenides does so as his text progresses. A
242
243
244
245
246
247
248

Cassin, Si Parmenide, 51.

Taran, L., Perpetual Duration and Temporal Eternity in Parmenides and Plato, The
Monist 62 (1979) 49.
OBrien, D., Temps et intemporalite chez Parmenide, Les etudes philosophiques 3 (1980)
258.
Cf. Heitsch, E., Sein und Gegenwart im fruhgriechischen Denken, Gymnasium 18 (1971)
427.
Emmenai is the epic form of enai and is in the innitive because it is the verb belonging
to a complement phrase; in direct speech it would be estin.
This formula could have other possible translations. I will justify my choice later on.
I follow the text of the Aldina edition (cf. the summary critical apparatus that accompanies my version of the Poem). The Simplicius manuscripts propose oute.

64

(c) The Absolutization of the Fact of Being

symptomatic case is the long text preserved thanks to Simplicius and

known today as fragment 8, which begins with the same formula as 2.3a:
hos estin (that exists) and then expounds the characteristics or proofs
of this exists, proofs that all revolve round a subject: eon (the rst of
these is: [that which is] being [eon] is unbegotten . . . (8.3). As I said above
about my interpretation of verbs called impersonal, these attest the real
and effective presence (in the corresponding verb tense) of the activity related
to the formation of the verb. As estin is related with the activity of enai
(just as raining is related to the activity of to rain), is on its own
means there is being (just as it is raining means there is rainnow),
that is to say, the fact of being is present.

(c) The Absolutization of the Fact of Being, the Negation

of the Thesis and the Ways of Investigation
However, the rst way of investigation is a way to think, as well as the
enigmatic estin, that ouk esti me` enai (2.3b). This second statement, linked
to the rst, confers an absolute and unique value upon the rst statement. I
mean that in a twofold logic like that of Parmenides, it is not possible (ouk
esti) not to be (me` enai) necessarily means that only being is possible.
This statement, which could be deduced from the elimination of the two
negatives in 2.3b, and which will appear literally in 6.1, esti ga`r enai
(since it is possible to be) is summed up in the estin standing on its own
in 2.3a. Insofar as only existence exists, insofar as there is only the fact of
being, this term implies in itself both the subject and the predicate of
Parmenides thesis.
The absolutization of the concept of being is achieved by the negation of
the contradictory concept of not-being. Thus we arrive at a crucial point in
Parmenides thought, which, I believe, has not been analyzed to date with
the attention it deserves. This involves an analysis of non-being, the fact of
not existing, together as the other side of the coin with the fact of being.249
When we set the statement of the two ways side by side, we will tackle this
point more precisely. For now, we may say that Parmenides not only re-

249

One of the few scholars to take an interest in the question, Zucchi has a very radical
opinion on it. For him, Parmenides ontology is preceded by a medenology (Zucchi, H.,
Estudios de losofa antigua y moderna, [Tucuman: Universidad Nacional de Tucuman, Instituto de Filosoa, 1956], 919, chapter El problema de la Nada en Parmenides). In a
slightly exaggerated way, Colombo (Colombo, A., Il primato del nulla e le origini della
metasica [Milan: Publicazioni della Universita Cattolica Sacro Cuore, 1972]) shows no
doubt in stating that Parmenides starting point is the thesis il nulla e` nulla (12) and
that being non e` altro che non-nulla (37).

The Meaning of Parmenides Thesis (and of Its Negation)

65

lates the impossibility of non-existence to the necessity of existence,250 but

he puts the radical statement about being (there is being) together in the
same way with the only thing that can be said about non-being: it is not
possible not to be (really a double negative). These two formulas constitute
Parmenides thesis.
This brief incursion into the domain of the rst way (which, as I have
said, contains a thesis) offers us decisive elements for understanding the
negation of the thesis expounded in the second way, formulated in 2.5.
Each of the hemistiches in this second formulation is set against251 those in
2.3 and, as I have already pointed out, in order to keep the opposite values
ascribed to each way, we have to presume that the tacit subject of both
ways is the same. So the rst hemistich of the second way (2.5a) (which we
have called not-A) denies what the other way (which we have called
A) statesthe absolute reality of existenceand therefore proclaims that
the fact of being does not exist, that there is no being, that it is not being.
And the second hemistich, which is presented in modal form, reinforces
this thesis by stating that it is necessary that there should be no being.
Thanks to this game of oppositions, this second way establishes a thesis
diametrically opposite to that of the rst way (that is, it is its negation),
since it has to be said of the fact of being that it does not exist, and the
necessity of the fact of its not existing must be proclaimed. These two statements constitute the negation of Parmenides thesis.
So now let us look in a bit more detail at the content of the thesis and
its negation. I may say at once that, unlike various interpreters, I do not
consider that the statements in 2.3 and 2.5 are simply empty formulas, abstract statements applicable to any content. This interpretation, which goes
back to G. Calogero, nds its most distinguished representatives among
Anglo-Saxon scholars, including A. P. D. Mourelatos,252 M. Furth,253 and J.
Barnes.254 This position is not very far from those scholars who see in 2.3
250

251
252
253

254

Basson says that 2.3 is really the conclusion of a reasoning that will appear later in 2.78,
a reasoning that establishes the impossibility of nonexistence (Basson, A. H., The Way
of Truth, Proceedings of the Aristotelian Society 61 [196061] 75). I believe, vice versa, that
it can be deduced from the impossibility of not being, stated in 2.3b, that it is not possible
to say or to express that which is not (2.78).
Kahn has observed that there is an opposition between contraries in the rst hemistiches and a contradictory opposition in the second hemistiches. (Kahn [1968/69], 707.)
Mourelatos (1970), Chapter I.
Whatever Parmenides is saying will apply equally well to ascertain whether there is
animal life on Mars, or a rational root to a certain equation, or an amount of tribute that
will satisfy the Persians, or whether Socrates can y, etc. (Furth, M., Elements of Eleatic
Ontology, Journal of the History of Philosophy 6 [1968] 117.)
According to Barnes, estin has to be translated into English as it is, but it does not
mean anything: the term only has an ordinary anaphoric role, indicating that if something is inquired into, then either it exists . . . or . . . (Barnes, J., Parmenides and the
Eleatic One, Archiv fur Geschichte der Philosophie 61 [1979] 19).

66

(c) The Absolutization of the Fact of Being

and 2.5 simply the operative premises on the basis of which Parmenides
philosophy will be expounded. It is probable that this interpretation contains an echo of the desire to insert Parmenides philosophy into formal
schemes, or, directly, into syllogisms, as rst proposed by W. J. Verdenius255
and then by A. H. Basson,256 J. Mansfeld,257 U. Holscher,258 and D. Gallop.259
It is obvious that the success or failure, the luminosity or obscurity of these
attempts, are imputable to their authors and not to Parmenides, who did
not have any idea that at some later date Aristotle would propose certain
rules to follow in a scientic demonstration and praise the virtues of the
syllogism. This mania for formalizing Parmenides thought led certain interpreters to try to complete260 his original ideas.261 The fact that we can
detect in Parmenides the implicit presence of the principles of identity
(that which is, is, 6.1) and of non-contradiction (it is or it is not, 8.16)
and of the excluded middle (it is necessary to be absolutely or not to be
at all, 8.11), as well as arguments from the absurd (cf. fr. 8), does not
authorize these excesses.
Both Parmenides thesis and its negation are presented as ways of
investigation. According to Mourelatos, when the verb to be is given an
existential value, this fails to respect the character of being a way, possessed by both possibilities, since existential propositions can be the starting point of a route, or they could be the goal of the route, or they could
be stations along the route but they are not themselves a route.262 This
criticism by Mourelatos can be refuted. But this refutation leads us into
greater depth in the presentation of the two ways. As I have already said,
Parmenides sets out the only ways of investigation there are to think and
the statement of both ways begins with two pairs of declarative conjunctions: hopos-hos (2.3) and hos-hos (2.5). Little has been said in general
about these conjunctions, except in the particular case of M. Untersteiner,
255
256
257
258
259
260

261

262

Cf. Verdenius (1942), passim.

Cf. Basson, Way of Truth, passim.
Cf. Mansfeld (1964), Chapter II.
Cf. Holscher (1969), 83.
Cf. Gallop (1984), 69.
For example, Manchester imagined a dialogue between the Goddess and the missing
interlocutor in order to explain the obscure stages of Parmenides argument (Manchester,
P. B., Parmenides and the Need of Eternity, The Monist 62 [1979] 87 ff.).
This is the case with Calogero, who suggested placing the current fragment 3 as a complement to line 2.8 and then reading as much as you think to say (Calogero [1932]. 16).
For his part, Mansfeld suggests this conclusion to fragment 2, as a link with fragment 3:
since only that which is, is for thinking and saying (Mansfeld [1964], 82). The most
coherent and well-founded position among these is that of Wiesner, who places the current fragment 3 (since it is the same to think and to be) after 2.8a and before fragment
6, which, according to him, follows on from it (Wiesner [1996], 139 ff.).
Mourelatos (1970), 275.

The Meaning of Parmenides Thesis (and of Its Negation)

67

who gives them an interrogative value, and thus arrives at a very special
position on the subject of estin.263 As we have already seen, most scholars
translate them as a complemental that, or state the implicit existence of a
verb that needs a that of this type. I do not think it is necessary to introduce a new verb to explain these declaratory links, since line 2.2 already
offers us noesai, whose value is complemented in 2.3 and 2.5. Indeed, the
content of the only ways of investigation there are to think is given in what
follows. On the one hand, there is a way to think that it exists and it is not
possible not to be (2.3) and, on the other hand, a way to think that it does
not exist and that it is necessary not to be (2.5).264
Our version of 2.3 and 2.5 enables us to explain the four declarative
conjunctions coherently and conrms that the nucleus (in 1.29 the Goddess
called it the heart) of the thesis, which will be true, is the statement of
the effectiveness of the fact of being. The way and its content fuse, since
the way is correlative with a way of thinking. As Heitsch states, the way
is the content of thinking.265
In this respect we can say that, although it would be dubious to seek a
theory of language levels already in Parmenides, there are certain expressions in the Poem that operate on a plane that, with J. Jantzen, we might
call metasprachlich.266 In particular, this is the case with judgments about
the thesis and its negation that state the content of each way. Given that
this content is expressed by expository phrases (statements or negations),
the judgment stands on a higher, or at least different, level. This allows
us to justify certain apparent contradictions in Parmenides exposition: for
example, the statement of positive references with respect to the fact of notbeing, even though this fact is denied. We know that thinking must move
away from not-being, but the expression it is not possible not to be (2.3b)
is true (2.4). Anything that is not being is inexpressible (2.78), but the
Goddess mentions (2.6) that the way that states the existence of something
that is not is absolutely unknowable.267 As we shall see, theseperhaps
263

264
265
266
267

Untersteiner (1958), lxxxv. Robinson proposes a compromise solution: for him the four
terms are deliberately ambiguous, since Parmenides wishes to indicate both existence and
a type of existence (Robinson, T. M., Parmenides on the Real in Its Totality, The Monist
62 [1979] 54).
The fact that further on, in 2.78, it is stated that the content of this way cannot be thought
or stated does not deny that, a priori, as a possibility, this way must be taken into account.
Heitsch, Sein und Gegenwart, 429, and Gegenwart und Evidenz bei Parmenides (Wiesbaden: Franz Steiner Verlag, 1970), 15.
Cf. Jantzen, Parmenides zum Verhaltnis, 118.
Heitsch, who does not take these nuances into account, maintains that the ways are mere
metaphors for the principal alternative (being or not being) and that, therefore, Parmenides can say the same about the way of being that he says directly about being (cf.
Heitsch, Sein und Gegenwart, 430, note 37). But this scholar does not explain how this
principal alternative can be reached before the values of estin and ouk estin have been

68

(c) The Absolutization of the Fact of Being

spontaneousnuances of language justify the possibility of being abreast

of everything (panta), even if this everything implies, in addition to a
positive content, a theory about something that is wrong or false. Knowing
what the falseness of the false entails, is true.
If this is possible, it is because the rst way of investigation is a thesis,
and the second way is its negation. I do not share Mourelatoss criticisms
about this assimilation (way = thesis). The thesis, which like any thesis, is
made more explicit further on, is completed, reinforced; it is a road to
travel. The thesis in itself is not the object that the Goddess proposes for
study. On this point I agree with Mourelatos, for whom in any investigation (dzesis) information is gathered about the object of study.268 Investigate does not have a special meaning in Parmenides. As we can see from
8.6 what origin will you seek for it (dizeseai)?dzemai means seek,
investigate, and the same happens with the noun dzesis (investigation), apparently invented by Parmenides.269
So each way of investigation is a way for investigation to follow,270
to develop271 the supposed or suggested content of the respective thesis.272
Consequently, both the thesis expounded in 2.3 and its negation, contained
in 2.5 are possibilities for directing investigation, one of which will be
eliminated because of its contradictory content, whereas the other will unfold in a long series of characteristics or properties or proofs (semata), which
will form part of fragment 8.
The way in which both the thesis and its negation propose to direct
thinking consists in presenting a specic message. The thesis, expounded
in 2.3, states or proposes the fact of being, of being present now. Additionally, as we have already seen, it confers a necessary and absolute character

268
269

270

271
272

xed, especially when in 8.16, after having set forth this alternative, Parmenides recalls
that the choice has already been decided (kekritai, perfect). This decision has been
made in fragment 2, where the theoretical impossibility of one way has been shown, and
in fragments 6 and 7, in which the practical impossibility of following that same way was
demonstrated.
Mourelatos (1970), 67.
Cf. Mourelatos (1970), 67. The verb indicates a thorough search. Heraclitus used it for
seekers of gold (fr. 22), and the famous fragment 101 (I sought for myself) refers to the
deep nature of the self (cf. Guthrie, W. K. C., A History of Greek Philosophy, Vol. I [Cambridge, England: Cambridge University Press, 1965], 418), which, like every nature,
loves to conceal itself (cf. fr. 123).
Note that dizesios (of investigation) is a subjective genitive. The same occurs with the
other genitives related with the notion of way: the ways of night and day (1.11),
the way of persuasion (2.4), and the way of all (6.9). This means that the ways in the
fragment are ways that investigation (subject) has at its disposal.
I prefer to avoid the term deduce or assume in order not to involve Parmenides in
logical procedures that will be systematized later.
When we analyze fragment 6 we shall see that this development, from the original thesis,
will be represented by the preposition apo.

The Meaning of Parmenides Thesis (and of Its Negation)

69

upon the fact of being. These two aspects reappear, in an explicit way, in
8.11, where the khreon esti of 2.5b is repeated, referring to the fact of being:
pampan pelenai273 khreon esti, it is necessary to exist wholly. Here the
adverb pampan characterizes the fact of being wholly, absolutely and the
necessity of this is indicated by khreon esti. In fact, the same thing had already been said in 2.5b, although negatively, and if we admit that 2.3 and
2.5 are opposites, we can assume that in the rst way estin on its own
already implied the necessity of the fact of being. In all cases, the necessity
refers to the predicate, but I believe that it can be extended to the possible
subject. This is how the matter was considered, for example, by G. Buroni
(for whom 2.3 means that essere e` necessariamente274) and R. Falus (who
sets the absolute necessity of being against not being).275

(d) The Opposition Between the Thesis and Its Negation

If we admit the unique character of the fact of being, on the basis not
only of the maltreated hen (one) of 8.6,276 but also of expressions such
as oulon mounogenes (total, only-begotten) (8.4) and no other thing is
or will be, apart from that which is being (8.3637), we can maintain that
Parmenides thesis states the effectiveness of necessary, absolute and unique
existence. This existence, this fact of being, this presence is rst stated in
positive form, that is, inasmuch as it fullls its role of being present, almost
tautologically, and then in a sort of reductio ad absurdum in 2.3b, through a
double negative that denies the possibility of its nonexistence. The purpose
of this double negative is to reinforce the previous statement,277 the absolute
assertion of the fact of being. On the one hand, Parmenides states that by
being, it is (eon estin); this is his fundamental idea, his thesis, the statement
273
274

275

276

277

Pelenai and enai are synonyms: cf. the interchange of these verbs in 6.8 and in 8.4445.
Buroni, B., Dellessere e del cognoscere: Studii su Parmenide, Platone e Rosmini, Memorie della Reale Accademia delle Scienze di Torino, Serie II (Scienze morali, storiche e lologiche) 29 (1878) 334, note 1.
Falus (1960), 273. Cf. also Guazzoni Foa`: In line 2.3 there is not only the explicit assertion
of the existence of being, but also that of the necessary existence of being (Guazzoni
Foa`, V., Attualita` dellontologia eleatica [Turin: Societa` Editrice Internazionale, 1961], 39).
The thesis of Untersteiner, according to which Parmenides being is not one, but all
(oulon), (Untersteiner [1958], Chapter I, passim) won few followers. Even so, it is clear
that Parmenides said nothing about the One (cf. Barnes, J., Parmenides and the Eleatic
One, Archiv fur Geschichte der Philosophie 61 [1979] passim, especially 2021).
The double negative is a long way round in relation to the simple statement, but this
long way round can often be very useful. Its purpose is to conrm and reinforce the
conviction: I consolidate the value of a statement by negating its negation. (Morot-Sir,
E., La pensee negative [Paris: Aubier, 1947], 369.) Cf. also Sigwart, C., Logik, 5th ed. (Tubingen: Mohr, 1924), 200 ff.

70

(d) The Opposition Between the Thesis and Its Negation

that of itself justies Parmenides eminent place in the history of philosophy. But if Parmenides thesis had been limited to this statement, it could
have been held that, hypothetically (since the rest of the Poem refutes it),
as well as that which is being, something else could also have existed.
So, to rule out this possibility, Parmenides also states his rst thesis in
negative form. Nothing except the fact of being can exist, since it is not
possible not to be (2.3b). With this second statement, which is simply a
consequence of the previous statement,278 the whole conceptual eld is covered, as the summary of the alternative in 8.16 shows: exists or does not
exist (estin e` ouk estin). For Parmenides, any exhaustive study of reality
must take into account as possible objects of investigation (that is, as a
priori possibilities) existence and that which is not existence and which,
therefore, deserves to be called not being.279 Everything corresponding to
the terms of the alternative that Parmenides takes as his starting point (the
reality of existence, of the fact of being) is excluded from the opposite term,
and there is no intermediate possibility. Parmenides considers that the conceptual eld is constituted, a priori, by only two terms or spheres, and because of the basic opposition within these, there can be no intermediate
states. This is the scope of the thesis presented in 2.3: on the one hand, it
determines the positive aspect of the doctrine and immediately denies anything that does not respect the postulation of the fact of being stated in
2.3a. If the starting point had been white, Parmenides would have denied
not only black, but also red or green, that is, everything that, by not
being white, belongs to the sphere of the non-white. The notions of being
and not being are not contrary, but contradictory.280
This incursion into the dichotomous structure of Parmenides thought
enables us to pose a thorny problem that we have not yet given the attention it deserves. This is the relationship that exists between the thesis in
fragment 2 and its negation. There are scholars who state that 2.3a and 2.5a
are contradictory expressions, whereas 2.3b and 2.5b are contrary expres278

279
280

According to Ralfs (Ralfs, G., Der Satz von Widerspruch bei Parmenides, in Lebensformen des Geistes, ed. Glockner, H. [Kant-Studien, Erganzungsheft 86] [1964] 12), the value
of the conjunction te ka in 2.3 is only understood when we get to 6.12, where the correlation ga`r . . . d (since . . . whereas) reinforces the complementarity existing between
the two hemistiches. Cf. also Taran (1965), 191, note 44: existence implies the impossibility of nonexistence.
Cf. Simplicius, Phys. 116.23: [For Parmenides] it is obvious that what remains besides
that which is (to` para` to` on), does not exist, and that which does not exist is nothing.
If this is so, the other than being, discovered by Plato in the Sophist as the substitute
for non-being, would have been assimilated by Parmenides into his notion of not-being
and the parricide would not have taken place. If the crime was committed it was because
for merely chronological reasons, Parmenides could not defend himself.

The Meaning of Parmenides Thesis (and of Its Negation)

71

sions. This is the viewpoint of Kahn,281 Mourelatos,282 and Lloyd.283 From

this, Lloyd deduces that the alternative between 2.3 and 2.5 is not exhaustive (but his analysis, as J. Klowski284 aptly points out, does not take into
account the non-modal expressions, which are obviously contradictory).
Even so, it occurs to us that Aristotelian logic would demonstrate that the
statements in 2.3b and 2.5b are also contradictory. The impossibility represented by ouk esti (it is not possible) is contradictory to the notion of
possibility, but in order for something to be necessary, as postulated by
khreon esti (it is necessary), it must rst be possible (something impossible cannot be necessary). If we admit that adunaton (impossible)
is a synonym of ouk estin (it is not possible) and anagkaon of necessary, Aristotle comes to our aid when he states that to` anagkaon (the
necessary) ka` to` aduaton (the impossible), antestremmenos (contradictory)
(De int. 22b8). As P. Aubenque points out, Parmenides merely applies
avant la lettre, Aristotles denition of necessity, which is the impossibility
of its contrary,285 and therefore, with respect to the rst way, he could
have said that it is and it is necessary that it should be.286 For E. Heitsch,
it is not possible not . . . (es ist nicht moglich nicht . . .) signies necessity (die Notwendigkeit)287 and J. Wiesner shares this view, since, for him
cannot not be (kann nicht nicht sein) is synonymous with must be
(muss sein).288 When Parmenides expresses himself through exclusive alternatives, whether these are modal or non-modal, he wants to show that
he is thinking in terms of contradictory notions. A symptomatic example
occurs in 8.11, e` pampan pelenai khreon esti, e oukh (it is necessary either
to exist wholly, or not): the adverb pampan confers a modal character
upon the verb exist, and its negation is not, that is, its contradiction.
In the negation of Parmenides thesis we nd once more the absolute
systemization that we already found in the thesis itself. This negation of
the absolute systemization, as in the previous case, broadens the scope of
the thesis to cover the whole conceptual eld and prevents the appearance
281
282
283
284
285
286
287
288

Kahn (1968/69), 707.

Mourelatos (1970), 71: The modal clauses in the statement of the two routes in B2 are
opposed not as contradictories but as contraries.
Lloyd, G. E. R., Polarity and Analogy (Cambridge, England: Cambridge University Press,
1966), 104.
Klowski, J., review of Lloyd, Archiv fur Geschichte der Philosophie 51 (1969) 196.
Aubenque, Etudes, 109. The author bases his point on the text of Met. V.5.1015a34 and
b23.
Aubenque, Etudes.
Heitsch, H., Parmenides: Die Anfange der Ontologie, Logik und Naturwissenschaft (Munich:
Tusculum, 1974), 143.
Wiesner (1996), 116.

72

(e) Structural Difference Between Statement and Negation

of any new element. The negation of the thesis tries to afrm that ouk estin
[enai or eon] (it does not exist, there is no being), but it immediately
goes on to maintain that at the same time khreon esti me enai (it is necessary not to be). What is said in the rst hemistich acquires the quality of
necessity in the second.

(e) Structural Difference Between Statement and Negation

Nevertheless, there is an essential difference between the structure of the thesis and the structure of its negation. In the thesis there is a statement, reinforced by a double negative. In the negation of the thesis there is a rejection
of a term, reinforced (with the addition of necessity) by the statement of a
negative term. This strongly didactic dissymmetry leads us to study the
meaning of Parmenides thesis, as well as the scope of its negation.
The two ways of investigation, which contain Parmenides thesis and
its negation, are concerned with both being and not-being.289 In the rst
(that is, the one the Goddess states rst, in 2.3, which is the original thesis),
it is stated that there is being and that it is impossible not to be. In the
second (2.5, which is the negation of the thesis) it is stated that there is no
being and that it is necessary not to be. Consequently, I hold that it makes
no sense to speak of a way of being as opposed to a way of not-being,
since both ways speak of being and not-being;290 the difference between the
two is that in one case the fact of being is stated and non-existence is denied, and in the other, the fact of being is denied and the fact of not-being
is stated. To persist in speaking of the ways of being and of not-being, we
would have to call the rst way the way of being that is and the second
the way of not being that is (or of being that is not). It is worth pointing
out that the difference between both ways does not lie in the terms of which
they are composed (which are practically the same), but in the way in
which they are related, that is, in the type of predication established between being and not being. In the case of the rst way, we have a statement
([eon or enai] estin) and a double negative (ouk esti me` enai), which, in my
analysis, is equivalent to a positive statement.291 But, as we saw above, we
289

290
291

I use the terms being and not-being for the sake of convenience. The patient reader
who has followed my text to this point knows that by these terms I am alluding to the
fact of being, of existence, the effective reality of a presence, and to their negations.
That is why I prefer to speak of the rst thesis and the negation of the thesis or, as
we shall see later, of the way of Truth and of opinions.
For a double negative to be equivalent to a positive statement, it is necessary that between the true and the false there should be no possible third solution, and thus the
principle of the excluded middle should be explicitly admitted. (Morot-Sir, La pensee
negative, 371.) This is the case with Parmenides.

The Meaning of Parmenides Thesis (and of Its Negation)

73

construed the expression estin eon (or enai) through an analysis of estin
standing on its own, which produced itself as its only possible subject.
In contrast, in the second hemistich (2.3b), the formula me` enai is the
direct subject of the impersonal it is necessary. It is worth pointing out
that the rst way, as a whole, asserts that of eon (or enai), we can only say
that it is, and of me` eon (or me` enai), we can only say that it is not. The
way that supports this thesis accompanies the truth (2.4). So the primary
structure of the way of truth consists in stating or predicating a notion regarding
itself: it establishes the being of being and the non-being of not-being.292
There can be no doubt: we have here a tautology, or if you like, the principle of identity, but without the postulation of this tautology, any type of
thinking is impossible.
In contrast, in the second way, exactly the opposite occurs, since it is
the negation of this thesis. In the second way, we nd the rejection of a
positive term (ouk estin [eon or enai]) and the assertion of a negative notion
(khreon esti me` enai). In this second way, of enai (or eon), it is said that it is
not, and of me` enai it is said that it is necessary. This way is unknowable
(2.6): it is the way of error. The primary structure of the way of error consists
in the assertion or predication of a concept with respect to its own negation: it
establishes the non-being of being and the being of not-being.293 This dissymmetry between the conceptual structure of the thesis and its negation
covers a much more profound difference, essential for the understanding
of Parmenides thought.
The thesis expounded by the rst way is formulated, we might say, on
a single level: both the statement and the double negation operate between
similar notions (cf. the schemes in footnote 292). In both cases there is a
notion that splits into two and that is predicated or attributed to itself (the
reason why I have spoken, formally, of a tautology). Being is attributed to
being and it is said: there is being; not being is attributed to non-being and
it is said: it is not possible not to be. The thesis operates on one basic,
fundamental level. I call it basic because this level will act as a basis for
further reasoning. We could say that this thesis is the thesis of Parmenides,
his only thesis.
Therefore, its negation, represented by the second way, is, in contrast,
secondary to it. Parmenides Poem is an eminent example of the secondary
and derivative character of any negation in relation to the positive statement. This
hierarchy can be explained in the following way.
292

293

If, didactically, we give the positive terms the symbol X (enai, estin, eon, khreon esti)
and the negative terms the symbol Y (ouk estin, me` enai, me` eon), the rst way follows
the scheme XX and YY.
According to the symbolization assigned in the previous note, the structure of the second
way would be YX and XY.

74

(e) Structural Difference Between Statement and Negation

In a system like that of Parmenides, there is a primordial concept: the

fact of being, existence, presented at rst by estin standing on its own and
later claried. Everything that does not correspond to this presence does
not exist.294 The only existence of that which does not exist (provisional
existence, we might say) consists in its presence in formulas such as nothing, non-being, and so forth, within the body of the argument: its only
reality resides in the negation of the only real; it is a mere empty name.
This scheme belongs preeminently to Parmenides thought. It is not the
case with other philosophical systems, in which nothing or negation have
had a positive role. In Hegels philosophy, for example, being is the thesis
and nothing[ness] the antithesis, but despite this opposition that makes
nothing[ness] something secondary with respect to being295 (it is anti-being,
since it is an antithesis): the intuition or thought of nothing[ness] has a
meaning; the two are different and, consequently, nothing[ness] exists in
our thinking and our intuition.296
The philosophy of existence also recognizes the positive nature of both
principles, although they are radically opposed; human existence is suspended in nothing[ness] and it is in this rootlessness that it nds the possibility of an authentic philosophy.297 Therefore, this primacy of nothing[ness]
confers autonomy upon negation, which does not have to be the negative
aspect of a positive statement, but is itself an independent expression: the
necessary condition for it to be possible to say no is that not being should
be a perpetual presence in us and outside us, and that nothingness besieges
being.298
Everything is different in Parmenides, whose absolutization of the fact
of being (which is the only thing that can be grasped whose reality cannot
be questioned) is expressed both by a statement and by a double negation.
294

295
296
297
298

Parmenides extremism is such that he does not dare formulate the negation of being in
an afrmative way, which would have been possible: non-being is non-being. However,
fragment 8 presents some examples of negations referring to terms, which, in their term,
imply a negation: [that which is being] is not divisible (8.22); [that which is being] is
not decient (8.33). Even so, Parmenides would never have subscribed to the phrase
with which Plato paraphrases his thinking: that which is not (to me` on) is really (ontos)
non-being (me` on) (Soph. 254d).
As Sartre wrote, nothingness is logically posterior [to being], since it presupposes being
in order to negate it (Sartre, J. P., Letre et el Neant [Paris: Gallimard, 1943], 51).
Hegel, G. F., Wissenschaft der Logik, 3rd ed. (Leipzig: Felix Meiner, 1929), 67.
Cf. Heidegger, M., Was ist Metaphysik? (Bonn: Cohen, 1929), 20.
Sartre, Letre, 47. Certain philosophers of the analytic tendency, generally situated at the
antipodes of the philosophy of existence, agree with it on this point and even admit the
existence of negative facts. Cf. Ryle: there really are negative facts (Ryle, G., in Knowledge, Experience and Realism, ed. The Aristotelian Society, Vol. Supp. 9 [1929] 80). Cf.,
contra, notes 303 and 304.

The Meaning of Parmenides Thesis (and of Its Negation)

75

In Parmenides the negation is the rejection of the statement. Various scholars

followed his example, although none of them went as far as Parmenides.
Below, I mention certain cases that have treated the negation as secondary
to the statement. For H. Bergson, for example, that which is perceived is
the presence of one thing or another, and never the absence of what might
have been,299 and consequently, as H. Hoffding had already noted,300 negating always consists in setting aside a possible positive statement.301 This
position found its most enthusiastic supporters in two eminent logicians:
W. T. Krug and C. Sigwart. The former stated the problem thus: there is
no negation conceivable without the concomitant conception of an afrmation, for we cannot deny a thing to exist, without having a notion of the
existence which is denied.302 This reliance of negation on the corresponding
positive statement was also indicated by C. Sigwart, for whom the object
of negation is always a formulated or attempted judgment.303 This position
is a consequence of the thesis according to which, in reality, nothing can be
based on a negative judgment,304 and that is why negation is relegated to a
subjective or psychological plane. Thus, according to J. Thiedewho appears to base himself on Trendelenburg305 being in which we nd ourselves, given reality, is composed of simple positive facts;306 negation derives its origin from thought and does not exist without thought.307 Finally,
as a curiosity, I may add that similar positions can be found in some thinkers from India. Prabhakara, for example, does not admit the metaphysical
reality of negation, and for this reason, the problem does not exist from the
epistemological viewpoint either.308 In this system, negation is nothing but
299
300

301
302
303

304
305
306
307

308

Bergson, H., Levolution creatrice, 3rd ed. [Paris: Alcan, 1907], 305.
The negative judgment always presupposes the corresponding positive judgment (Hoffding, H., La base psychologique des jugements logiques, La revue philosophique 26 [1901]
374).
Bergson, Levolution, 311.
Krug, W. T., System der theoretischen Philosophie, Vol. I: Denklehre oder Logik (Konigsberg: Goebbels und Unzer, 1806), 118.
Sigwart, Logik, 155. Mabbott shares this opinion: The real foundation of the negative
judgment is the corresponding afrmative judgment [ . . . ]; therefore negation is subjective (Mabbott, J. D., in Knowledge, 72).
In the complete grasp of experience of truth, no negative judgment would remain
(Mabbott, in Knowledge, 73).
Trendelenburg, F. A., Logische Untersuchungen, Vol. I, 3rd ed. (Leipzig: S. Hirzel, 1870), 44.
Cf. Hobbes: Sunt autem positiva negativis priora (Hobbes, T., Elementorum philosophiae,
Vol. II [London: A. Crook, 1655], 7).
ber die Negation, den Widerspruch und den Gegensatz, dissertation, (Berlin,
Thiede, J., U
1883), 6. Cf. also 4.0621 of the Tractatus of Wittgenstein: the sign - does not correspond
to anything in reality (Wittgenstein, L., Tractatus [Frankfurt/Main: Suhrkamp, 1989]).
Cf. Mookerjee, S., The Buddhist Philosophy of Universal Flux (Calcutta: University of Calcutta, 1935), 409.

76

(e) Structural Difference Between Statement and Negation

the absence of a knowable fact, and consequently, the negation of a negation must assume that which must be negated.309
Parmenides reasoning evolves in a similar atmosphere. The assertion
of estin, as J. Jantzen stresses, states a fact (Tatsache, a state of things).310
The statement is primordial and the negation relies on (that is sup-poses)
the corresponding statement. That is why I maintain that whereas the thesisthat is, the rst wayoperates on a single level, the negated thesis
the second wayassumes two levels: the statement and its negation. It is
precisely this fact that leads to Parmenides second way being condemned
because of its internal contradiction. To grasp the scope of this contradiction
we have to take into account the fact that Parmenides thesis, precisely
because it is a thesis, is a statement. But because of its object, it is a privileged, exceptional, unique, statement: the word be assumes the implicit
assertion that the object designated by it exists; if any word uttered expresses a reality, then, we could say, the word be expresses a reality
squared.311
I think this statement by Verdenius is apt, so I can state that simple
estin is not a neutral term. It is a stated term, whose strong value is
shown throughout the rest of the Poem. We may also adduce the viewpoint
of Kahn: estn calls for no argument, and in fact Parmenides offers none.
He merely asserts that his thesis is true.312 In Parmenides, estin is the exclusive content of the rst way, and as such, reappears with the recapitulation
of the only muthos that remains, once the wrong way has been eliminated.
This muthos is a word that is also a statement hos estin (that there is).
So, estin standing on its own, as a statement, can be true or false, but Parmenides cannot fail to state that it is true, since its negation is impossible,
and for this reason estin becomes the basis of his system. The subsequent
unpacking of the term into a subject and a predicate enables him to
give proofs (semata) of its necessary and absolute character in the extensive fragment 8. But it must not be forgotten that predicate and subject
are indissociable, since there is only that which is being, and only that
which is being is. As A. Baumann states, the predicate is that which is
thought with the subject.313 For Parmenides, both plain estin and plain enai
mean there is being, the fact of being exists, it is being.
309
310
311
312
313

Das, A. C., Negative Fact, Negation and Truth (Calcutta: University of Calcutta, 1942), 134.
Jantzen, Parmenides zum Verhaltnis, 115.
Verdenius, W. J., Der Logosbegriff bei Heraklit und Parmenides, II, Phronesis 12 (1967)
112.
Kahn (1968/69), 711.
Baumann, A., Formen der Argumentation bei den vorsokratischen Philosophen, dissertation
(Wurzburg, 1906), 41.

The Meaning of Parmenides Thesis (and of Its Negation)

77

If we take this analysis into account, we see immediately that the negation of Parmenides thesis (=ouk estin [eon]) is a combination of statement and
negation: it negates a statement (eon). If my above analysis is correct, saying
estin assumes the statement of the fact of being. Therefore, by statingas
the negation of the thesis doesouk estin [eon], we are stating that being
that exists does not exist, since the notion of existence is inseparable from
eon, and we are postulating non-existence of this eon that exists. To put it another way: in the negation of the thesis there is a mixture of positive and
negative notions (cf. the symbolization proposed in note 292), of being and
not-being, since when we say there is no being, we are saying that being
exists and does not exist.
The scope of this second thesis was wonderfully grasped by Parmenides ercest enemy, the only philosopher who decided polemically to set
out on the second way: Gorgias. According to him, if non-being exists, it
will exist and at the same time, it will not exist, since insofar as we think it
does not exist, it does not exist; but as it is non-being, it will exist (Sextus,
Adv. Math. VII.64).314 In this intrinsic contradiction lies the error of the negation of the thesis, which does not respect the principle postulated by the
thesis itself: the concept of being can only be stated or predicated with regard to
itself.
But everything gets worse when the negation of the thesis also says
that it is necessary not to be. Now we not only have the statement of a
negative notion, but of the necessity of the existence of that notion, the
postulation of the effective reality of non-being, as against its impossibility
pointed out in 2.3b. From this, as I have argued up till now, it must be
recognized that in Parmenides the existence of a negative term regarding
the fact of being already represents a contradiction, since the fact of being
is afrmative and cannot be negated. F. M. Cornford is right when he says
that the words the non-existent (absolute nonentity) cannot be uttered at
all without self-contradiction.315 In the best Parmenidean tradition, H. Bergson states that by the mere fact of saying the object A, I already attribute
a sort of existence to it.316 Consequently, thinking of the object A as inexistent is to think, from the start, of the object, and consequently to think of it
as existing,317 and then replacing it with another object, which is its nega314

315
316
317

According to Guthrie, in this passage Gorgias argues in ultraparmenidean terms (Guthrie, W. K. C., A History of Greek Philosophy, Vol. III [Cambridge, England: Cambridge
University Press, 1965], 196).
Cornford, F. M., Platos Theory of Knowledge (London-New York: K. Paul-Harcourt, 1935),
208.
Bergson, Levolution, 314.
Bergson, Levolution, 309.

78

(e) Structural Difference Between Statement and Negation

tion. This shows us that every negative object possesses a positive moment, as G. Kahl-Furthmann318 says: that in which the object in question is
considered and the particle not is associated with it to obtain the corresponding negation.
However, according to Parmenidean postulates, it is not possible negatively to state a term referring to the fact of being (because it is not possible
not to be), and 2.5b falls into this contradiction. A negation referring to the
fact of being can only be negated, as is seen in 2.3b. Even in fragment 8,
where Parmenides is concerned with the proofs or properties of the
fact of being, there is a long series of negative terms, but these terms originally assume a negation.319 The two negatives cancel each other out and
Parmenides does not violate his own principle, which consists of stating
positive terms and negating the negative ones when talking about being.
This already implicit contradiction in the negative terms referring to
being is reinforcedwith the modal nuance of necessityin the second
hemistich of the second way (2.5b). Here we nd again the mixture of being
and non-being in 2.5a, and we state it is necessary not to be. So, as a
whole, this second way combines, mixes, and interchanges notions of being
and not-being, predicating one of the other, attributing to each concept the
contrary concept (there is no [being], 2.5a; it is necessary not to be,
2.5b). In order to do this, the negation of the thesis has to assume the thesis
itself, which had postulated the two concepts (but at the same time had set
out the rules of the game: predicating each notion of itself).
For this reason I say that the second way is secondary in relation to the
rst: every negation of the fact of being presupposes the fact of being and
decides to negate [deny] it. In the same way that someone who denies the
white must know what white is (indeed, what meaning would the expression non-white have for someone who did not know what white is?),
and as an atheist is someone who denies that that which is considered or
called god can exist, any attempted negation of the fact of being must
start from an understanding of the notion of the fact of being. But the analogy and the temptation to nd already in Parmenides a sort of ontological
318
319

Kahl-Furthmann, G., Das Problem des Nicht, 2nd ed. (Meisenheim/Glan: A. Haim, 1968),
129.
For example, unbegotten (ageneton, 8.3): in begotten the idea of generation is implicit,
which, for absolute being, implies originating from non-being. The prex a (un) of
ageneton denies this generation, which would be against the everlastingness of the fact
of being. So Parmenides can say that that which is, is unbegotten, since this means that
it is un-un-everlasting; ergo, that which is, is everlasting. The same occurs with anolethron: in-destructible = un-un-solid; therefore, the fact of being is solid. Atremes and
akneton: im-mobile = un-un-xed; therefore, being is xed. Anarkhon and apauston: nontemporal = un-un-everlasting; ergo, that which is, is everlasting. According to Frankel,
even there it is a case of double negatives. (Frankel [1951], 402, note 12).

The Meaning of Parmenides Thesis (and of Its Negation)

79

argument ends here, given that any type of negation could be legitimate
if the notion negated does not exist (for many people, for example, god
does not exist, and it could be imagined that white did not exist); in the
case of the fact of being, that is impossible: denying that it is is denying itself.
That anything should be, by not beingaccording to G. Imbraguglias320
subtle formulamakes no sense.

(f) Why Is the Negation of the Thesis Impossible?

After expounding the thesis and its negation (and expressing various value
judgments that the rst accompanies the truth, whereas the second is
completely unknowable), Parmenides tries to justify why one way must
be adopted as a valid way and the other must be rejected. We should not
be surprised that Parmenides begins by expounding the reasons for condemning the negation of the thesis, since it is precisely these elements that
justify the necessity of the thesis itself. Indeed, it is the impossibility of notbeing (the necessity of which attempts to be the negation of the thesis) that
will make the fact of being necessarily evident. If you like, Parmenides
invents arguing from the absurd: the impossibility of the negation of the
thesis proves the validity of the original thesis.
Line 7 of fragment 2, which immediately follows the negative judgment
the Goddess expresses about the second way, begins with the formula gar
. . . (since . . .). This particle usually has a causal meaning; indeed, in line
8 it begins the exposition of the reasons that make the second thesis completely unknowable, unknown:321 since (gar) you will not know (gnoes) or
utter (phrasais) that which is not (to ge me` eon) (since it is not possible).322
The nucleus of the expression is the formula to ge me` eon,323 the direct
object of the verbs know and utter and antecedent of the parenthetic
phrase (. . . it is not possible.) In this formula me` eon sums up the content
of the negation of the thesis: it is the negation (me`) of that which is (eon),
whose content is synonymous with me` enai (not to be) and whose necessity is postulated in 2.5b. In other passages of the Poem, Parmenides always

320
321

322
323

Imbraguglia, C., Teoria e mito in Parmenide (Genoa: Studio Editoriale di Cultura, 1979), 99.
The Greek term is panapeuthea, from the verb punthanomai (be abreast of, inform oneself about, know). In 8.21, Parmenides uses apustos in the same sense: unknowable,
referring to the possible corruption of that which is. In Homer, these terms also have an
active value: ignorant (Od. 1.242; 3.88; 3.184). In the case of Parmenides, stating that the
second thesis is completely ignorant would open up a worrying perspective.
Wiesner (1996), 16566, strongly stresses the value of gar in this passage.
The particle ge cannot be translated, but as we shall see (cf. infra), perhaps it plays an
important part in this passage.

80

(f) Why Is the Negation of the Thesis Impossible?

uses ouk eon (that which is not) and meden (nothing)324 as synonyms. As the contradictory notion to eon, me` eon has the same characteristics as it, but negated: instead of being possible, it is impossible; instead of
being absolutely, it is absolutely not. In other words: these are two contradictory ways: in fragment 2 there is absolute statement or absolute negation.325 As O. Becker remarks, the negation of the thesis is the absolute
[bloss] negative complement of the rst way.326 Non-being, in Parmenides,
is absolute non-being; but as there is no intermediate term between being
and non-being, any type of relative non-being is also excluded (if this
were not so, Plato would not have written the Sophist in order to invent
it).327 As there is only being, any negation of the fact of being (relative,
absolute, provisional) is impossible. Me` eon is the term that contradicts eon
and, as with any contradictory opposition, there are no intermediate terms.
The conceptual eld is divided into two areas: being and the negation of
being, and as the latter is impossible, only the former remains.
Despite Parmenides insistence on presenting his thought in a priori
dichotomous schemes (which then become monadic, as one element is eliminated), there are scholars who believe they have discovered nuances
within each part of the alternative. This is the case with Verdenius, who
sees a difference between absolute Nothingness (meden) and that which
is not (me` eon).328 Likewise, Loenen is opposed to me` enai representing
absolute non-being, since, according to him, the expression refers to the
phenomenal, concrete, non-necessary world.329 I think that to refute these
attempted subtleties, we need only examine the interchange of the terms in
question in different passages of the Poem; this shows that Parmenides says
exactly the same about meden, me` eon and me` enai.330 As we shall see below,
324

325

326
327

328
329
330

The exhaustive reasoning in 8.710 clearly shows that all these terms are synonyms. Cf.
also Mondolfo: the antithesis between eon and meden, me` eon, even ouk eon . . . (Mondolfo, R., Discussioni su un testo parmenideo [fr. 8.56], Rivista critica di storia della
losoa 19 (1964) 313).
Schwabl, H., Sein und Doxa bei Parmenides, Wiener Studien 66 (1953), reprinted in
Um die Begriffswelt der Vorsokratiker, ed. Gadamer, H. G. (Darmstadt: Wissenschaftliche
Buchgesellschaft,1968), 412.
Becker, O. Das Bild des Weges und verwandte Vorstellungen im fruhgriechischen Denken, Hermes, Einzelschriften 4 (1937) 142.
Plato is a subtle reader of Parmenides. Indeed, when in the Sophist, he begins the criticism
of Parmenides position, Plato starts from the notion of absolute non-being (to` medamos
on, 237b), but a few lines later he suppresses the adverb and simply attacks non-being,
since he knows that, for Parmenides, any possible non-being is absolute.
Cf. Verdenius (1942), 42.
Loenen (1959), 24.
That me` eon and meden are synonyms can be seen from 8.710: that which is being cannot
arise from that which is not (me` eon), since it cannot begin from nothing (meden).
With respect to me` enai and meden, cf. the equivalence between 2.3b, it is not possible
not to be (me` enai) and 6.2a, nothingness (meden) does not exist.

The Meaning of Parmenides Thesis (and of Its Negation)

81

nothing can be said of any possible Parmenidean theory about the phenomenal world. Either this problem did not interest him, or he left to his readers/listeners the difcult task of propounding a coherent theory about it
(that is, one that respects the only possible way).
I said that in the expression to me` eon in 2.7 we nd the mixture of
being and not being belonging to the negation of the thesis, but I must add
that the neuter article to accentuates the contradiction. Parmenides rarely
uses the article to turn the notion of that which is being or that which
is not being into noun form, and probably there is a sort of assimilation
in Parmenides mind between the article and a possible ti, or something. For example, R. J. Ketchum suggests translating to me` eon as what
is not anything,331 as if the article particularized certain existence. Furthermore, the important position the article occupies in the phrase, reinforced
by the particle ge, which to an extent separates the article from the participle it turned into a noun, has also attracted the attention of some scholars. For example, W. Brocker says the article has a demonstrative character
and roundly maintains that here it is synonymous with touto, this,332
and Holscher says that the article, with the participle does not mean it is
turned into a noun, but it emphasizes the generality of the predicate: a
thing or something (that is not).333 The same author discovers this generalization in fragment 4.2, but in relation to the opposite concept: you cannot force that which is being to be separate from that which is. Finally, for
J. Klowski, the particle ge absolutizes me` eon, which thus become absolute non-being.334
It is true that in Parmenides we nd examples of the archaic use of the
article as a demonstrative,335 but this does not mean that the same thing has
happened here. Certainly, there are passages in which it is impossible to
decide whether we have an article or a demonstrative or relative pronoun.336 In 8.37, for example, to ge Moirepedesen (fate [Moira] forced it),
331
332
333
334
335
336

Ketchum, R. J., Parmenides on What There Is, Canadian Journal of Philosophy 20 (1990)
172.
Cf. Brocker, W., Die Geschichte der griechischen Philosophie vor Sokrates (Frankfurt: Klostermann, 1965), 57.
Holscher (1969), 80, note 37.
Klowski (1977), 113. For Ruggiu, too, me` eon is equivalent to absolute nothingness
(Ruggiu [1975], 215).
According to Kranz, the article has a deiktische (demonstrative) force, for example in
8.60: this (ton) (Kranz [1916], 1172).
The typical case is 8.58, toi dheteroi. In 6.1, the to preceding legein is a relative, according
to Diels (Diels, Vors., 3rd ed., 143) and according to Gomperz (Gomperz, H., Psychologische Beobachtungen an griechischen Philosophen, Imago 10 [1924] 7). Since I reestablished a second to in the same line, this opinion has become unsustainable (cf. Cordero,
N. L., Les deux chemins de Parmenide dans les fragments 6 et 7, Phronesis 24 [1979]
2425).

82

(f) Why Is the Negation of the Thesis Impossible?

where the particle ge recalls our passage, obviously a relative. Despite

everything, the case that concerns us, 2.7, is more problematical,337 since
to could be the neuter article of the participle. Nevertheless, whether it is
an article or pronoun, to individualizes the concept to which it refers, and
if we apply to Parmenides the analysis in Platos Sophist338 (which does,
indeed, refer to Parmenides), we see that the singular article implies the
unity of that to which it refers (just as the plural article refers to multiplicity), that is to say, that the article indicates something as one thing or
as several things. Be that as it may, if the formula me` eon is a contradiction in itself (combining, as it does, afrmation and negation), the addition
of to ge does no more than reinforce this.

337
338

Some scholars, like Calogero, translate this to as if it were a relative, without saying
anything about it: you could not know that which is not (Calogero [1932], 16).
Cf. the passage 238de.

Chapter V: Parmenides Thesis, Thinking,

and Speaking
Parmenides thesis states that that which is being is, that there is being.
Presenting this certainty as a way, a road to travel, suggests that a welldirected and methodical argument will unfold its richness, deduce the nal
consequences arising from accepting it, and possibly even justify it. Already
in fragment 2 the presentation of the thesis announced that it was a way of
thinking, and as we saw, the content of that thinking was expressed
through an utterance (or, if you prefer, in an attributive phrase) in which
the truth resulted from the type of predication that was established between the terms. It is worth pointing out that already in fragment 2 (and
even in the allegorical presentation of the thesis at the end of fragment 1)
Parmenides establishes a relationship between an object, thinking about
that object, and expressing it by means of speech.
So lines 3 and 5 of fragment 2 showed that this object had an absolute
and necessary character, since its negation was impossible. This then leads
to a consequence: thinking and speaking cannot dispense with this object.
Further, it is necessary to think and to say that by being, it is; that there is
being; that that which exists, exists. Part of fragment 8 and the rst two
lines of fragment 6 set out to demonstrate this necessity.339 Indeed, if nothing existed, there would be nothing to think about. Without that which is
being (eon), says Parmenides, you will not nd thinking (8.3536). Why?
the pupil might ask his master. Because it is the same to think and that
by [because of] which there is thinking (8.34). Thinking and being are
inseparable. Thinking recognizes one single cause (and it is by virtue of
this that there is thinking): the fact of being, which is its single object. It is
thanks to this that thinking is found. Let us now analyze that passage of
339

Wiesner has demonstrated with certainty that fragment 3, it is the same to be and to
think, plays an essential role in this argument (Wiesner [1996], passim), and that this
phrase is the premise of the theory of knowledge in B 2.78 and 6.1a (162). We take
the value of this fragment as relative, and Wiesner is right when he says that the role of
B3 in the argument is left in complete obscurity by Cordero (Wiesner [1996], 200). The
reason for my mistrust is simple: this brief text has come down to us isolated from any
context. Cf. also Tarans opinion in agreement, where he hesitates to accept a literal interpretation of the phrase (Taran [1965], 42 and 198). I have only taken it into account when
its terms reappear together in (and, in my judgment, claried) other passages, especially
in 8.34. Cf. infra the commentary on this passage.

84

(a) Thinking Is Expressed Thanks to Being

the Poem in detail, because in it we have the foundations of Parmenides

theory of knowledge.

(a) Thinking Is Expressed Thanks to Being

The Greek text of lines 34 and 36 in fragment 8 is not as uniform as the
critical apparatus of the classical edition of H. Diels claims. For this reason,
direct consultation of the manuscript tradition has led me to modify in one
respect the version unanimously accepted to date. The two single sources
of the passage are Simplicius (the whole of the passage in Phys. 86, 87, and
143) and Proclus (lines 35 and 36a in In Parm. 1152). Line 34 does not present major problems and can be translated in the following way: thinking
and that because of which [houneken]340 there is thinking [esti noema] are the
same. In contrast, in line 35 there is a minor variation in the manuscript
tradition,341 but there is also an important difference between the versions
of Proclus and Simplicius, radically changing the meaning of the passage
and, I believe, solving various problems. Let us look at this point. In line
35, interpreters have unanimously accepted the preposition in342 as accompanying the relative (hoi) and have translated, with slight variations,
since without that which is being, in which it is expressed. I incline toward eph, (= ep)343 and translate thanks to which. The rest of the
passage (lines 36 and 37) presents us with major problems. Let us now look
at this crucial passage in detail.
340
341
342
343

Cf. infra the justication for this translation.

Pephotismenon (illuminated, claried) instead of pephatismenon (expressed).
Like all the other interpreters I follow the second option.
This reading is only found in Simplicius.
This version is in all the Greek manuscripts deriving from Proclus, though in very different families (cf. Klibansky, R., and Labowsky, L., eds., Plato Latinus, Vol. III: Parmenides
nec non Procli Commentarium in Parmenides [London-Leide: In Aedibus Instituti Warburgiane-Brill, 1953], xxxvi), which suggests that it was already in the original. It also
appears in the Latin translation of de Moerbeke, G., which is itself based on a very
ancient text (cf. Plato Latinus, xxv). Dillon holds that this version has the advantage of
enabling one to discern fairly clearly the original Greek from which he was working
(Morrow, G. R., and Dillon, J. M., eds., Proclus Commentary on Platos Parmenides
[Princeton, NJ: Princeton University Press, 1987], xliv). The most important codex among
those containing this Latin translation (Ambrosianus A 167 sup.) keeps the quotations from
Parmenides in Greek, and the reading eph hoi clearly appears in folio 189 v (cf. Proclus, Commentaire sur le Parmenide de Platon, trans. de Moerbeke, G., critical edition of
Steel, C., Vol. II [Louvain: Presses Universitaires de Louvain, 1985], 426). On the value of
this manuscript, cf. Klibansky and Labowsky, Plato Latinus, xii, and Steel, Commentaire
sur le Parmenide, 3*, for whom the Ambrosianus is the most important witness to the
tradition, since it is the one closest to the archetype. This scholar also states, The terms
in Greek must have been found in the model (Commentaire sur le Parmenide, 4*).

Parmenides Thesis, Thinking, and Speaking

85

We saw that in these lines there is a relationship among the fact of

being, the expression of being, and thinking about it. In line 34, the term
houneken may have two different meanings: as a synonym of hou heneka
(that by [because of] which) it has an adverbial-causal meaning; but, as
a synonym of hoti, it can have the two normal uses of this term: complemental (that) and causal (because). In Homer there are various examples of the adverbial-causal use344 and the simple causal value.345 In contrast,
the complemental meaning only occurs with verbs that expect a complement clause (Od. 5.216, oda, know; Od. 7.300, noen, think), but never
with a noun. As this is the case in 8.34, it is very probable that here houneken
does not have a complemental value, even though some authors have held
that it does.346
The previous use of the same term in line 8.32 with a clearly causal
value (houneken ouk ateleuteton, because it is not innite) and even before,
in 8.13, of its synonym tou heneken with the same value,347 allows us to state
that the meaning of the term in 8.34, as H. Diels,348 K. von Fritz349 and J.
Wiesner350 have maintained, is it is the same to think and that by [because
of] which there is thinking. This meaning could also be deduced from
Simplicius commentary on the passage 8.3437 (if that exists by virtue of
which thought exists, it is clear that it must be something intelligible, since
it is by reason of the intelligible that thinking and the intellect exist [Phys.
144.2224]),351 and it would offer a plausible explanation of the causal nexus
gar, which completes the passage. That because of which there is thinking
is, clearly, a paraphrase of eon (or of enai).
The following lines offer the reasons for this statement: since without
that which is being (eon) [. . .] you will not nd thinking. The unknown
has revealed itself: that which is being is synonymous with that because
of which there is thinking. Without that which is, thinking has no foundation, since (gar, 8.36), as Parmenides observes as a conclusion to this argument, there is only that which is being (indeed, there is not and there will
344
345
346

347
348
349
350
351

For example, Od. 3.61: allow us to accomplish that by [because of] which (houneka) we
are here.
For example, Il. 9.505: Ate has light feet; that is why (houneka) she arrives before all of
them.
This is the case with Conche, who translates: . . . the thought that there is (Conche
[1996], 128); Cassin: . . . and the thought that is (Cassin [1998], 89); and Gomez-Lobo:
. . . and the thought of what is (Gomez-Lobo [1985], 113).
Cf. Taran (1965), 103: tou heneken means because of which.
Diels (1897), 85.
von Fritz, K., Nous, noen and Its Derivatives in Presocratic Philosophy (Excluding Anaxagoras), [I: From the Beginnings to Parmenides] Classical Philology 40 (1945) 237.
Wiesner (1996), 151.
On the quotations from Parmenides to be found in Simplicius, cf. Stevens, A., Posterite
de letre: Simplicius interpre`te de Parmenide (Brussels: Ousia, 1990).

86

(a) Thinking Is Expressed Thanks to Being

not be anything apart from that which is being [eon], 8.3637). Thinking
is condemned to be thinking about that which is: Denken heisst: Seiendes
denken.352 Nothing remains apart from that which is being.
We may add, in parentheses, that these lines 8.3437 offer us the only
context in which to set any possible interpretation of the succinct fragment
3,353 since it is the same to think [noen] and to be [enai].354 If we compare
this enigmatic text, word for word, with 8.34, we can state that since it is
the same to think (to` ga`r auto` noen estin) in fragment 3 is equivalent to it
is the same to think (tauton desti noen) (8.34), and and to be (te ka` enai)
is echoed in and that by [because of] which there is thinking (te ka` houneken esti noema). Being is that because of which there is thinking.355 There
is identity between thinking and being, but any idealist interpretation is
excluded, since it is being that has priority. The fact of being is the cause
of thinking356 and therefore all thinking is necessarily thinking of being. In
the formula thinking of being the genitive is obviously objective, since
the fact of being is the object of thinking. If this were not so, we would
share the extravagant idea of Phillips, according to whom the totality of
being thinks of its own totality.357 Being and thinking are very closely related, but in another sense: I only perceive that which is present, and that
which is present is the only thing I perceive.358
At the heart of this clear and precise reasoning, the relative phrase containing the pronoun hoi (which in the dative) plays an essential part,
but the phrases whole structure has given rise to erce polemics. W. Leszl
shows no doubt in stating that whereas the rest [of the passage] is clear,
the meaning of the interpolation en hoi pephatismenon esti is not.359 Let us
start from the unanimously accepted version (a version which, as I said, I
do not share: cf. infra): en hoi pephatismenon esti (8.35). The interpretation
closest to Parmenides thought holds that that which is, is expressed (pephatismenon, from phatzo) in thinking. If this is so, en hoi refers to thinking
352
353
354

355

356
357
358
359

Hoffmann E., Die Sprache und die archaische Logik (Tubingen: 1925), 8.
On fragment 3, cf. supra, note 339.
Vuia even proposed setting the line that constitutes fragment 3 between 8.33 and 8.34, as
the premise of 8.34 ff. (Vuia, O., Remontee aux sources de la pensee occidentale: Heraclite,
Parmenide, Anaxagore [Paris: Centre Roumain de Recherches, 1961], 82).
Wiesner rejects the apparent similarity between fragments 3 and 8.34 since, according to
him, the rst case refers to the gnoseological aspect of the theory, and the second to the
ontological aspect (Wiesner [1996], 162).
Eon is the conditio of noen (von Fritz, Nous, noen and ts Derivatives, 238). Thinking
implies that by [because of ] which there is thinking (Mansfeld [1964], 85).
Phillips, E. D., Parmenides on Thought and Being, Philosophical Review 64 (1955) 558.
Heitsch, E., Sein und Gegenwart im fruhgriechischen Denken, Gymnasium 18 (1971)
428.
Leszl, W., Approccio epistemologico allontologia parmenidea, La parola del passato 43
(1988) 309, note 40.

Parmenides Thesis, Thinking, and Speaking

87

and the subject of the participle is to` eon: without that which is being, you
will not nd thinking, in which that which is being (to` eon) is expressed.
Despite its coherence with Parmenides thought,360 this version has attracted
practically no adherents. Perhaps the contorted syntaxas Verdenius361
calls itof this version caused it to be set aside, seeing that the relative
precedes its antecedent.
So let us try to respect the order of the terms (that is, put the antecedent
rst and then the relative) and consider, somewhat imaginatively, that the
participle expressed alludes to thinking and that the relative picks up
that which is being. This is the classic position of Diels,362 which has had
many supporters. The general scheme of this structure of the terms is as
follows: Without being, in which it is expressed, you will not nd thinking. This is how Verdenius, among others, translated it: you will not nd
knowing apart from that which is, in which is utterly;363 Taran has it as
without Being, in what has been expressed, you will not nd thought.364
Bormann adopts this view as knowledge of being is communicated or
expressed in being,365 and in his turn, P. A. Meijer justies this translation
as, according to him, thinking is in being.366 Looking at these possibilities,
I have to say that the relationship between that which is being and thought
or thinking does not emerge clearly from the relative phrase in 8.35. Indeed,
how could thinking or a thought be expressed, expounded, or communicated in that which is? Being is the cause of thinking, and the effect cannot
be expounded or expressed in the cause. That which is cannot include
thought; if it did, the risk of idealism would be enormous, since that which
is thought would be, when in reality Parmenides, who is not Gorgias,367
says the opposite: that which is being, is thought.
As the traditional version led us down a blind alley (since the structure
most adapted to Parmenides thought was contorted, and that which appeared impeccable from the syntactic point of view did not agree with what
Parmenides was proposing), I have adopted the text unanimously transmit360

361
362
363
364
365
366
367

Indeed, this version maintains that, as thinking and the cause of thinking are the same,
without that which is (that is, without the cause of thinking) there can be no thinking, in
which, precisely, that which is, is expressed. Moreover, the priority of being over thinking
is reinforced, since if that which is were not, nothing could be expressed in thinking,
since thinking is the expression of that which is.
Cf. Verdenius (1942), 39.
Diels (1897), 37.
Verdenius (1942), 40.
Taran (1965), 86.
Bormann (1971), 84.
Meijer (1997), 83.
As we know, Gorgias decides to set out polemically along the way prohibited by Parmenides and says that if being and thought are the same, everything that is thought, is (cf.
fr. 3 7782).

88

(a) Thinking Is Expressed Thanks to Being

ted by all of Proclus manuscripts, eph [and not en] hoi. There are various
reasons supporting this choice. The relative phrase refers to a statement,
the expression of something, and when something is named in Greek the
formula ep (eph in our case) plus dative is used. When we want to say in
Greek that X applies the name B to A, we use this formula: X names
(that is, puts the name) B upon [ep] A.368 The complement in the dative
expresses the object receiving the name, or with a phrase in the passive
voice, the name by which the object is called.369 This is the causal use of the
preposition ep (the Liddell, Scott, and Jones Lexicon gathers various examples; see ep in section III: various causal senses, especially 5 on
names), since the object is considered as the cause of the name applied to
it, and this name is borne or carried like something carried or supported on
your back, like a label (ep also means carry upon, sup-port). Parmenides himself gives us a clear example in fragment 9: things which have their
own characteristics are named [onomastai] thanks [ep] to these or those.
According to Woodbury, this proposed interpretation arises naturally from
Simplicius paraphrase (Phys. 180.8): cold is [so] called thanks (ep) to density; the construction onomazein (or kalen) ep tin (put a name upon
something) is used of the relationship between names and reality.370
This same value of ep` + dative is present in the text I propose to adopt
in 8.35. The participle pephatismenon alludes to the fact of thinking (thought
is that which is expressed), and the relative picks up the notion of eon.
Intellectual activity is possible thanks to that which is, which serves as its
basis and which is exhibited through expression. Expression or utterance
(legein, phatzein, phrazein) makes thinking become concrete in thoughts
(noemata), but the support of thinking is that which is being, which is the
matter of all thought. Line 8.35 brings out this fact: thinking is expressed
(becomes concrete) ep onti, thanks to (or because of: ep never loses its
causal force371) that which is. This idea is expressed in the relative phrase
ephhoi, which is the reading I am proposing to adopt. Without that which
is being (aneu tou eontos, 8.35, to which the relative hoi refers), thinking does
not exist.372 This is the meaning of 8.3436: Thinking and that because of
368
369
370

371

372

Cf. Mourelatos (1970), 184.

Cf. Plato, Parm. 147d: you pronounce each of the names upon something (ep tini); Soph.
218c: the fact upon which (ephhoi) we are speaking.
Woodbury, L., Parmenides on Names, Harvard Studies in Classical Philology 63 (1958)
149. In the case of a person, this construction expresses the name given to someone, or
the thing that is named in (ep) the person: the name or the thing ts the person (160).
About my reading, Stevens says that this causal value of ep was already found in the
houneken of 8.32 and the tou heneken of 8.13 (Stevens, A., Posterite de letre: Simplicius
interpre`te de Paemenide [Brussels: Ousia, 1990], 4849).
Without alluding to my reading of the text, Wiesner arrives at a fairly similar version to
mine: Thinking is expressed within [that is how this scholar interprets the preposition

Parmenides Thesis, Thinking, and Speaking

89

which there is thinking are the same, since without that which is being, thanks
to which [ephhoi] it is expressed, you will not nd thinking. In other terms:
thinking does not exist except when it expresses something about that which
is being. And as there is not and there will not be anything else apart from
that which is being (8.3637), thinking only has one single correlative: that
which is. When being is absent from what is thought or said, there are only
empty words, deceptive names (8.52), which belong to opinions.
This necessary relationship between that which is being and thought
or thinking, to which utterance will be added in lines 6.12, appears again
in negative form in 8.89: It is not sayable [phaton] or thinkable [noeton]
that it is not [hos ouk estin]. Once again Parmenides leaves out the subject,
which should not surprise us: at this point in the Poem we already know
that the only thing that exists is that which is being, and that therefore it is
necessary to think it and express it. The negation of the thesis, which tried
to assert that that which is being does not exist, must necessarily be abandoned: It has been decided, through necessity, that one remains unthinkable and unnamable (since it is not the true way) and that the other exists
and is genuine (8.1718). The notion of necessity picks up on the impossibility denied in fragment 2. Necessity presupposes an a priori possibility.
This necessity to say and think that that which is being is, is formulated at
the beginning of fragment 6.
Lines 34 and 36 of fragment 8 have established the indissoluble relationship among that which is being, the fact of thinking it, and the possibility of expressing it in speech. We can say without exaggeration that philosophy nds its justication in this passage. Any philosophical system tries
to express a thought about reality in speech. Parmenides shows, for the rst
time (if there was anyone before him, no texts remain to prove it), that
thinking and speaking must grasp and express that which is; if they do not,
they are condemned to stray, wander off, and reproduce illusions, wishes,
and opinions. And for Parmenides, that which is, that which is being
(eon), is inseparable from its existence, since there is only that which is
being. Parmenides says nothing about possible later developments in philosophical thought; he lays the foundations for all possible reection: nothing
can be investigated if it does not start from the basis that that which is
being is, that the fact of being characterizes (how? each philosopher will
give his or her own reply) that which is being. And this is so for a very
simple reason: because there is being and nothing[ness] does not exist. The
rst lines of fragment 6 expound this obvious fact.
berlegungen zu Paren] thought, since there is nothing outside being (Wiesner, J., U
menides, fr. VIII, 34, in Etudes sur Parmenide, Vol. II, ed. Aubenque, P. [Paris: Vrin, 1987],
187).

90

That Which Is Being, Is

(b) It Has to be Said and Thought That

That Which Is Being, Is
Parmenides text known today as fragment 6 completes the formulations in
fragment 2 and relates them to two special ways of grasping that which is
being: thinking and speaking. The trilogy be-think-say we found in lines 34
to 36 of fragment 8 had already been given their foundation in this fragment 6.373 The content of this fragment 6 provides a series of essential elements for the understanding of Parmenides thought and, as was the case
with fragment 2, it is odd that over more than a millennium, no classical
author has seen t to quote any passages from it (unless possible quotations
have not come down to us). Indeed, this fragment 6which, as we shall
see, has been the cause of all kinds of misunderstandingshas come down
to us exclusively from Simplicius. If this commentator on Aristotle had not
had the brilliant idea of quoting these bare nine lines of the Poem, today
we would not know why Parmenides justies the necessity of saying and
thinking that there is being and why, at the same time, he fears that any
future philosopher might be tempted to follow a wrong way (or method)
that is unaware of this necessity. Simplicius cites the complete passage (the
whole nine lines) just once. The quotation is to be found in his commentary
on Aristotles Physics, page 117, lines 413. Part of the passage (the beginning of the fragment, lines 6.1a2a) also appears in the same work, on page
86.2728; and a fragment from the end of the text, lines 6.89a, is quoted
on page 78.34.
The text begins by expounding in a direct, clear, and distinct way the
need to say and to think that that which is being, is. The rst term in the
passage is an impersonal khre, it is necessary. There is nothing against
the innitives to say and to think being the subjects of it is necessary.
To reach this conclusion, rst we must reestablish Simplicius original text
(which, we may suppose, faithfully transcribed Parmenides original), since
from 1835 to date374 a modied version of this text has been used, in which
only one of the innitives, to say is preceded by the article to (the),
which led a considerable number of scholars to think that that to was not
373

374

Although we shall never know the order in which the nineteen quotations today called
fragments were set in Parmenides original, given the methodological nature of the
presentation of his philosophy, there are texts that have to come before or after others
(cf. the chapter Introduction to Parmenides). There can therefore be no doubt that the
text called fragment 6 today precedes fragment 7, and that the latter continues directly
in fragment 8.
I presented this original text in 1979 (cf. Cordero, Les deux chemins), but few scholars
took any notice.

Parmenides Thesis, Thinking, and Speaking

91

an article, but a demonstrative, and caused them to construe the phrase as

if it were saying: this [to] is necessary to say and to think . . .375 In Parmenides original text, as cited by Simplicius, each innitive is preceded by an
article, and then it makes no sense to speak of two demonstratives; each to
is simply the article that precedes the innitives that are used as nouns:
khre` to` legein to` noen, that is, it is necessary the saying [Greek: to say] and
the thinking [Greek: to think]. The story of the disappearance of the second
article from 1835 onwards is an odd one and deserves to be told, since it
shows us that understanding of classical texts usually depends on details
foreign to their authors. Indeed, it was S. Karsten and C. A. Brandis who
proposed replacing the to before noen with a te, despite the fact that
this to had been transmitted unanimously in all Simplicius manuscripts.376
Karsten considered that the reduplication of the articles (since there was
already a to before the innitive legein) was vulgar,377 and changed
the second to to te.378 For his part, C. A. Brandis suggestedin a work
also published in 1835379 not only changing the to before noen to te,
but also changing the to in to` legein to se (according to Professor
Grauert). From then onward, this Karsten-Brandis conjecture has been accepted, and it reappears in fundamental works, such as those by Riaux,380
Mullach,381 and Stein,382 until H. Diels nished off the job of disinformation
by inverting the terms in the critical apparatus of his 1882 edition: te noen
libri [that is, the original manuscripts], to` noen Karsten. Diels should have
written the opposite: te noen Karsten, to` noen libri.383 As P. A. Meijer lu-

375

376

377
378

379
380
381
382
383

A single example will be enough: Questo bisogna dire et pensare . . . (Giannantoni, G.,
Le due vie di Parmenide, La parola del passato 43 [1988] 211, note 7). The same construction was already to be found in Diels-Kranz, Fragmente I, 153; and in Cornford, F. M.,
Parmenides Two Ways, Classical Quarterly 27 (1933) 99.
I owe this information to Leonardo Taran, who some years ago now told me about the
true text of line 6a. Later consultation of all the Simplicius manuscripts containing this
text enabled me to conrm the correctness of Tarans information.
Karsten, S., Philosophorum graecorum veterum praesertim qui ante Platonen oruerunt operum
reliquiae, Vol. I, Pars Altera: Parmenidis (Amsterdam: J. Muller & Soc., 1835), 77.
Kahn shares Karstens viewpoint: he believes the second to to be the result of a copyists
error, repeating the rst to where Parmenides had written te (Kahn, C. H., Being in
Parmenides and Plato, La parola del passato 43 [1988] 261). The application of this kind of
intuition to other passages in the Poem would produce very odd results.
Brandis, C. A., Handbuch der griechisch-romischen Philosophie (Berlin: G. Reiner, 1835).
Riaux (1840), 210.
Mullach, F. G. A., Aristotelis, De Melisso, Xenophane et Gorgia disputationes, cum Eleaticorum
Philosophorum fragmentis (Berlin: 1845), 114.
Stein, H., Die Fragmente des Parmenides per physeos, in Symbola Philologorum Bonnensium in Honorem F. Ritschelii (Leipzig: Teubner, 1864/67), 783.
A detailed explanation of this philological tragedy can be found in Cordero, Lhistoire
du texte, in Aubenque, Etudes, 1920.

92

Impossibility of Thinking and Saying

cidly pointed out, this error caused a little philological tragedy. Since I
restored the second to, as Meijer says, the problem is resolved,384 as any
hope of considering the term as a demonstrative now makes no sense.385
To` legein and to` noen ([the] to say and [the] to think) are therefore
the subjects of the impersonal it is necessary (khre),386 but both innitives
have a content. It is necessary to say and to think something. What? That
that which is being is; that by being, it is: eo`n emmenai. This necessary content of thinking and speaking is Parmenides thesis. Emmenai is the Aeolian
form of the epic innitive of enai,387 that is, the phrase eo`n emmenai, in direct
speech, would take this form: eo`n estin. As we have already said, plain estin
standing on its own appears, in this fragment 6, accompanied by its subject
eon, that which is, that which is being, by being. This is nothing new.
We have already seen that the only indubitable possible subject of is is
that which is being, since this speaks for itself, once we have grasped the
scope of is as the basic and fundamental thesis of any attempt to do
philosophy. But Parmenides does not conne himself to presenting his thesis, he also states that thinking and speaking must necessarily admit it. Why?
Because the negation of the thesis is inconceivable, unimaginable, inexpressible. It is the impossibility of admitting the negation of the thesis that
gives it its absolute and necessary character. As that which is not being is
impossible, unthinkable, and inexpressible, it is necessary to say and to
think that only that which is in being, is. Parmenides expounds the impossibility of following the negative way in lines 7 and 8 of fragment 2.

(c) Impossibility of Thinking and Saying

That Which Is Not Being
In 2.78 we see the rst appearance in Parmenides of the relationship we
analyzed at the beginning of fragment 6, which will reappear in various
384

385

386

387

Meijer (1997), 11821. Inexplicably, a year before this optimistic statement by Meijer,
Wiesner, who knew about my restoration of the original text (cf. Wiesner [1996], 10),
continued to support the demonstrative nature of to (Wiesner [1996], 818).
Cassin said that Bollack and Wismann were tempted by this possibility. Both suggested
giving the article its full demonstrative meaning (Cassin, Si Parmenide, 54). This suggestion was dropped, and today Cassin kindly admits that convinced by Nestor Luis Cordero, I abjure the te that I retained in Si Parmenide (Cassin, B., Leffet Sophistique [Paris:
Gallimard, 1995], 557, note 9).
It is true that it is more common to nd innitives without articles as subjects of impersonal verbs, but in Parmenides we also nd the use of innitives with article as subjects:
cf. fr. 6.8, to` pelein, which is undoubtedly a subject, related to ouk enai (being and not
being).
Cf. Il. 2.249.

Parmenides Thesis, Thinking, and Speaking

93

passages of the Poem: that between the fact of being (or, negatively, between the notion of not being) and different modes of verbal or conceptual
reference to the fact of being. Lines 7 and 8 of fragment 2 state that the way
that expounds the negative thesis is completely unknowable since you will
not know that which is not (as it is not possible) or utter it (oute ga`r an
gnoes to ge me` eon [ou ga`r anuston] oute phrasais). This passage is concerned
with that which is not, knowing (gignoskein), and expressing (phrazein). If we take into account other synonyms of these notions, we have seen
that this trilogy has already appeared in 6.1 (eon, noen, legein; that which
is being, thinking, saying) and that we shall meet it again in 8.79 (eon,
estin, noen, phanai; that which is being, is, think, say), in 8.17 (hodos
[hos estin], noen [anoeton], onomazein [anonumon]; way [that is], think [unthinkable], name [unnamable]), and in 8.3436 (aneu tou eontos, noen, phatzein; without that which is, think, say). In all of these expressions there
are no signicant differences in the terms referring to the act of thinking
or those referring to the act of expressing. For think we only nd
noen and gignoskein, which are equivalents in Parmenides. Verdenius,388
Taran,389 and Mansfeld390 agree on this point. M. Untersteiner is of the same
opinion, but the reasons he gives to explain the presence in Parmenides of
the verb gignoskein (know) do not seem to me to be relevant.391
As for the expression of the fact of being, the ve verbs I have listed
(phrazein, legein, phanai, onomazein, phatzein) certainly are not synonymous,
but they denote similar nuances of the possibility of referring orally, by
means of speech, to that which is being. Phrazein stresses the possibility
of indicating, showing392 especially in wordswhereas onomazein
refers to giving a name, which is the reason Parmenides reserves this
verb for the way of error, in which reality is supplanted by empty words.
Legein has no special meaning in Parmenides; as for any Greek, the word
means say something signicant, whereas phanai and phatzo allude
to saying in the sense of uttering words.
So Parmenides forbids indicating and knowing that which is not,
that which is not being. There are various ways of explaining this prohibition, but the rst way had already stated that it is not possible not to be.
Therefore, it is also impossible to indicate, utter, know, think, etc. that which

388
389
390
391

392

Cf. Verdenius (1942), 35.

Cf. Taran, L., El signicado de noen en Parmenides, Anales de lologa clasica 7 (1959) 135.
Cf. Mansfeld (1964), 57, note 1.
According to this scholar, Parmenides should have written nooes (you will think),
but to respect the meter, he used gnoes (you will know) (Untersteiner [1958], cviii,
note 28).
Cf. Mourelatos, A. P. D., Phrazo and Its Derivatives in Parmenides, Classical Philology
60 (1965) 261.

94

Impossibility of Thinking and Saying

is not. This relationship between the double negation of the rst way (2.3b)
and the impossibility of referring (verbally or mentally) to that which is not
reappears in the parenthetic expression since it is not possible.393 A. P. D.
Mourelatos considers that this expression in fragment 2.7 (by means of gar)
offers the explanation of the impossibility of knowing that which is not, not
the impossibility of that which is not.394 Other authors who share the same
position translate the expression as non e fattibile395 or as it cannot be
consummated.396 There is no doubt that anuston can have this meaning
of realizable,397 but in the philosophical terminology of the fth and
fourth centuries it is the meaning possible that predominates. In fragment
2 of Melissus we nd the expression ou ga`r ae` enai anuston, which Albertelli translates as it is not indeed possible that it should always be398 and
W. Kranz has it as denn unmoglich kann immerdar sein.399 The same
meaning reappears in fragment 7 (3): alloude` metakosmethenai anuston
(neither is it possible that it should change structure;400 neither is it possible that it should change organization).401 In Diogenes of Apollonia we nd
the expression hos anusto`n kallista (fr. 3), with the meaning of in the best
possible way,402 and in Democritus (fr. 279) the formula malista ton anuston means as far as possible403 or with the greatest possible generosity.404
Finally, in Anaxagoras, an expression appears that is identical to that of
Parmenides: ou ga`r anusto`n panton pleo enai (fr. 5), whose meaning is it
is not possible that there should be something more than the whole.405 The
same occurs in Parmenides. In 2.7 ou[k] anuston returns to the impossibility of not-being formulated in 2.3b. The subject of the expression is to
ge me` eon; it is that which is not that is not possible, and therefore it
cannot be uttered or known.
Anything attributed to that which is not being remains bereft of any
reference. Parmenides reasons thus: the way that tries to state non-being is
393
394
395
396
397
398
399
400
401
402
403
404
405

We follow Simplicius reading instead of ephikton (accessible), transmitted by Proclus.

On this preference, cf. Diels (1897), 67.
Mourelatos (1970), 23, note 36.
Untersteiner (1958), 131.
Mourelatos, A. P. D., Some Alternatives in Interpreting Parmenides, The Monist 62
(1979) 9.
Cf., for example, Xenophon, Anab. I.8.11.
Albertelli (1939), 230.
In DielsKranz, Fragmente, I, 271.
Albertelli (1939), 235.
Bernabe, A., De Tales a Democrito (Madrid: Alianza, 1988), 186.
In the best possible way, KirkRavenSchoeld (1983), 440.
Solovine, M., Democrite (Paris: Alcan, 1928), 153.
Bernabe, De Tales, 325, note 63.
Bernabe translates: it is impossible that there should be something greater than all
things (Bernabe, De Tales, 262).

Parmenides Thesis, Thinking, and Speaking

95

completely unknowable (panapeuthea, 2.6). Why? Because it is impossible to

refer mentally or verbally to that which is not. Why? Because there is no
non-being. The causal particle gar in 2.7 explains the cause of the impossibility of following the second way of investigation, that is, of adopting the
negation of the thesis.406 But it had already been postulated in the rst thesis
that there is no non-being. Therefore, it can be stated without taking any
risk that Parmenides reasoning is circular, even though his circle has nothing vicious about it. He himself says so: it is common for me that where I
begin, there I shall return again (fr. 5).
Line 6.1a states the necessity of saying and thinking that that which is
being, exists: that by being, it is. This necessity is a novelty since what we
know up until now is that non-being is not possible; necessity had been
postulated about the erroneous content of the second way: not-being is
necessary. Now this necessity belongs to the rst thesis, which is thereby
enriched. We have moved on from possibility (it is [not] possible [not] to
be, 2.3b) to necessity. Now we know that the thesis estin is necessary,
since, as I said, estin standing on its own develops into estin eon, literally,
by being, [it] is. But this rst thesis has been enriched in another way
too. Whereas in fragment 2 its presentation unfolded on what we could call
an ontological level, the beginning of fragment 6 extracts a rst consequence on what we could call the gnoseological or referential level: as
only X exists, it is necessary to think and to express that X.407 Speaking and
thinking have a privileged position in Parmenides system, but they depend
on an ontological content, determined by the way of truth. Only that which
exists is thinkable and expressible by a logos. And as the presence of that
which is being is absolute and necessary, all thinking and speaking must
refer necessarily and absolutely to that which is.
It is worth pointing out that the beginning of fragment 6 returns, in
afrmative form, to the relationship between that which is being, and saying and thinking, but the negative proof presented in 2.78 (you will not
know or utter that which is not being) adds a basic positive foundation: it
is necessary to say and think that by being, it is, since it is possible to be,
whereas nothing[ness] does not exist. This new formulation of the thesis,
in its afrmative and its doubly negative aspect (that is, newly afrmative)
appears in the second hemistich of 6.1 and in the rst hemistich of 6.2. Let
us see how Parmenides presents his thesis again in fragment 6.
406
407

Wiesner (1996), 165, stresses the force of the term gar in this passage.
This way of presenting things excludes any kind of idealist interpretation that might be
encouraged by an isolated reading of the problematic fragment 3, being and thinking
are the same. As Levi observes, Parmenides does not say that only the thinkable exists,
but that only that which exists is thinkable (Levi, A., Sulla dottrina di Parmenide e
sulla teoria della doxa, Athenaeum 5 [1927] 270).

This page has been intentionally left blank.

Chapter VI: Presentation of the Thesis

and Its Negation in Fragments
6 and 7

In the previous chapters we saw that Parmenides thesis and its negation
are represented by two ways of investigation, one of which, corresponding
to the thesis, is valid, that is, a way that can be taken, whereas the other,
corresponding to the negation of the thesis, is a blind alley leading nowhere. After presenting both ways in fragment 2, Parmenides takes up his
formulation again in fragment 6. This time the presentation of both ways
is a bit different, because they have already been proposed as possibilities,
worth considering a priori. Now it is a question of showing that one of the
ways necessarily must prevail, because it offers the basis of all thought and
all speech, whereas the other way must be abandoned because it contains
an internal contradiction.
Fragment 6 sets out this new formulation, but scholars differ appreciably in their interpretations of it. All the interpreters agree that fragment 6
begins by expressing the necessity of thinking and saying that that which
is being, is (fragment 6.1a), then goes on to expound the thesis and its
negation again, that is, the two ways of investigation already presented in
fragment 2. But interpretations begin to differ when it comes to specifying
which lines each of the ways is described in. An analysis of the huge bibliography devoted to this subject enables me to state that, in fact, there are two
possible interpretations. The great majority of scholars nd the formulation
of the thesis (i.e., the rst way) in the second hemistich of line 1 of fragment
6 (i.e., in 6.1b) and the formulation of the negation of the thesis (i.e., the
second way) in the rst hemistich of line 2 of fragment 6 (i.e., 6.2a). That
position is mainly based on the presence of the term nothing (meden) in
6.2a, whichaccording to that viewdemonstrates that this way is concerned with not-being. So this means that, according to these interpreters,
there is already a formulation of two ways in the rst two lines of fragment 6. A
secondary consequence of this interpretation has devastating effects for the
understanding of Parmenides philosophy, because from line 6.3 onward,
there can be no doubt that another way of investigation is presented, and
as these interpreters have already found two ways at the beginning of fragment 6, a third way naturally appears.

98

(a) 6.1b2a Reintroduces the First Way of Investigation

A few interpreters (including myself) also nd the formulation of two

ways of investigation in fragment 6 (i.e., the thesis and its negation), but
now the presentation of the rst way takes place in the whole of 6.1b2a,
that is, at the end of line 1 and the beginning of line 2 of fragment 6. This
would mean that in the rst two lines of fragment 6 only one way is being
presented. Therefore, the presence of the term meden in line 6.2a is reintroducing the formula me` enai from line 2.3b, in which Parmenides stated
that it is not possible not to be. Here, as it is said of nothing[ness] that
it does not exist, line 6.2a is reformulating the double negation of the
positive thesis, that is, the rst way. For this position, the beginning of fragment 6 presents one single way, the true one. So the new way that Parmenides
presents immediately afterward is the second way. There is no room for a
third way, either theoretically or practically. So let us look in detail at
lines 6.1b and 6.2a, because it is this passage that determines the number
of ways that we nd in Parmenides.408

(a) 6.1b2a Reintroduces the First Way of Investigation

The formula that concludes the rst line of fragment 6 (second hemistich
of 6.1 = 6.1b) continues naturally (. . . gar . . . since, because) in the rst
part of the following line (rst hemistich of 6.2 = 6.2a) (. . . d . . . ). It is a
single formula, split into two lines for the sake of the meter esti ga`r enai/
mede`n douk estin. The variations to be found in the manuscript tradition
are not important. I have already mentioned that the only source for the
passage is Simplicius, but, as we know, various manuscripts of Simplicius
text have come down to us. In 6.1a, the codices Laurentianus 85.1 (B) and
Riccadianus 18 (C) have esti to` enai. In 6.2a, instead of meden d some
codices have the corrupt text medeoid, which does not mean anything
(but which stimulated the imagination of R. Vitali, who proposed the reading me` d od, ma non vedo).409 However, this can be explained by a
lowercase transliteration, at a given moment, of the primitive manuscripts,
which, as we know, were written only in capital letters.
Before pronouncing on the thesis (or theses) formulated in this passage,
we must analyze the syntactic structure of 6.1b2a. Essentially, the interpretations differ according to the value that they place on the conjugated verbs
408

409

An impartial observer would say that this question is irrelevant, since when Parmenides
presents the two possibilities in fragment 2, he says that they represent the only (mounai) ways of investigation. The only ways are just two.
Vitali (1977), 35.

Presentation of the Thesis and Its Negation in Fragments 6 and 7

99

esti and ouk estin, whose subjects are enai and meden. The main
question is this: does the verb esti have the same value in 6.1b and 6.2a? In
6.1b, as the subject of esti is an innitive, enai, the verb must have potential
value: It is possible . . .; in 6.2a, on the other hand, the subject, meden, is
not an innitive, and given that it has no attribute, in such a case the verb
generally has an existential meaning. Therefore, on the basis of this difference, some authors have translated the verb in a different form in each case
(position A), whereas other authors do not take these nuances into account
(position B) and consider that the translation should be the same in both
lines.
Position B already emerges clearly in the rst translations of the Poem:
namque est ens, nihil vero non esse (S. Karsten);410 car letre existe et le
non-etre nest rien (F. M. Riaux);411 denn das Sein existiert, das Nichts
existiert nicht (H. Diels).412 It also reappears in more modern translations:
There is Being, Nothing is not (L. Taran);413 denn das Vorhandensein ist
vorhanden, Nichts aber ist nicht vorhanden (J. Klowski);414 car il y a etre,
et rien il ny a pas (M. Conche);415 denn Sein gibt es, Nichts aber gibt es
nicht (J. Wiesner).416 However, some of these authors have interpreted 6.1a
differently than I have and made that which is being (eon) the object of
think and say: it is necessary to say and to think that which is being.417
For that reason, these authors nd in 6.1b the cause of the necessity of
thinking that which is: it is possible because there is being (or because
being exists) and nothing[ness] is not.418
There can be no doubt that position B offers an excellent reading of the
passage, but in nearly all the examples enai is regarded as a noun, whereas
it is not clear that this is so.419 As I have said several times, Parmenides
expresses the central idea of his system in very different ways: by means
of an innitive (enai), a conjugated verb (estin), a participle (eon), and even
410
411
412
413
414
415
416
417
418

419

Karsten (1835), 35.

Riaux (1840), 211.
Diels (1897), 35.
Taran (1965), 54.
Klowski, J., Die Konstitution der Begriffe Sein und Nichts durch Parmenides, KantStudien 60 (1969) 414.
Conche (1996), 100.
Wiesner (1996), 252.
These scholars considered that emmenai was related to khre. This was the case with Taran
(Taran [1965], 58).
Untersteiner suggests a very personal version of the passage. He states that in 6.2a, the
same innitives should be understood as in 6.1a: on the other hand, [thinking and speaking] nothing[ness] does not exist (Untersteiner [1958], 135).
Unless we adopt the version given in some Simplicius manuscripts (cf. above), to` enai, a
hypothesis that no one has held.

100

(a) 6.1b2a Reintroduces the First Way of Investigation

an innitive used as a verbal noun (to` pelein).420 All these terms are synonymous, from the semantic point of view.421 However, with regard to the syntax,
each term must be analyzed in function of the value that Parmenides gives
it, and in that case an innitive is an innitive and not a noun. Bear in
mind that this syntactic specicity of the terms forms part of the Parmenidean method, which aims to bring out the wealth of nuances in the verb
to be.
In 6.1b, as in fragment 3, enai is an innitive.422 If this is so, the expression esti ga`r enai is made up of the verb to be in the third person (esti),
and to be (enai) as the subject. In this case, esti must be read as modal
(as was the case in 2.3b, ouk esti [it is not possible] me` enai [not to be])
in 6.1b, but not in 6.2a, and therefore the translation of the verb must be
different in each hemistich. This position offers a more coherent understanding of the text, because the necessity of saying and thinking that that
which is being exists, proclaimed in 6.1a, is based on the causal statement
made in 6.1b: that which is exists necessarily because it is possible to be.423
But above all, this position takes into account the fact that the subject of the
verb esti/ouk estin is syntactically different in 6.1b and 6.2a: an innitive in
one case, a noun in the other. In fact, esti can only have modal value (as in
2.3b) when its subject is an innitive, and therefore the only possible translation of esti ga`r enai is since it is possible to be. On the other hand, this
possibility does not exist when the subject is a noun, which prevents us
translating mede`n douk estin as nothing[ness] is not possible. As G. Calogero observes, this formula simply means and nothing[ness] does not exist.424 Various scholars have translated the phrase thus: Es ist namlich
moglich su sein, Nichts aber ist nicht (K. Bormann);425 es muss sein; denn
Nichtsein ist nicht (K. Riezler);426 and denn das Sein kann sein; Nichts ist
nicht (U. Holscher),427 among others.428 My position is as follows: 6.1b presents the possibility of being as the cause of the necessity of saying and
thinking that that which is being, is (6.1a), and this possibility is conrmed
420
421
422
423
424
425
426
427
428

An extreme case could be 6.8, where to` pelein is coordinated with ouk enai, and it could
be said that, by hendiadys, the latter innitive is also used as a noun.
That is why we do not hesitate to use the expressions being, that which is being,
present existence, etc. as synonyms here.
Just as the terms me` enai used in fragment 2 were innitives, which we have always
translated as not to be and never as a nonexistent [the] non-being.
Bormann (1971), 75.
Calogero (1936), 159, note 3.
Bormann (1971), 37.
Riezler, K., Parmenides (Frankfurt am Main: Klosterman, 1934), 31.
Holscher (1969), 17.
Cf. also Cornford (1939), 31; Ranulf, S., Der eleatische Satz vom Widerspruch (Copenhagen:
Gyldendal, 1924), 161; Mansfeld (1964), 81; Loew, E., Das Verhaltnis von Logik und
Leben bei Parmenides, Wiener Studien 53 (1935) 11.

Presentation of the Thesis and Its Negation in Fragments 6 and 7

101

because nothing[ness] does not exist (6.2a). Thus the translation of the passage is: since it is possible to be, whereas nothing[ness] does not exist.
Without any doubt, this refers to the expression of the thesis contained in
line 3 of fragment 2, but because this evidence has not been accepted by a
great many researchers, we must analyze this passage at length.

(b) Relation Between 6.12 and Fragment 2

The beginning of fragment 6 completes the reasoning begun in fragment 2.
We saw that in line 2.3 Parmenides expounded his rst thesis (broadly
speaking, being exists and it is not possible not to be), and after having
considered it as the way of persuasion because it accompanies truth, he
presented the formula for the second thesis (broadly speaking, being does
not exist and it is necessary not to be). This thesis is rejected because that
which is not, which is impossible, is unthinkable and unsayable (2.78).
In fragment 6, Parmenides inverts the terms of this rst refutation of the
second thesis (other criticisms appear throughout the Poem, relating especially to its nonviability). In this way this criticism, now negated, becomes
the necessary characteristic of the rst way: It is necessary to think and to
say that that which is being exists (6.1a). Why? Since it is possible to be,
whereas nothing[ness] does not exist (6.1b2a). This way of arguing follows the following pattern:
2.5:
2.78:
6.1a:
6.1b2a:

statement of the second way

basis of the error of the second way
basis of the possibility of the rst way
statement of the rst way

The Goddess urges the proclamation of this statement of the rst way,
since, as we shall see, it is not enough merely to listen to it. Long habit (7.3)
drives mortals into a blind alley with no way out. To understand the scope
of this true command of the Goddess, we must briey look at the Greek
text of the formula phrazesthai anoga, I order to proclaim, which is really
a Homeric and Hesiodic cliche.429 In the case of Parmenides, the phrase
contains a direct object ta, these things (plural: we will come back to
this detail), and the sentence is completed by T. Bergks unanimously
adopted conjecture: s, a pronoun alluding to the hearer, [I] order you
[s] to proclaim. This conjecture makes no sense. The critical apparatus
presented by Diels offered other possibilities, ta gego (D), tou ego (E),
and ta ge (F). If to these possibilities we add others, veried by me (and
429

Cf. Homer, Od. 1.269, 13.279, 16.312, 20.43, 23.122; Hesiod, Works 367.

102

(b) Relation Between 6.12 and Fragment 2

ignored by Diels): ta` ego (Mut. 184 [= III.F.6]), ta` dego (T), and ta se
(P), we see that the pronoun se only occurs in one instance, and that this
one does not respect the meter. Doubtless, it was to correct this drawback
that Bergk added the pronoun ego, which occurs in various versions, and
proposed a hybrid conjecture that fuses two different readings. However,
the conjecture is also dangerous, because it restricts the exhortation to the
hearer; he is the one who is to proclaim what he has just heard. We have
followed the text of codex D, which most agree, even Diels, is the most
important one in the basic DEF group.430 It presents not only an acceptable
text but one that is coherent with the meaning of the Poem. The Goddess
orders, in general, that what she has just said should be proclaimed.
The content of the proclamation, referred to by the pronoun ta, is the
statement of the rst way of investigation: since it is possible to be,
whereas nothing[ness] does not exist (6.1b2a, i.e., two statements, which
is why the Goddess uses the plural, although, for convenience, the formula
can be translated as this).431
But this new formulation adds certain nuances exclusively concerning
the syntactic value of the expressions. To put it another way, the presence
of the subject enai in 6.1b confers a potential character on the verb esti,
which had an existential value in 2.3a, where it stood on its own; but, inversely, the use of the noun meden in 6.2a takes away from ouk estin the
modal value that this verb had in 2.3b, where the innitive me` enai was its
subject. Thanks to this sort of interchange of terms, the new formulation of
the Parmenidean thesis in 6.1b2a is completely complementary, term for
term, with its rst formulation: each of the parts of the verbal statement
(which is the basic one, since Parmenides starts from is) gains the nuance
that it lacked: the existential estin of 2.3a gains a modal value in 6.1b, and
the modal ouk esti of 2.3b gains an existential value in 6.2a. This double
formulation of Parmenides rst thesis reinforces the identity it already possessed in its rst formulation in the two hemistiches: 2.3a was conrmed
by the double negation in 2.3b, just as 6.1b is conrmed by the negation
(ouk esti) of a negative term (meden = me` den) in 6.2a. If we link the modal
and nonmodal elements of both formulations of the thesis, its complete

430
431

Diels, H., D doctior est, Preface to Simplicius Commentary on Aristotles Physics, 6.

The correlation since (gar), whereas, can be interpreted as a coordination, and.
Cf. Denniston: in such cases there is no substantial difference between de and ka
(Denniston, J. D., The Greek Particles [Oxford: Clarendon Press, 1934], 162. Cf. another
example in Parmenides: 1.45. Wiesner has criticized my position in this respect (Wiesner
[1996], 82, note 392). For me, even if the two parts of the formula are adversative, that is
because they are concerned with opposing objects, but both form part of the same
thesis: it is said of being that it is, whereas (if we want to stress the adversative value) it
is said of nothing[ness] that it is not. And that is true.

Presentation of the Thesis and Its Negation in Fragments 6 and 7

103

statement is as follows: there is being because it is possible to be, and there is no

not-being because non being is not possible. Accepting this authentic apology
for the fact of being makes the Goddesss hearer become a man who
knows (1.3). And that is why the Goddess addresses her listener and tells
him, I order to proclaim this. This is the content of the rst way, which,
as was said in lines 2.4, is the way of persuasion (petho), since it accompanies the truth. That means that for Parmenides, truth persuades, and
this persuasion is strong enough to become an obligation. The Goddess
orders the proclamation of a truth that becomes imperative because it is
capable of persuading. So we must devote a few lines to the links, in Parmenides, between truth and persuasion, and given the disjunctive character
of his thought, to the relationship between error and the notion of deception (apate), which, as we shall see, is set against persuasion.

(c) Truth, Persuasion, and Deception

When we analyzed the allegorical presentation of the two ways in fragment
1, we saw that the Daughters of the Sun persuaded (pesan, 1.16) Dike to
open the heavy gates giving entry to the realm of truth. And we also saw
that in order to persuade her, the maidens had to offer enveloping arguments (logoi). Thus, from the beginning of the Poem, persuasion and logos
go together. In each case, persuasion will work, according to the type of
reasoning that is presented on each occasion. To convince the Goddess,
enveloping arguments were enough; to convince the future philosopher,
the arguments must revolve around the truth. Let us recall the passage
from fragment 8 in which the Goddess ends the rst part of her philosophy
course: here I end for you my trustworthy reasoning (logos) and thought
(noema) about the truth (8.5051). The logos about the truth is trustworthy
(piston); it assumes a true conviction (pstis, 1.30, 8.28), capable of rejecting
the possibility of either origin or destruction of that which is being: the
force of conviction (pstios iskhu`s, 8.12) does not allow it either to be born
or to die. A. Oguse demonstrated the close relationship between the verbs
pethomai (allow oneself to be persuaded) and pisteuo (trust),432
which are practically synonymous in Parmenides.
But even here Parmenides thought is presented in the form of an alternative. So, as only truth persuades, non-truth (which in his dichotomous
scheme is opinion) deceives. Let us recall that as soon as she has ended
her true argument, the Goddess announces that now she will expound the
432

` propos de la syntaxe de petho et de pisteuo, Revue des etudes grecques 76

Oguse, A., A
(1965) 31.

104

(c) Truth, Persuasion, and Deception

opinions of mortals, which will consist of a deceitful order (apatelon) of

words (8.52). The alternative between persuasion and truth on the one hand
and deception and opinion on the other corresponds to the Parmenidean
double message, which, as I said, concerns both an account of the truth of
the fact of being and a description of the error of opinions, and just as the
former convinces, the latter can deceive. Therefore precautions must be
taken: I tell you of this probable cosmic order so that no viewpoint of
mortals will prevail over you (8.6061). The power of conviction must
prevail over the deceitfulness of opinion.
Several times I have already said that Parmenides inherits schemes
from traditional epic. Persuasion and deception as ambivalent aspects of
logos are found in the Homeric poems, in which M. Untersteiner sees a
legacy of ancient conceptions of the magic power of words.433 In Iliad 9, for
example, Achilles complains to Odysseus that Agamemnon deceived him
(apatese) when he seized his female slave, and he lets his anger y: Dont
let him now try to persuade me (pesei)! (345). And in Odyssey 14, Odysseus
tells how a Phoenician expert in deceit (apatelia) had persuaded him (parpepithon) to go to Phoenicia (28890). The same pair appears in the Homeric
Hymn to Zeus, in lines 7 and 33. The author of this work says that, despite
her power, Aphrodite, who can prevail over both men and gods, is unable
to persuade (pepithen) and deceive (apatesai) the trilogy made up of Athena,
Artemis, and Hestia, the incorruptibles. J. P. Vernant, who looked at this
passage in a famous article, wrote that Aphrodite possesses this power because she is accompanied by the goddesses Peitho and Apate.434 Vernants
derivation of the verb petho from the goddess Peitho is completely legitimate, because this goddess often appears together with Aphrodite, and her
function is to exercise erotic persuasion. R. G. A. Buxton observes that it
is Peitho who in Hesiod (Works 7374) offers Pandora the jewels that will
awaken desire in men, especially in the naive Epimetheus.435 Peithos power
of enchantment will be taken up again by the tragic authors and, after the
Parmenidean interval, it will descend directly to Gorgias Helen. If we look
at the pre-Parmenidean literature, we can sum up, with Mourelatos, that in
every context Peitho expresses the power residing in an agent, a power that
is perceived as a sort of attraction exercised over a patient.436 For M. Detienne, Peitho is without doubt the power of the word as it is exercised
over another, its magic, its seduction, as the other feels it.437
433
434
435
436
437

Untersteiner, M., Les Sophistes, (French translation), Vol. I, (Paris: Vrin, 1993), 164.
Vernant, J. P., Hestia-Herme`s, in Mythe et pensee chez les Grecs, Vol. 1 (Paris: Maspero,
1971), 130.
Buxton, R. G. A., Persuasion in Greek Tragedy. A Study of Peitho (Cambridge, England:
Cambridge University Press, 1982), 3637.
Mourelatos (1970), 139.
Detienne, M., Les matres de verite, 62.

Presentation of the Thesis and Its Negation in Fragments 6 and 7

105

As for Apate, we may say that as a goddess she is characterized by the

will to deceive. Deception means the deliberate desire to pass one thing off
as another, and in this sense, Apate, as M. Untersteiner remarks, has an
even stronger signicance than pseudos, which originally only means taking
one thing for another.438
The Parmenidean alternative is as follows: either persuasion or deception. It is the presence or absence of truth that decides whether the logos
convinces or, on the contrary, deceives. In order to convince, the logos must
offer proofs, evidence, arguments; thanks to these, long habit will be deprived of its deceiving power. Against what is said Parmenides sets
what must be said.

(d) The Exhortation to Proclaim That It Is Possible to Be

and That Nothing[ness] Does Not Exist
If the Goddess urges the proclamation of the thought-content expounded
in lines 6.1b2a, it is, of course, because this is her thesis, that is, the rst
way of investigation. Although this is obvious, very few scholars have admitted that the way stated in 6.1b2a is the rst way of investigation, or Parmenides thesis. The ones who have include, in particular, L. Taran, who
vigorously maintained this view; R. Kent Sprague; K. Bormann; and R. Vitali. The vast majority of scholars of Parmenides thought distinguish between 6.1b, where they nd a true thesis (which is more than obvious) and
6.2a, where, according to them, Parmenides suddenly goes on to talk about
a wrong way, since they claim the formula mede`n d ouk estin (nothing[ness] does not exist) is an expression of the negation of the thesis, that is, the
second way. First of all, let me say that even if a strongly adversative content is ascribed to the particle de in 6.2a and it is translated as on the
other hand, however, nevertheless, and so on, the formula nothing[ness] does not exist (or according to the scholars who mistakenly give
this verb a modal value, nothing[ness] is not possible)439 is indubitably
true, since it is merely a reformulation of it is not possible not to be (2.3b),
a persuasive thesis, since it is accompanied by the truth (2.4).
So we may wonder why the interpreters almost unanimously follow
this tendency to consider that in 6.2a Parmenides is alluding to the way of
error? Doubtless it is because of the notion of meden, nothing[ness]. For
this reason there are scholars who consider that here Parmenides is presenting the way of non-being. This is the case with F. M. Cornford,440 W.
438
439
440

Untersteiner, Les Sophistes, 166.

Later, perhaps on the basis of Parmenides, Aristotle writes: that that which is, is, and
that that which is not, is not, [is] true (Met. G.7.1011b27).
Cornford (1939), 32.

106

The Exhortation to Proclaim That It Is Possible to Be

Jaeger,441 U. Holscher,442 W. R. Chalmers,443 H. Frankel,444 Gomez-Lobo,445 J.

Wiesner,446 and P. A. Meijer,447 among others. According to these scholars,
ber6.2a is the way that states the existence of nothing[ness].448 Or, as U
weg maintains, it is the way that holds that besides being, there exists
nothing[ness],449 but this way [which one must take ad sensum to be the
way of thinking that nothing can exist] is the rst to be avoided.450 Doubtless all these arguments are valid, but applied to the wrong way, that is,
the thesis stated in 2.5, which will reappear later in 7.1. They are also valid
applied to the notion of not-being, or nothing[ness]. But in 6.2a there is no
such notion: there is a phrase; something is said about nothing[ness]. What
is said? It is said that nothing[ness] does not exist (meden douk estin). This
formula states exactly the opposite of what we have just met in the arguments quoted above. It does not say nothing[ness] can exist, or that
nothing[ness] exists or that there is nothing[ness]. It says the opposite.
The copulative or existential verb appears negated: nothing[ness] does not
exist, nothing[ness] is not possible, there is no nothing[ness]. These
latter phrases are not only true at rst sight, but have already been proclaimed as true by the Goddess herself in line 2.4, since they are all simply
reformulations of 2.3b, there is no not-being, it is not possible not to
be, non-being is not.451 It is because this is a true thesis that the Goddess
orders these things are to be proclaimed.452 If 6.2a had expressed the second
way, that is, an erroneous thesis, the Goddess would not have been able to
order that its content should be proclaimed, since she had already said that
that which is not is not utterable or proclaimable (oute phrasais, 2.8;
the verb is the same: phrazo). The only thing that can be said, thought, and
proclaimed is that there is being; that that which is being, exists; or, if you
prefer, that nothing[ness] does not exist, that it is impossible not to be.
I have already said that I am against calling the rst way the way of
441
442
443
444
445
446
447
448
449
450
451
452

Jaeger (1947), 100.

Holscher (1969), 85.
Chalmers, W. R., Parmenides and the Belief of Mortals, Phronesis 5 (1960) 7.
Frankel (1951), 404.
Gomez-Lobo (1985), 95.
Wiesner (1996), 86.
Meijer (1997), 153.
Cf. Untersteiner (1958), cxi.
berweg, F., Grundriss der Geschichte der Philosophie, Vol. I, 12th ed. (Basel: Schwabe, 1953),
U
84.
Guthrie (1965), 22.
On the equivalence between not-being and nothing[ness] in Parmenides, cf. notes
324 and 330 in Chapter IV.
It tends to be forgotten that the Greek text has a neuter plural:ta. As Taran observes,
the exhortation refers to the two preceding statements: it is possible to be (6.1b) and
whereas nothing[ness] does not exist (6.2a) (Taran [1965], 60).

Presentation of the Thesis and Its Negation in Fragments 6 and 7

107

being and the second way the way of not-being, since what characterizes each way is not a notion (being, not-being, or nothing[ness]) but
what is said about it. For this reason the rst way is a thesis, and the second
is the negation of that thesis; and in each of the two ways (both in the thesis
and its negation) the notions of being and not-being are present. It is the
kind of relationship that Parmenides establishes between the notion in
question and the fact of being that determines the truth or falsehood of the
thesis in question. When it attributes a contradictory notion to a notion, the
thesis is false; when an analogous notion is attributed to each notion, then
the thesis is true. The rst way is valid because it holds that there is being,
but also because it states that it is not possible not to be. The negation of
the thesis is erroneous because it states that there is no being, but also
because it maintains that it is necessary not to be. As we can see, in the
thesis, which is true, the notion of not-being is present, and in the negation of the thesis, which is erroneous, the notion of being is present. It is
not the presence of a notion that characterizes each thesis but what is said
about it. The notion of nothing[ness] is present in the rst, the true thesis,
because its existence is denied; it is true to deny the existence of nothing[ness].
On the other hand, stating that nothing[ness] exists assumes that the second
way is being followed and this way contradicts the rst. This rst way is
clearly and distinctly expressed in 6.1b2a.
I said that very few scholars admitted this evidence. R. Kent Sprague
states that in 6.1a2b there is a recommendation of the way of being;453
K. Bormann holds that contrary to what the second way states (Das Nicht
ist), 6.2a belongs to the rst way, which is the true one;454 and we can
deduce from R. Vitalis strange translation that this scholar also saw a formulation of the rst way in 6.12 (Vedo infatti essere le` ma non il non e`,
cio` che io ti invito a considerare ovviamente di questa prima via di ricerca).455 But it is L. Taran who offers the most solid arguments in this
respect: ta (6.2b) in the plural must allude to more than one phrase, and
given that all these clauses must refer to a single way, this has to be the
rst way of inquiry, because we have shown that mede`n douk estin may
refer to it while it is impossible for esti ga`r enai to be a part of this way.456
If Parmenides urges the proclamation of this fundamental thesis, why, in
the following line, does he tell his listener to withdraw from this way?
This delicate question deserves special treatment.

453
454
455
456

Kent Sprague, R., Parmenides: A Suggested Rearrangement of Fragments in the Way of

Truth, Classical Philology 50 (1955) 125.
Bormann (1971), 97. This scholar adds that the second way says the opposite of 6.12.
Vitali (1977), 35.
Taran (1965), 60.

108

Parmenides Does Not Recommend Withdrawing

(e) Parmenides Does Not Recommend Withdrawing

from the Thesis Expounded in 6.1b2a
So in 6.12 we nd a new formulation of the rst way, that is, the true
way, accompanied by an exhortation to proclaim or manifest this truth. The
following line unequivocally alludes to this way by means of the demonstrative this (tautes): from this rst way of investigation I <withdraw>
you (protes gar saphhodou tautes dizesios <ergo>).457 The manuscript tradition has transmitted this line in a mutilated form, and the verb withdraw,
which stands between angular brackets, was added by H. Diels as a conjecture. So according to the present state of the text, the Goddess exhorts the
disciple she is addressing to withdraw from this way of investigation, that
is, from the way that, as we have seen, is the true way. As Taran observes,
it is probable that this prohibition led the vast majority of scholars to consider that the condemned way was the second way, the way of error,458 and
then adapt the translation to t its negative character. Nevertheless, there
can be no doubt that this way, mentioned in the preceding line, is the
way of truth. How can this obvious fact be harmonized with the Goddesss
order to withdraw from it? The scholars who saw in 6.2a a formulation
of the way of error did not ask themselves this question, but as we saw,
this position is unsustainable. Given the clarity of the text, we must try to
explain why the Goddess orders withdrawal from a true way. I believe
there are four possible explanations, three of which have been proposed by
scholars who were aware that this passage presents a serious problem: (1)
the contradiction is eliminated if the passage is analyzed in a very free way,
almost allegorically; (2) the lines in the passage could be rearranged; (3) it
is possible to imagine a gap after line 6.2 so that the term this would not
refer to the way of truth; and, nally, (4) it is possible to question Dielss
conjecture, the source of all the problems.
(1) For K. Bormann, there is no problem whatever in the passage,
whose spirit would be, you have to follow the way of nothing[ness] is
not; if you do not, a rst false way appears.459 Doubtless, the rst part of
Bormanns argument can be found in Parmenides text, but there are no
457

458
459

The reference to the previous way does not necessarily depend on the term rst (protes),
since Parmenides never calls the way mentioned rst by the Goddess the rst way. It
depends on the demonstrative this, which refers to what has just been said immediately
beforehand.
Taran (1965), 59.
Bormann (1971), 98. For Heitsch, the prohibition refers to the phrases negative term, ouk
estin, which is not even a way (Heitsch [1979], 87). We may answer that in 2.5a ouk
estin is a way, and although in the alternative in line 8.15 ouk estin has no subject, in
6.2a the subject is meden, nothing[ness] and that, thanks to the double negative, the
negative way ouk estin becomes positive: ouk estin . . . meden.

Presentation of the Thesis and Its Negation in Fragments 6 and 7

109

indications for the conditional phrase. The Goddess withdraws the disciple
from that which she has just presented and not from a hypothetical disobedience of her order. W. Kranz had already offered a similar solution to
Bormanns when he translated 6.3 thus: dies (die Annahme von Nichtsein)
ist namlich der erste Weg der Forschung, von dem ich dich fernhalte.460
Here, too, the solution is arbitrary, since the phrase in brackets does not
exist in Parmenides text and there is no right to imagine what the philosopher might have thought without venturing to say it. Finally, for A. P. D.
Mourelatos, 6.12 presents both the rst way of investigation and the doctrine that reinforces it by denying the second way.461 But precisely this rejection of the second way (the negation of the negation of the thesis) is true,
and Parmenides cannot withdraw us from the truth.
(2) The second possibility was attempted by R. Kent Sprague. Aware
that the rejection expressed in 6.3 could not refer to a true way, this scholar
proposed inserting between 6.2, where there is a true way, and 6.3, where
there is the idea of rejection, a line alluding to a clearly wrong way: line
7.1, which states a way according to which there are things that are not
(enai me` eonta). Immediately after the last line of fragment 6, Kent Sprague
places fragment 7, but now with fragment 7 starting from line 3 (she has
already set line 7.1 between 6.2 and 6.3, and she considers 7.2 to be inauthentic, a paraphrase composed by Plato).462 In fact, Kent Sprague has inherited an old tradition that already envisaged the independence (even, in
some cases, the doubtful character) of lines 7.12. G. G. Fulleborn had already eliminated 7.1 since, according to him, this line was prosaica non
nulla Parmenidis dicta,463 and consequently he placed 7.26 after fragment
1. S. Karsten did the same (he regarded 7.1 as sententiam Parmenideam
Platonis verbis expressam).464 And so did F. M. Riaux (who states that the
line is inauthentic).465 On the other hand, H. Diels admits both the authenticity and autonomy of the group 7.12, and like Fulleborn, sets the rest of
fragment 7 after fragment 1, according to the evidence of Sextus (VII.111).466
This viewpoint was shared by J. Burnet467 and later by G. Calogero, who
proposed placing 7.12 before fragment 6.468
Kent Sprague takes her lead from these examples, but her originality
lies in placing 7.1 between 6.2 and 6.3. It is true that then the passage ac460
461
462
463
464
465
466
467
468

Kranz, W., Vorsokratische Denker, 3rd ed. (Berlin: Weidmann, 1959), 95.
Mourelatos (1970), 77, note 7.
Kent Sprague, Rearrangement, 125.
Fulleborn (1795), 98.
Karsten (1835), 81.
Riaux (1840), 230.
Diels (1897), 34.
Burnet, J., Early Greek Philosophy, 4th ed. (London: A. & C. Black, 1930), 174.
Calogero (1932), 20.

110

Parmenides Does Not Recommend Withdrawing

quires a monolithic coherence: . . . whereas nothing[ness] does not exist; I

exhort you to proclaim it (6.2), since this can never be maintained: that
there should be things that are not (7.1); I <withdraw> you from this rst
way of investigation (6.3). But up to what point can we consider that Parmenides Poem is a sort of puzzle that can be arranged as we like? Simplicius, who says himself that he has Parmenides Poem in front of him, quotes
the whole of fragment 6 without interruption (Phys. 117). Kent Sprague proposes inserting 7.1 into the middle of it. We could assume a gap in Simplicius text.469 But it is difcult to see how the content of this gap (viz. line 7.1)
could have reappeared in three more passages of Simplicius (Phys. 135, 143,
and 244) always accompanied by 7.2. But at the same time, we should not
forget that Simplicius knows that Plato cites 7.12 (in Soph. 237a and 258a)
and that perhaps, for that reason, he does not venture to separate these two
lines to allow 7.1 to become independent, as Kent Sprague claims. Moreover, Simplicius resistance to separating 7.1 from 7.2 is also doubtless due
to the fact that the philosopher considers that 7.2 is authentic, contrary to
the opinion of Kent Sprague. For all these reasons, I believe that Kent
Spragues solution could be taken into account only if the problem in question leads us down a blind alley with no way out. But, as we shall see,
there is a possible way out.470
(3) In order to avoid the obvious contradiction between a true statement and the invitation to withdraw from it, some have assumed the existence of a gap after line 6.3. This hypothesis was proposed by O. Becker,
for whom the whole passage 6.13, as it has come down to us, is incomprehensible.471 According to Becker, the demonstrative this (tautes) alludes to
a nearby notion, but this nearness can be either before or after it, and he
chooses the second possibility. If this is so, the Goddess would withdraw
the disciple from the way stated after 6.3, in a passage now lost which
would have begun with I withdraw, a word miraculously added by
Diels, and which would continue in a hypothetical line 6.3a, which might,
for example, have this form: in no way can this be said or thought: that
469

470

471

Taran states that if there is a gap, this must come after line 6.3 and not between 6.2 and
6.3 (Taran [1965], 60). Nevertheless, this hypothesis does not refute Kent Spragues thesis, since there could be a gap between 6.2 and 6.3 (and Taran gives examples of fragmentary quotations by Simplicius) and another, perhaps smaller one, at the end of line 6.3.
Before abandoning this possibility (b), we may note that Bicknell proposed another rearrangement of the fragments, but his hypothesis does not concern our passage. This
scholar maintains that 6.39 should be completed by fragment 4, and then fragment 8
should follow, but he does not say what kind of relationship there is between 6.12 and
the new fragment made up of 6.3 ff. plus fragments 4 and 8 (Bicknell, P. J., A New
Arrangement of Some Parmenidean Verses, Symbolae Osloensis 42 [1968] passim).
Becker, O., Drei Abhandlungen [but Bemerkungen in the contents] zum Lehrgedicht
des Parmenides, Kant-Studien 55 (1964) 256.

Presentation of the Thesis and Its Negation in Fragments 6 and 7

111

there are things that are not (that is, a mixture of 6.1 and 7.1). If this is so,
this means that in 6.3 Parmenides would be saying: I withdraw you from
this rst way of investigation (6.3); in no way can this be said or thought:
that there are things that are not (6.3a); and then also . . . (6.4). This idea
of Beckers is ingenious, but I do not understand why his attempt at reconstruction is really an attempt to justify the notion of withdrawal, which
only arises from the hypothetical ergo proposed by Diels. If Becker quarrels with the traditional textas he doeshe should begin by questioning
Dielss conjecture, which forms part of the traditional version even though
it does not belong to Parmenides.
A year after Becker but independently, L. Taran also asserted the existence of a gap at the end of line 6.3, much larger than that imagined by
Diels. According to Taran, this gap resulted from Simplicius habit of only
retaining passages that were closely concerned with the topic he was analyzing. The object of study on page 117 of his Commentary on Aristotles
Physics is the identity of being and not-being, and in the lines he did not
quote, according to Taran, there would have been a reference to the second way of inquiry or some word (e.g., now), which coming after [ergo]
would qualify the temporary abandon of the rst way.472
Taran is aware of the fact that this is a possibility and, as such, cannot
be rejected completely.473 Nevertheless, it is highly improbable that Simplicius would have omitted a whole passage from Parmenides. If Simplicius
only mentioned the lines closely linked to the topic he was expounding, his
quotation in our case should have begun with line 6.3, since 6.12 says
nothing about those who consider being and not-being to be identical.
Moreover, Simplicius himself says that he intends to quote the greatest possible quantity of passages from Parmenides, given the rarity of the book
(Phys. 144), and he keeps his word, since he goes on to quote a block of
fty lines from fragment 8. Finally, we may say that both aphhodou (6.3)
and apo` tes (6.4) depend on the same verb, whether we accept Dielss conjecture or any other, and with or without a gap, the meaning of 6.3 is picked
up again in 6.4. And in 6.4 the wrong way is presented. So the problem
subsists, since how can the same verb be applied to two contradictory ways,
especially if it refers to a rejection?
472
473

Taran (1965), 61.

Brumbaugh accepts the hypothesis (Brumbaugh, R. S., Review of Taran, International
Philosophical Quarterly 4 [1966] 496); Long rejects it (Long, A. A., Review of Taran, Journal of the Hellenic Studies 86 [1966] 223). According to Mourelatos, the verb ergo, accepted
by Taran, has too strong a meaning to assume that it could apply to a merely temporary
abandonment of a way (Mourelatos [1970], 77, note 7). Finally, Heitsch accepts the gap
and adds other quotations from Simplicius in support of Tarans hypothesis (Heitsch, E.,
Gegenwart und Evidenz bei Parmenides: Aus der Problemgeschichte der Aequivokation [Wiesbaden: Franz Steiner Verlag, 1970], 42 ff.).

112

(f) The Origin of the Notion of Withdrawing as a Conjecture in 6.3

It cannot be denied that the Taran-Becker hypothesis is ingenious, but

it has to be admitted that these are desperate, even dangerous, measures.
Indeed, we can postulate a gap with respect to any passage in the Poem, a
providential gap that would give the text its (true?) meaning, that is, the
one desired by whomever is making the conjecture. My position in this
matter is as follows: the text must speak for itself, and if at any given moment there is a silence (corresponding to a gap), elements affecting the continuity of the argument, like the withdraw conjectured by Diels, should
not be introduced into it.
(4) The last possibility consists of critically analyzing the hypothesis
proposed by Diels. Everything I have said in the previous pages offers us
a rm starting point. In 6.12 there is an unequivocal statement of the rst
way, and line 6.3, by virtue of a conjecture by Diels, states that it is necessary to withdraw from this way. We are facing a contradiction, but Parmenides is innocent; the suggestion to withdraw from a true way does not
belong to him. It belongs to Diels.

(f) The Origin of the Notion of Withdrawing

as a Conjecture in 6.3
So why did Diels resort to the notion of withdrawing as a conjecture,
and what reasons did he offer in defense of his choice? Let us begin with
the rst point. A conjecture is necessary in this case because all the manuscripts of Simplicius Commentary on Aristotles Physics, the only source of
the passage, present a mutilated text. I have personally consulted all these
manuscripts, and I have veried that in every case line 6.3 ends with the
word dizesios (of investigation), followed immediately by the beginning
of line 6.4, auta`r epeit . . . In line 6.3, the last half of the hexameter is
missing, but the fact that the text is transcribed in a continuous manner in
all the manuscripts leads us to suppose that the gap must go back at least
to the archetype of Simplicius text, or perhaps to the Parmenides text that
Simplicius had in his library. So it is obvious that all the copyists who
reproduced this passage found themselves with a text that was already incomplete.474 Therefore, no attempt to restore the lost word can be based on
efforts to decipher a content that has become hidden or distorted by material defects in the manuscripts (for example, a word erased but recoverable
thanks to restoration techniques, a damaged folio, etc.). The terrain is free
and the last word lies with the interpreter.
474

Cf. Cordero (1997), Appendix II, the list of currently existing Simplicius manuscripts that
contain Parmenides line 6.3.

Presentation of the Thesis and Its Negation in Fragments 6 and 7

113

Diels was not the rst to try to offer a coherent version of the passage.
Everything began when Simplicius text containing fragment 6 of Parmenides passed from being in manuscript to being printed. Indeed, once
printing was used to reproduce an identical text in hundreds and later
thousands of copies, manual copying of ancient texts was abandoned. Simplicius book was printed for the rst time by the publishing house of Aldo
Manuzio in Venice in 1526. As may be imagined, a printed edition must
offer a comprehensible text, especially in the case of the Aldine editions
(named for the rst name of the publisher, Aldo), already famous in their
time for the precision and rigor of their editing. That is how, when it came
to printing the mutilated Parmenides text, the editor responsible completed the phrase thus: withdraw thought (erge noema) from this rst way
of investigation. That is, the paternity of the verb withdraw does not belong
to Diels. The text of the 1526 Aldine edition is of exceptional interest for our
investigation into the origin of the lost term, and that is why I have studied
it intensively over a number of years,475 since the printed version depends
on the manuscripts used by the publisher as a model. I have been able to
establish that the copious bibliography dedicated to Manuzios press, as
well as the documents and correspondence of those in charge of it, give us
extremely interesting data about most of the works carried out in 1526;
unfortunately, nothing concrete can be found about the edition of Simplicius we are concerned with. The editor in charge appears to have been Francesco dAsola (Asulanus), Aldos son-in-law, since it is he who dedicates
the edition to Cardinal Hercules Gonzaga in a sort of prologue. However,
unlike what occurs in other cases, nothing is said in that prologue about
the Greek manuscripts used in editing the work.
Today we know the manuscripts of Simplicius Commentary on Aristotles Physics were very plentiful at that time. Even now more than forty
codices are preserved, containing at least the rst book of the Commentary
(in which Parmenides text is to be found), either complete or in fragments.
We may presume that these were even more numerous in Manuzios time.
A search through the repertories of existing manuscripts in the libraries of
Italy at that time showed us that an editor did not usually use a large
number of codices.476 Today any researcher who wants to can examine the

475

476

Cf. Cordero, N. L., Analyse de ledition Aldine du Commentaire de Simplicius a` la

Physique dAristote, Hermes 105 (1977), and Les sources venitiennes de ledition Aldine
du Livre I du Commentaire de Simplicius de la Physique dAristote, Scriptorium 38(2)
(1984).
With regard to the St Marks Basilica library (Biblioteca Marciana) in Venice, most of
the manuscripts bequeathed to it by Cardinal Bessarionincluding all those containing
Simplicius bookwere stored away in chests until 1530, and loans outside the library
were exceptional. Cf. Cordero, Les sources venitiennes.

114

(f) The Origin of the Notion of Withdrawing as a Conjecture in 6.3

forty-odd existing manuscripts, which would have been practically impossible during the period of Manuzios successors (1526; Aldo had died in
1515). Be that as it may, I was able to draw up a list of sources that theoretically could have been used as a model for the 1526 edition, and in all of these
the text of the line in question is cut short.477
So where does the expression erge noema (withdraw thought) come
from? We do not know. Clearly the editor, probably Francesco dAsola,
could have had at hand a manuscript in good condition, unknown today,
but even leaving aside our passage, it would be suspicious that no traces
remain of such an important manuscript. I do not think it likely that the
editor could have consulted manuscripts more ancient than E, F, and D,
from which all the others derive. If my hypothesis is correct, and on the
basis of the documents I have analyzed, I will risk saying that the authorship of certain terms absent from the original belongs purely and simply to
Francesco dAsola. We know today that dAsola did not always follow Aldos mottonon enim recipio emendaturum libros,478 as we know that
he corrected the texts he edited to a considerable extent. Although he did
not know of the existence of Asola, H. Diels rightly wrote that in the case
of Simplicius, Aldini exempli editor haud pauca novavit, infeliciter plurima.479 A clear example of this emmendatio infelix is his conjecture to
complete line 6.3 of Parmenides, proposed to make a mutilated text comprehensible, but whose secondary consequences distorted the philosophers
thought for centuries.
After 1526 the version erge noema was accepted unanimously, and
only minor corrections were proposed to adapt it to the requirements of
the meter. That is how the unpublished version of the Poem made by J. J.
Scaliger suppresses gar taphhodou (for of the way) and proposes from
this rst investigation withdraw thought.480 For his part, G. G. Fulleborn
eliminates tautes and reads for from the rst way of investigation withdraw thought,481 a text also adopted by Brandis, since, according to him,
the word tautes was added by Simplicius contrary to the meter (a curious argument, since Brandis places more trust in Fulleborn than in Simplicius).482 S. Karsten proposes his own version: in the rst place (proton) with477

478
479
480
481
482

In the works cited in note 475, I propose this list: Marc. Gr. 219 (G 1), Marc. Gr. 227 (F),
Marc. Gr. 229 (E), Marc. Gr. Cl. IV.15 (G IV), Paris. Gr. 1908 (P), Laur. 85.1 (B), Laur. 85.2
(D), and Mut. 184 (III.F.6).
Manuzio, A., Theocritus (Venice: In Aedibus Aldi, 1496).
Diels, H., Preface, in Simplicii In Aristotelis Physicorum [ . . . ] Commentaria (Berlin: G.
Reimer, 1895).
Cf. Cordero, N. L., La version de Joseph Scaliger du Poe`me de Parmenide, Hermes 110
(1982) 11722.
Fulleborn (1795), 59.
Brandis (1813), 104.

Presentation of the Thesis and Its Negation in Fragments 6 and 7

115

draw thought from this way of investigation,483 basing it on the fact that
autar epeita (but then) in 6.4 cannot be coordinated with protes in 6.3.
F. G. A. Mullach restores tautes and takes up T. Bergks version (from
this rst way [protaphhodou tautes] of investigation withdraw thought484),
and H. Stein simply changed the order of certain terms.485 This is the state
of the question until 1882, the year in which H. Diels published the second
edition (three and a half centuries after the rst) of Simplicius Commentary. Between 1526 and 1882, the expression withdraw thought (erge
noema) was always accepted, and line 6.3 of Parmenides was read in this
way: from this rst way of investigation withdraw thought. The existing
contradiction between the way of truth stated in 6.12 and this rejection in
6.3 does not seem to have bothered anyone.
As we know, the Simplicius edition published by Diels forms part of
the series Commentaria in Aristotelem Graeca, and on page 117 of volume IX
the quotation from line 6.3 of Parmenides reads thus: for from this rst
way of investigation I withdraw you. There are three changes from the
version in the Aldine edition. For one, ergo, I withdraw, is in the rst
person,486 whereas in the Aldine edition it was in the second person, withdraw. The complement thought (noema) is absent: it is the disciple himself and not just his thought who must withdraw from this way. And the
third change, as the Goddess addresses the disciple, is the appearance of
the pronoun you (s), which gured in most of the manuscripts but which
had been left out of the Aldine edition: I withdraw you. This version of
6.3 has been accepted almost unanimously from 1882 until a few years ago.
I say almost because, before my rejection of it, I only found one author
who did not accept this verb withdraw, and after I published my view
for the rst time, two or three researchers agreed with my rejection. The
only other author who rejected Dielss conjecture is Vitali. In a work published in 1977, this author presented a very special version of the text, since
in order to complete the meter in line 6.3 he added a term from 6.4 and
then introduced a conjecture into this line 6.4. Consequently, 6.3 acquired
this form: . . . which I invite you to consider obviously in this rst way of

483
484

485
486

Karsten (1835), 77. Riaux accepted this version (Riaux [1840], 211).
Mullach, F. G. A., Fragmenta philosophorum graecorum (Paris: Didot, 1860), 131; Bergk, T.,
Commentatio de Empedoclis Proemio, Kleine philologische Schriften II (Halle: Buchhandlung des Waisenhauses, 1886), 35.
Stein, H., Die Fragmente des Parmenides per phuseos, Symbola philologorum bonnensium
in honorem F. Ristchelli (Leipzig: Teubner, 186467), 804.
According to Patin, Dielss conjecture is preferable as it keeps the verb in the rst person
(which allows for the reintroduction of s) and in the indicative (as in 6.2, anoga) (Patin,
A., Parmenides im Kampfe gegen Heraklit, Jahrbuch fur klassische Philologie, Supplementband 25 [1899], 516, note 1).

116

(g) Rejection of the Conjecture I Withdraw You

investigation, but which I reject in that one . . . (6.4).487 Finally, although

my rejection of ergo was rst published in 1979 in an article in Phronesis,488
after thatbut clearly independentlyA. Garca Calvo also contested the
verb ergo and proposed this version of 6.3: for from this way of inquiry I
rst drove you back (osa).489 That is, this scholar does not accept the verb
proposed by Diels, but keeps the idea of withdrawal now represented by
the verb otheo.

(g) Rejection of the Conjecture I Withdraw You

Let us return to Dielss text. Unlike the editor of the Aldine edition, Diels
gives a reason for proposing the conjecture I withdraw you. Indeed, in a
footnote to his edition he says that this reading is based on a parallel expression found in a line belonging to the same context, 7.2: but you, withdraw thought [erge noema] from this way of investigation.490 Let us examine this argument from Diels. It is highly probable that the editor of the
Aldine edition followed the same reasoning and that therefore he directly
proposed the formula erge noema found in 7.1 for 6.3 as well. Diels keeps
the meaning of withdrawal expressed by the verb ergo, but adapts the text
to the meter and suppresses the complement noema, putting the verb in the
rst person, which leads him to add the pronoun you: [I, sc., the Goddess] withdraw you. The reasoning appears to be impeccable, since there
can be no doubt that, as Diels says, 7.1 (withdraw thought from this way
of investigation) appears to be a parallel passage to 6.3 (for from this rst
way of investigation I [ . . . ] you).
If it exists, this parallelism is deceptive: there are parallel passages in
other parts of the Poem but, if words are missing in one of them these
cannot be explained by terms provided by the other.491 In the case concern487

488
489
490
491

Vitali (1977), 35. It is true that I published my criticism six years before the publication
of Vitalis book in 1977 in my doctoral thesis Letre et le non-etre chez Parmenide (Paris IV,
Sorbonne), but clearly Vitali did not know of this work. With respect to Vitali, let us say
that it is not surprising that this writer rejects Dielss conjecture, since his book is characterized by acute hypercriticism. Although this scholar confesses to rejecting corrections,
he departs from the orthodox text of the Poem on seventy-two occasions, and some of
his corrections are rather grotesque. For example, in 1.2, instead of pempon (they
brought me), Vitali adopts pempton, which is found in codex N, and translates at the
fth hour. And in 1.14, instead of accepting Scaligers correction, dike, he keeps diken
with an adverbial accusative meaning and translates for the use of these [sc., gates].
Cordero, Les deux chemins, 132.
Garca Calvo, A., Lecturas Presocraticas (Madrid: Lucina, 1981), 192.
Cf. also Diels (1897), 68.
The typical example is lines 2.3 and 2. 5. They are very similar (almost identical) but they
maintain opposite theses. There is also an obvious parallelism between 1.1 (the mares

Presentation of the Thesis and Its Negation in Fragments 6 and 7

117

ing us, by means of tautes in 6.3 and tesd in 7.2, each line alludes to a way
mentioned in the previous line, that is, in 6.2 and 7.1, respectively. For the
parallelism to be complete, both lines must refer to the same way, from which
the Goddess says, in one case, that it is necessary to withdraw, and in the
other, that it is necessary to withdraw thought. However, we have already
shown that the way presented in 6.12 is the way of truth (hence the difculty of admitting that we must withdraw from it). But which way does
the this (tesd) refer to in 7.2?

(h) The Thesis Expounded in Fragment 7

The Goddesss exhortation to withdraw thought, which we nd in line
7.2, refers to a way (this) proclaimed immediately before, in line 7.1. Let
us analyze this text. This line, rst cited by Plato in the Sophist (in 237a and
258d) and then by Aristotle (Met. N.1089a4), by Pseudo-Alexander (Met.
N.2 [805]), and by Simplicius (Phys. 135, 143, 244), consists of two parts.
The rst hemistich, which presents serious textual problems492 (to the point
where, as we saw, there are authors who consider it to be apocryphal: cf.
supra), states for this [touto] shall never prevail [damei]. The second hemistich makes the meaning of this explicit: that there should be [enai; estin
in direct speech] things that are not [me` eonta].493 However, nobody ever
stated that this formula might belong to the rst way of investigation. All authors
admit that this is a presentation of the negative way (the false way, for
those who think that Parmenides presents only two ways, or the second
false way, for those who hold that there are three ways). If this is so, the
parallelism between 6.3 and 7.2 collapses from the start, because even
though it appears to exist, it is deceptive: each exhortation refers to a different way (6.3 would command withdrawing from the rst way; in contrast,
7.2 orders the withdrawal of thought from a wrong way, that is, from the
thesis there are things that are not, that which is not being, exists). It
is worth pointing out that it is not allowable to complete a mutilated text

492

493

that carry me as far as my spirit reaches) and 1.25 (the mares that carry you to my
home), but if a word was lost from one of the passages, the other passage could not ll
the gap.
The text generally accepted today is that proposed by several manuscripts of Aristotle (E
and J) and Simplicius (E on pages 135 and 244, and D and E on page 134): touto damei.
The verb is the epic form of the passive subjunctive.
The variant onti, proposed by the Aldine editions and taken up by Estienne, has no
manuscript authenticity (Estienne, H., Poiesis philosophica [Geneva: 1573], 42). A more
academic translation would be that non-beings exist, but the intelligent reader knows
that when a Greek philosopher asks questions about beings (ta` onta), he wants to know
what things are.

118

(h) The Thesis Expounded in Fragment 7

(6.3) with a term referring to a context that is not only different but contrary
(7.2). It is clear that Diels made a wrong inference: ergo, which makes sense
when referring to a negative way (the one proclaimed in 7.1), is not appropriate for a positive way (that stated in 6.2).
Let us analyze this line 7.1 in detail. The expression esti me` eonta
contradicts the formula meden douk estin (nothing[ness] does not exist)
in 6.2a494 and is therefore a new way of formulating the second way of
investigation, that is, the negation of the thesis.495 This has been maintained,
among others, by K. Reinhardt,496 M. Untersteiner,497 L. Taran,498 A. H. Coxon,499 and G. Giannantoni.500 Other scholars have seen in 7.1 the statement
of a third501 or even a fourth502 way, but in both cases, once more, these are
wrong ways, which are to be avoided. If we turn to Plato, there can be
no doubt that the quotation from Parmenides in the Sophist refers to the
impossibility of the existence of that which does not exist. Indeed, the rst
time that Plato quotes 7.12 he does so as an illustration of a logos that
some have dared to suppose that that which is not exists [to` me` on enai],
and when he is convinced he has refuted Parmenides, in 258d he says that,
really, the opposite of Parmenides statement is what must be said, that is,
that ta` me` onta, hos estin,503 since now that which is not has the right to
be. Aristotles case is similar: when he transcribes 7.1 he says that there
Parmenides shows that that which is not [to` me` on], is [estin] (Met.
N.1089a). And, nally, for Simplicius there is in that passage an allusion to
a way seeking that which is not [to` me` on zetouses] (Phys. 78.5).
The opposition between the way indicated by the formula enai me` onta
and the true way is even more obvious if we take into account the expres494

495

496
497
498
499

500
501
502
503

We have already shown how, in Parmenides, that which is not and nothing[ness] are
synonymous. The thesis in 7.1 states that nothing[ness] (the things that are not) exists
and therefore is contradictory to 6.2a.
Like the vast majority of interpreters, I consider that me` eonta is neuter plural. So I do
not share Reichs exotic hypothesis, according to which the formula is an accusative singular alluding to who no longer exists, that is, the dead. This viewpoint leads Reich to
see in Parmenides an allusion to the Pythagoreans metempsychosis (Reich, K., Parmenides und die Pithagoreer, Hermes 82 [1954] 289).
Reinhardt (1916), 36.
Untersteiner (1958), cxxx.
Taran (1965), 76.
Coxon (1986), 191. This scholar adds that Plato, Aristotle, and Simplicius had already
understood that 7.12 rejected the negative way of 2.5, and that this opinion is conclusive.
Giannantoni, G., Le due vie di Parmenide, La parola del passato 43 (1988) 216.
This is the case with Gomez-Lobo (1985), 101, and Wiesner (1996), 99.
This is the case with Meijer (1997), 147.
In the rst case, Plato uses the singular, me` on; in the second, the plural, me` onta; and,
clearly, in both cases he is speaking of the same thing. This shows that the unusual
plural in 7.1 is completely irrelevant. It left even Plato indifferent.

Presentation of the Thesis and Its Negation in Fragments 6 and 7

119

sion ou mepote . . . damei (shall never prevail), which as soon as an order

or wish is assumed, is parallel to is not possible . . . in 2.3b. What is not
possible, what shall never prevail, is nothing[ness], that which is not, considered already impossible in 2.7. Line 7.1 as a whole (hemistiches 1 and
2) is synonymous with 2.3b: it is an expression of the principle of noncontradiction, as S. Ranulf504 and J. Moravcsik505 have observed. The complement clause (hemistich 2) reintroduces the way of error. Indeed, in enai me`
eonta we have the assertion of a negative term, which, according to our
analysis is equivalent to the assertion of a concept with respect to its contradiction, that is, the negation of the thesis. This contradiction must be
avoided. Ruggiu stated that the Goddess wants to prevent things being
both eonta and me` eonta at once.506 This internal contradiction (all in all, the
non-respect of the principle of non-contradiction) is the principal characteristic of the negation of the thesis, the second way.
The Goddess orders the withdrawal of thought from this way of investigation (7.2). That is logical. On the other hand, there can be no obligation
to withdraw from a way that is the opposite of this way. As this is the case
with 6.12, we cannot admit the only reason adduced by Diels to propose
his conjecture (viz. the parallelism between both ways). The ways referred to in 7.2 and 6.3 are not parallel: they are opposite. Therefore I shall
try to solve the problem of the gap in line 6.3 in a different way.

(i) A Possible Solution for the Gap in Line 6.3

After having stated his thesis once more (and only his thesis: it is possible
to be, whereas nothing[ness] does not exist) in the rst two lines of fragment 6, from line 4 onward in the same fragment (as I shall try to show),
Parmenides expounds the negation of the thesis, according to the same
procedure already used in fragment 2 (thesis: line 3; negation of the thesis,
line 5). However, between the presentation of both possibilities, in line 6.4,
the Goddess expresses something concerning both the thesis she has just
expounded in 6.12 (this rst way of investigation . . .) and the one she
will expound from 6.4 onward (and then . . .). As I have already shown,
the phrases verb is missing: we do not know what the Goddess says about
the two possibilities. Be that as it may, we must not forget that in the original text there was a verb (although this has been lost in successive transcrip504
505
506

Ranulf, S., Der eleatische Satz vom Widerspruch (Copenhagen: Gyldendal, 1924), 160.
Moravcsik, J. M. E., Being and Meaning in the Sophist, Acta philosophica fennica 14 (1962)
25.
Ruggiu (1975), 147.

120

(i) A Possible Solution for the Gap in Line 6.3

tions of the Poem), and it is essential that this verb should have been valid
for the two formulas, both for that in 6.3 and that in 6.4, which express contrary
contents. The expressions protes gar . . . and auta`r epeita . . . (rst . . . and
then . . .) depend on this unknown verb. Therefore the missing verb must
have had a meaning capable of being applied to two contradictory ways,
since, as I have shown, 6.3 alludes to the thesis, that is, the rst way, and
6.4 ff. to its negation, that is, the second way. This fact presents no difculty
at all. In various parallel passages Parmenides uses the same verb to refer to
two contradictory ways. In 1.2830, truth and opinions (that is, the two
ways) must be the object of inquiry (puthesthai), and therefore they are
lumped together in the notion of panta, everything. In 2.25 the expression to think (2.2) is valid for both the ways that will then be stated.
Finally, in 8.1718, the verb ean (allow) is valid both for the way of error
(which has been allowed as unthinkable) and for the way of truth (which
has been allowed as genuine).507 This means that, regardless of each ways
merit, the same verb (or similar verbs) can be applied to both ways: both
must be the object of inquiry (1.2830), both constitute the muthos that must
be listened to (2.2), both are possibilities offered to thought (2.3, 2.5) and
both merit precise assessment (8.1718).
However, there is an essential nuance that must not be forgotten: both
ways are stated or proclaimed by the Goddess and listened to or accepted
by the traveling disciple. And once she has stated both ways, the Goddess
explains their content to the disciple: truth corresponds to being, and opinion to human perception and the wandering intellect. The description of
the way of truth will be the object of most of fragment 8, up to line 51.
After that there will be an explanation of doxai, an explanation that will
take place even though there can be no pstis alethes (real trust, 1.30) in
them. This observation leads us to take into account one of the meanings
of the preposition apo, which goes with the missing verb in the two lines
and does not necessarily mean a rejection, as in Dielss conjecture (from
this . . . I withdraw you). Here, in fragment 6, we have the use of apo with
the meaning of by, with, and in particular, from.508 My hypothesis is
as follows: in lines 6.3 ff. the Goddess alludes, as in 1.2830, 2.35 and
8.5052, to the starting point of the mental journey the disciple must undertake, the journey that covers the content of the Goddesss course. This
journey, like any other, has an end, clearly mentioned by the formula en
507

508

In two other parallel passages there are different verbs, but their meaning is similar.
1.2627 speaks of traveling the road of truth that lies near the Goddess, whereas the road
of men lies far away. And in 8.50 the Goddess concludes (pauo) her explanation of one
way and invites the disciple to listen to her account of the second way; so we may say
that both ways are heard about from the Goddess by the disciple.
Cf. L. S. J., s.v. apo, I.

Presentation of the Thesis and Its Negation in Fragments 6 and 7

121

toi soi pauo (here I end for you) (8.5051). But there must also be a starting point for the teaching, and it occurs to me that the most appropriate
verb to indicate this start is the verb whose meaning is the opposite of
end: begin, arkhesthai, in the middle voice.
Parmenides uses this verb on two occasions: in fragment 5, arxomai (I
shall begin), and in 8.10, arxamenon (beginning). We may suppose that
he also used it in 6.3, but in this case it must not be forgotten that the
preposition apo, which appears on two occasions (in 6.3 and in 6.4) stands
in relation to this verb, just as it stood in relation to ergo (withdraw) in
Dielss conjecture. So can apo stand in relation to arkhesthai (begin)? There
can be no doubt that it can. Furthermore, the combination arkhomai (only
in the middle voice)509 + apo is a real cliche in ancient Greek literature, and
in line 2 of fragment 5 Parmenides himself uses arxomai accompanied by a
synonym of the preposition apo: hoppothen (where, whence). This
combination alludes to the starting point of something, be it a list, a series,
an account, a mental journey, even a philosophy course.510 There are no
examples of this construction511 in Homer. On a single occasion he uses a
parallel form in which ek replaces apo: ek de tou arkhomenos (beginning by . . .) (Od. 23.199). However, Herodotus gives us three representative examples of the construction: he gave an account consistent with the
truth, beginning from the beginning (arkhomenos . . . apo` arkhes) (I.116.5);
the Egyptians shave their heads, beginning from childhood (apo` paidion
arxamenoi) (III.12.10); and he expounded Cyrus paternal genealogy, beginning from Achaemenes (arxamenos de ap . . . ) (III.75.2). In Plato we nd a
number of relevant examples: Gorgias 471c, beginning with you (arxamenos apo sou); Phaedrus 228d, beginning with the rst [of the gods] (arxamenos apo` tou protou); Phaedo 100b, I begin with those (arkhomai apekenon); Parmenides 137b, I shall begin with you yourself (aphemautou
arxomai) (137b); Sophist 218b, beginning, in the rst place, with the Sophist (arkhomenoi . . . apo` tou sophistou);512 Sophist 242d, beginning from Xenophanes (apo` . . . arxamenon); Timaeus 28b, beginning from some beginning (aparkhes tinos arxamenos); Laws 771a, beginning from the sacred
(ap hieron ergmene); and Laws 771c, up to twelve, beginning from one (apo`

509
510

511
512

This fact invalidates A. Nehamas conjecture (cf. infra, Chapter VII, note 619), who, after
my work published in Phronesis (1979), adopted the same verb, but in the active voice.
Later, in grammar, arkhesthai + apo was used to mean the letter beginning a word. Cf.
Dionysius of Thrace: It begins with (apo` . . . arkhomenen) a vowel, like ergon; it begins
with (id.) a consonant, like Nestor (Ars. gramm. 6.33.26).
The only case to be found in Homer, kaprou apo` trkhas arxamenos (Il. 19.154), is a clear
case of tmesis: aperxamenos.
Cf. the parallel passage, what is the beginning (arkhe) from which it would begin (arxaito)
. . . ? (242b).

122

(j) Discovering the Foundation of the Two Ways in Fragment 6

mias arxamenos).513 Aristotle also uses this formula (Met. A.2.983a123: they
begin by being surprised, arkhontai apo` tou thaumazein; Z.2.1028b23, beginning from the one, apo` tou henos arxamenos), and so did Protagoras,514
Xenophon,515 Demosthenes516 and Simplicius.517
A decisive example can be found in Critias, since, as in Parmenides, he
is speaking about the starting point of a teaching: I begin from the origin
of man (arkhomai apo` tes genetes anthropou) (fr. 32). This would be the sense
of arkhesthai + apo in Parmenides, if he did indeed use this verb in fragment
6, that, as we saw, is to be found in two other passages of the Poem.
The principal consequence of my conjecture is the following: as it does
not assume a criticism (either of one or of two ways, as was the case with
eirgo), but a new presentation of the two possibilities given in fragment 2,
there is no need to imagine a third way, which would be the second way
to be criticized (since, obviously, it is not possible to criticize the rst).
Regarding this hypothesis, F. Fronterotta wrote that the meaning of the
message [of fr. 6] changes completely if different conjectures from that of
Diels are accepted: it is more reasonable to suppose that the way being
spoken of in lines 13 of fr. 6 coincides with the rst way in fr. 2.518 This
is what I have demonstrated.

(j) Discovering the Foundation of the Two Ways

in Fragment 6
I said that in the passage 6.12, Parmenides returned to formulating the
rst way of investigation, and that from 6.4 onward a description of the
second way, the way of error, clearly appeared. When the Goddess presented her program of studies, she had already told the future disciple that
he had to be abreast (puthesthai) of everything, both truth and mistaken
513

514
515
516
517
518

Cf. also the formula apo` Hestas arkhesthai (Euthyphro, 3a) or arkhometha (Cratylus,
401b), which is also found in Aristophanes (Frogs, 845) and in other fth-century authors,
although its origin is very ancient. The usual meaning is begin with the essential, since
Hestia represents the very center of the polis, the home (cf. Dorion, L. A., Euthyphro
[Paris: GF-Flammarion, 1997], 291, note 16). In Sophron and Crates we nd the equivalent
expression ex Hestas arkhesthai.
Protagoras, fragment 3, you have to begin learning from infancy (apo` neotetos, arxamenous).
Xenophon, Memories, 3.5.12: begin with the parents (apo` ton pateron arkhontai).
Demosthenes, 18.297 (= 325, 7), beginning with you (arxamenos apo sou).
Simplicius, Phys. 1014.26: [Zenos argument known as The Stadium] begins with
Achilles from the beginning of the stadium . . . (arxamenou . . . apo` tes arkhes tou stadou).
Fronterotta, F., Essere, tempo e pensiero: Parmenide et lorigine dellontologia, Annali
della Scuola Normale Superiore di Pisa, Classe di Lettere e Filosoa, Serie II, Vol. 24 (1994)
841.

Presentation of the Thesis and Its Negation in Fragments 6 and 7

123

opinions. Logically, learning must begin with the knowledge of a thesis, and
then begin with the other. Why must it begin twice? Because Parmenides
reasoning is circular: it is common [xunon] for me that where [hoppothen] I
begin [arxomai], there I shall return again (fr. 5). Years later, Plato followed
Parmenides example, since the method proposed in his Parmenides as intellectual gymnastics for the young Socrates consists in beginning (arxometha) with the examination of a certain hypothesis (137a), and once this is
exhausted, beginning with the opposite hypothesis (elthontes palin ep` ten
arkhen . . . ) (165e). As the starting point is arbitrary, the Parmenides of
Platos Parmenides proposes beginning with (apemautou arxomai) his own hypothesis (137b).
But the situation is more radical in Parmenides, since according to the
content of the thesis, it is enough to state it, to realize that its negation is
impossible; thus, even if you begin with the negation, you have to recognize that the positive statement is necessary. Therefore, the starting point
of the reasoning is common: the conclusion of one way is the starting point
of the other; you begin with being and come back to being.519 In fragment
5 it is the Goddess expounding her method, and therefore she says that
for her (moi), the starting point of the double argument is common
(xunos). This term has the same meaning in Parmenides as it does in Heraclitus (fr. 2 and 103) and in the ancient Etymologica:520 koinos, common,
that is, coincident.521 This does not mean an indifferent viewpoint, as
Taran522 maintains, but one that is shared (gemeinsamer523). It is from
this common point that you have to begin: the Goddess begins her explanation and the disciple begins rst to receive her teaching and then to test
it. He begins by one way and then begins again by another.524 So when the
Goddess ends her exposition of the rst way, she indicates: here I end for
519

520

521
522
523
524

This is the meaning of fragment 5, according to Meijer (Meijer, P. A., Das methodologische im 5. Fragment des Parmenides, Classica at medievalia 30 [1969] 1045). Cf. also
Meijer (1997), 2324.
Cf. Hesychius, Lexicon, Vol. III (Amsterdam: A. M. Hakkert, 1965), 173; Etymologicum
magnum, ed. Gaisford, T. (Amsterdam: A. M. Hakkert, 1967), 416; Etymologicum graecae
linguae (Leipzig: J. A. G. Weigel, 1818), 416.
Kirk, G. S., Heraclitus: The Cosmic Fragments (Cambridge, England: Cambridge University
Press, 1954), 115.
Taran (1965), 51.
Meijer, Das methodologische, 104.
It should be noted that the meaning of the correlation protes gar . . . autar epeita (6.34)
is not adversative, as it appears in most translations, but simply correlative. In Homer
this is a habitual formula corresponding to on the one hand . . . and then (cf. Il. 3.315,
11.420, 12.191, 23.237, 23.683, 24.791). The presence in all these cases of proton instead of
protes led Karsten to modify Parmenides text (cf. supra). Denniston says that the commonest use of autar is weakly adversative, or purely progressive (Denniston, J. D., The
Greek Particles [Oxford: Clarendon Press, 1934], 55) and that the formula autar epe often
indicates the progressive stages of an account. Cf. atar in 8.58.

124

(j) Discovering the Foundation of the Two Ways in Fragment 6

you (soi) reasoning and thinking about the truth (8.5051) and she begins
then (dapo` toude . . . ) (8.52) the presentation of the opinions of mortals,
about which she can only offer a deceitful series of words (8.52).
So the verb arkesthai offers a content that could be applicable to the gap
in line 6.3. In the rst place, since it does not assume a rejection of what
has just been expounded, it gives the connective gar (which joins 6.2 and
6.3) its normal value, since if Parmenides had wanted to distance himself
from what was said in 6.12, he would not have then said since or given
that. As Mourelatos observed about Tarans interpretation, Parmenides
would surely have written alla`but, howeverbut would never
have used a particle that implies continuity.525 Then, if what we have is not
a rejection, the Goddesss exhortation to her disciple continues normally in
6.4 (and that is why I believe that the verb might have been in the second
person). And as the teaching in question is situated in the future (the rst
thesis in 8.150; the second from 8.51 onward), the verb, I believe, must be
conjectured as being in the future tense. So my conjecture is you will begin,
arxei in the middle voice (as is the case with all the examples we have
already seen of arkhomai + apo; there are no examples of this formula in the
active voice). This conjecture leads us to leave out the pronoun s. Since
now the verb is in the second person, s would have been an apocopate
[shortened form] of su [you], you will begin, but the elision of u is
highly unlikely. However, if we take into account the manuscript tradition
of line 6.3, we nd two codices offering the reading t instead of s:
manuscripts B and C. Another codex, G IV, gives us no term between gar
and aph (this was the model followed by the 1526 Aldine edition), which
might mean that the copyist hesitated between s and t and preferred
to suppress the pronoun. I should say that I accept the pronoun te (t)526
and I complete the hexameter with the verb arxei: protes gar taphhodou dizesios
arxei. So my version of the beginning of fragment 6 is as follows: It is
necessary to say and to think that by being, it is, since it is possible to be,
and nothing[ness] does not exist. This I order to proclaim since you <will
begin> with this rst way of investigation, but then with that made by
mortals who know nothing . . . (6.14). We already know what is the foundation of the rst way, that is, the thesis: that which is being is, and nothing[ness] does not exist. In the next chapter we shall see what is the foundation of the negation of the thesis.

525
526

Mourelatos (1970), 77, note 7.

With regard to the construction gar te, cf. Denniston, Greek Particles, 529. Cf. also Il. 23.156,
Od. 7.307, 12.105.

Chapter VII: The Negation of the Thesis,

Opinions, and the
Nonexistent Third Way

After expounding his thesis once again in the rst two lines of fragment 6,
Parmenides presents the negative aspect of it (as he did before in fragment
2) still in fragment 6, from line 3 onward. But this time the negation of the
thesis is accompanied by its foundation, that is, its cause. This wrong
way is not autonomous, proved on its own evidence, as the rst one is
(since, indeed, who can deny that there is being, that that which is being
is?). This time we have an articial way, invented by those who ignore the
unbearable weight of the fact of being and therefore relativize it. Nevertheless, faithful to his program of study (in which the Goddess invited the
future philosopher also to inquire about human opinions), Parmenides
proposes studying this false way, in order to discover what its foundation
is. Once the origin of the error has been grasped, only one way will remain
as a real possibility, which will be discussed in the lengthy fragment 8. A
hypothetical third way has no foundation whatsoever.
If my interpretation is correct, the passage that begins in line 6.4 represents the negation of the thesis that, as we saw, is expounded in the second
way. A strong witness comes to my aid: Simplicius. Indeed, when this author cites lines 6.1b9, he does so to give an example of the position of
those who admit nothing besides (or apart from, para) being. W. Leszl,
who has studied this Simplicius passage in minute detail, states the following: in the rst place, Parmenides announces the fundamental alternative
[or separation], which constitutes the rst way: being is, but nothing[ness]
is not (6.1b2a). This is the alternative that mortals ignore when they set
being and not-being side by side.527 According to this same author, lines 4
ff. of fragment 6 illustrate this ignorance, and if Simplicius quotes these
lines it is to conrm in Parmenides the presence of two ways, and certainly
not three.528 F. Fronterotta arrives at the same conclusion, saying in Simplicius eyes, fragments 2 and 6 are not in contradiction, since Parmenides

527
528

Leszl, W., Parmenide e lEleatismo. Dispensa per il corso di Storia della Filosoa Antica,
Universita` degli Studi di Pisa, Dipartimento di Filosoa (May 1994) 13738.
Leszl, Parmenide,123.

126

No Distinction Between Being and Not-being

indicates only two ways of investigation in each of them,529 so that the

possibilities in Simplicius presentation are therefore certainly two and not
three, since Reinhardts hypothesis on the existence of a third way is
based exclusively on the opposition between fragments 2 and 6 and on
Dielss conjecture for the gap in 6.3.530
So now let us see how Parmenides describes this second way.531 The
rst words of line 4 of fragment 6 clearly show that the way that begins to
be expounded here is different from the previous one, presented in 6.12:
auta`r epeita, but then . . .. Nevertheless, in order to understand the true
structure of this way, we must clarify certain things about the original text,
which, once again, was subjected to certain abuses by interpreters. I say
this because in line 6.5 there is a verb plattontai (make, create) whose
subject is mortals. This enables us to state that, for Parmenides, the way
is the product of human activity. This verb plattontai occurs in all the manuscripts of Simplicius, the only source of fragment 6. Failing to take into
account this unanimity, the editor of the 1526 Aldine edition replaced this
verb with plazontai (stray, miss, wander). Nothing justies this change.
Nevertheless, this conjecture was adopted and even created a rule,
since in the rigorous Liddell, Scott, and Jones Lexicon we read that when
Parmenides writes plattontai he is merely using an unprecedented form of
the middle voice of plazo. H. Diels, who supports this change, offers similar
examples in the dialect of Tarento.532 This argument is more than weak. D.
Sider states ironically that regarding Parmenides as an Italian poet is like
believing that Joyce was also an Italian novelist.533 A. Capizzi demonstrated
that perhaps in the dialect of Tarento one might be able to nd plazo instead
of plasso (which, according to Diels, would have been the verb used by
Parmenides), but what needs demonstrating is the reverse: why does Parmenides use plasso instead of plazo,534 if he really wanted to refer to straying? Those who support the reading plazontai recognize that Parmenides
could have written plassontai in his usual Ionian, and that later a Byzantine copyist, used to the Attic of Simplicius, transformed it into plattontai.535
This hypothesis could be accepted, but it is difcult to take the nal step,

529
530
531

532
533
534
535

Fronterotta, F., Essere, tempo e pensiero: Parmenide et lorigine dellontologia, Annali

della Scuola Normale Superiore di Pisa, Classe di Lettere e Filosoa, Serie II, 24 (1994) 845.
Fronterotta, Essere, tempo, 389 and 845, note 27.
I repeat once more that as in fragment 2 there were only two ways to think, and as at the
beginning of fragment 6 Parmenides expounds the rst of these ways, the way that will
be described after that has to be the second way.
Diels (1897), 73.
Sider, D., Textual Notes on Parmenides Poem, Hermes 113 (1985) 364.
Capizzi, A., La porta di Parmenide (Rome: Edizioni dellAteneo, 1975), 75.
Cf. Diels (1897), 72.

The Negation of the Thesis, Opinions, and the Nonexistent Third Way

127

which is the assimilation of plasso/platto and plazo.536 When Parmenides

uses the notion of plattontai, he uses it with its normal, banal meaning of
make, create,537 and in the slightly subjective sense of make for yourself, imagine.538 If we take into account that the creation of this false way
is the work of mortals, that it consists of a series of words established by
them, we nd a very suggestive echo in a passage in the Laws, where Plato
says that the legislator must begin by creating with speech (plasasthai toi
logoi, 800b7). This interpretation of the verb enables us, nally, to nd a
justication for the relative hen (6.4), which is generally translated in an
arbitrary way as in (or by) which, as if it were a locative.539 The Goddess
simply says that this is a way that (he`n) mortals make, and that they believe is right.540
This impossible way is referred to by the pronoun tes (that) (6.4),
but the text does not give a complete formulation of it. This proves that
this way is not new: it has already been presented, in particular, in line 5 of
fragment 2. It is the wrong way, the negation of the thesis. In fragment 6
the Goddess expands on it, explains who made it and who its hypothetical
users are (hypothetical, since the way is untravelable). The reason is this:
this way was made by mortal human beings who know nothing, or, if you
prefer, who know nothing[ness]: broto` eidotes ouden (6.4). This formula,
which is a real stereotype in ancient lyric poetry541 is the polar opposite,
in Parmenides, to the formula eidota fota, the man who knows, which
characterizes the Goddesss disciple in fragment 1.3. If this is so, whereas
mortals who know nothing are condemned to make themselves a way that
leads nowhere, the philosopher, who is a mortal who knows, sets out
along the way of the Goddess, full of signs (hodo`s poluphemos damonos,
1.23), a way that lies far distant from the path of men (apanthropon
ekto`s patou estin, 1.27).542
536
537
538
539
540

541
542

Ballew notes that no important philologist has related plazo to platto/plasso (Ballew, L.,
Straight and Circular in Parmenides and the Timaeus, Phronesis 19 [1974] 193).
Cf. Hesiod, Works 70; Lysias, 19.60.
Ballew, Straight and Circular, 193.
Cf. Conche: upon which mortals stray (Conche [1996], 100); and Wiesner, auf welchem
. . . Sterbliche irren (Wiesner [1996], 252).
For the mortals of Parmenides, the world they believe is real has the same value as the
account given by Critias of Atlantis: it is not a myth created [plasthenta, participle of
platto], but a true report (Timaeus, 26e).
Mansfeld (1964), 4. This scholar picks up the subtle analysis of Pfeiffer, R., Gottheit und
Individuum, in Ausgewahlte Schriften (Munich: Beck, 1960).
Parmenides is thinking of ordinary men who have not decided to set out on a way in
search of truth. It remains to be said that in the Poem, the word broto (mortals) is
synonymos with anthropoi (men). Cf. 8.3839: onoma . . . broto` katethento (names . . .
that mortals assigned) and 19.3, onomanthropoi katethento (names that men assigned). Contra, cf. Coxon, A. H., The Philosophy of Parmenides, Classical Quarterly 28

128

No Distinction Between Being and Not-being

Here we nd a new conrmation of the erroneous character of the way

described in passage 6.47 that continues in 7.15: on this way the noos
(intellect) of mortals is driven, as if traveling in a chariot,543 but the driver
is incapacity (amekhane, 6.5). Deprived of their infallible guidesthe
Daughters of the Sun, the wise horses, and, especially, the Goddessmen
are now carried (phorountai)544 along a way made by an intellect that has
lost its object (which is, as we saw in our analysis of 8.3435, the fact of
being, the single possible thinkable object), and so this intellect lacks the
necessary resources to arrive at its goal. Incapacity545 is the mora kake
(evil fate) which plays the same role on the way of error as Themis and
Dike played on the way of truth (cf. 1.2628). As Vlastos says,546 this amekhane is the natural state of humanity before divinity (Prometheus, in his
example) brought help (skills) to it: men were blind and deaf.547 These same
defects belong to the mortals in fragment 6 (kopho` homos tuphlo te, deaf
and blind), whereas the formula by which Aeschylus sums up this state
of affairs (everything happened to them at random, confusedly, Prom.
vinc. 450) reappears in Parmenides with the term tethepotes (amazed), a
true resume of the situation. Mortals, lacking the Goddesss teaching, are
swept along, deaf and blind, like automata:548 they do not know where to
head for, since, unlike the mortal who knows, as Coxon549 observes, these
people do not know what road they are on, which is not extraordinary,
seeing that this way is panapeuthea (completely unknown) (2.6) or, as P.
Destree proposes, they shufe on drunkenly, with faltering steps, backwards and forwards, forwards and backwards, as the term palntropos550

543
544

545

546
547
548
549
550

(1934) 134; and Bormann (1971), 101, which establish a difference between the broto of
fragment 6 and those of the rest of the Poem. Further on I will offer my interpretation of
the identity of these mortal men.
Drive is ithuno, the verb used for driving chariots and riding horses: cf. Homer Il.
11.528, 16.475, and Hesiod, Shield, 324.
The allusion to the images in fragment 1 is clear: this verb echoes a pherousin (1.1, 1.25),
pherei (1.3), pheromen and pheron (1.3), applied in the passage to the traveler who goes in
search of the Goddess and who is carried.
The term amekhane alludes to the impossibility, because of the lack of the necessary resources, to carry out any kind of task. Cf. Empedocles, fragment 12.1: amekhanos = adunatos
(incapable = impossible). For Mansfeld, amekhane = Machtlosigkeit (Mansfeld [1964], 11
ff.). Snell relates the term with action (dran, do), on the basis of Aeschylus, mekhane
drasterios (way of doing) (Snell, B., Aischylos und das Handeln in Drama, Philologus,
Supp. 20, I (1928) 14).
Vlastos, G., Equality and Justice in Early Greek Cosmologies, Classical Philology 42
(1947), 163.
Cf. Aeschylus, Prom. vinc. 44748: In the beginning, they had eyes, but their eyes did
not show them anything useful; they had ears, but they did not hear.
Cf. similar images in Homer, Il. 4.262, 21.64, and especially Od. 23.15.
Coxon, The Philosophy of Parmenides, 131, note 1.
Destree, P., La communaute de letre (Parmenide, fr. B 5), Revue de philosophie ancienne
18(1) (2000) 12.

The Negation of the Thesis, Opinions, and the Nonexistent Third Way

129

indicates. There can be no doubt that this human way is the negation of
the thesis, that is, the wrong way of fragment 2, and for that reason the
Goddess exhorts the disciple to withdraw thought from it in 7.2. It is true
that men usually move forward blindly, but Parmenides writes his Poem
to show them the way to go, so that long habit, their usual guide,551 does
not force them to use eyes that cannot see what they ought to see (and
therefore are as if blind), ears that do not hear what they ought to hear
(and therefore are as if deaf), and a tongue552 that only utters deceitful
words (7.35; 8.52).553

(a) The Senses and the Wandering Intellect Do Not

Distinguish Between Being and Not-being
This critique of the senses554 gathers the last six lines of fragment 6 and all
of fragment 7 into a homogeneous whole. By means of an argument we
have already come across in fragment 2 (and which M. Untersteiner, with
respect to fr. 1, calls composizione ad anello555), the whole of this passage
begins and ends with a critique not only of sense experience but also of the
wandering intellect. This coupling should not surprise us: the boundary
separating thought and sensation in the Presocratics is vague. Even in Xenophanes, scholars do not agree about the translation of noe in fragment
24, whose subject is divinity; as the verb is accompanied by see and by
hear, to translate it as think, as do the vast majority of interpreters, is
perhaps excessive. In Il. 15.422, Hector saw (enoesen, from the verb noen)
his cousin with his eyes, and in Il. 24.294, Hecuba says to Priam that soon
you will see [noesas] the portent with your eyes. If we take into account
551

552
553

554
555

According to Becker, here Parmenides is contrasting a usual way with a way being traveled for the rst time (Becker, O., Das Bild des Weges und verwandte Vorstellungen im
fruhgriechischen Denken, Hermes Einzelschriften 4 [1937] 142, note 13).
Cf. Bacchylides, 10.51.
For Aubenque, in this passage there is an allusion to the bavardage vide, la glossolalie,
in strong contrast to the true speech of the Goddess (Aubenque, P., Syntaxe et semantique de letre dans le Poe`me de Parmenide, in Etudes sur Parmenide, Vol. II, ed. Aubenque, P. [Paris: Vrin, 1987], 119). Cf. also Mansfeld, J., Parmenide et Heraclite avaientils une theorie de la perception? Phronesis 44 (1999) 331. Holscher, on the basis of Empedocles, fragment 3.11, proposes another interpretation and says that here the tongue
represents the sense of taste (Geschmack) (Holscher, U., Anfangliche Fragen [Gottingen:
Vandenhoeck & Ruprecht, 1968], 52. Contra, cf. Patin, for whom using the tongue here
means onomazein (naming) (Patin [1899], 633). Cf. also Verdenius (1942), 55, note 7;
and Mansfeld (1964), 43.
Vlastos speaks of insensible senses (Vlastos, G., Parmenides Theory of Knowledge,
Transactions and Proceedings of the American Philological Association 77 [1946] 69).
Untersteiner (1958), lxxix.

130

No Distinction Between Being and Not-being

that in an indirect version of Xenophanes text, transmitted by Diogenes

Laertius, the verb noe was replaced by anapne (breathe), perhaps a
translation such as observe would t fragment 24. Aristotle had already
said that for the ancient philosophers, thinking (to` noen) was like sensation, something bodily (De anima 426a26), and J. Mansfeld recently defended the idea according to which the pre-Platonic philosophers did not
make a distinction between sensation and thought,556 since both factors are
the cause, in Parmenides, of the wrong way habitually followed by mortals,
that is, the adoption of a method of investigation that denies the fundamental thesis and therefore leads nowhere. This wrong way is, without any doubt,
the second way expounded in fragment 2, a way considered a priori as a way of
investigation (let us not forget that in line 2.2 the Goddess says that she will
present the only ways of investigation, and then goes on to present just
two) and which subsequently shows itself to be completely impracticable. The
critique in line 7.2 refers to this way.
The formula in line 7.1, enai me` eonta (there are things that are not),
is a summary of the last two lines of fragment 6 (6.89): mortals way of
thinking based on the senses, allowing itself to be guided by a wandering
intellect, leads them to suppose that there are things that are not. This way of
thinking must never prevail (ou mepote touto damei) (7.1a). But if men
think like that it is because they are dkranoi (two-headed) (6.5). The time
has come to consider this fundamental characteristic of the human condition, as Parmenides conceives it. Men have two heads: with one they look
at being and with the other at non-being.557 This conjunction is the key to
the error of mortals. They are incapable of accepting the principle of noncontradiction and the excluded middle, which requires a decision: either
the one or the other. As they do not know what being is, and therefore, the
extent to which it is impossible not to be, they confuse the one with the
other, as if it were possible to say that there are things that are not. Remember that the way of error consisted in a combination of contrary notions.
This combination emerges clearly from lines 8 and 9 of fragment 6, which
describe the content of the wrong way as seen by its user. So we should
not think that these lines present a precise formula, proposed by the Goddess, as was the case in 2.3 and 2.5. The passage shows how men conceive (nenomistai) the only possible objects of all thought: being and non556
557

Mansfeld, J., Aristote et la structure du De sensibus de Theophraste, Phronesis 41 (1996)

158 ff., and Parmenide et Heraclite, 331.
The expression belongs to Loew (Loew, E., Ein Beitrag zum heraklitisch-parmenideischen Enkenntnisproblem, Archiv fur Geschichte der Philosophie 31 [1917] 45). We
should not make the mistake of believing that the expression dkranoi refers to doublethinking. In Parmenides time you did not think with your head but with your breast.
The head is the receptacle for the principal senses: sight, hearing, smell, taste.

The Negation of the Thesis, Opinions, and the Nonexistent Third Way

131

being, existence and nonexistence.558 By mortals, the text says, to` pelein te
ka` ouk enai (being and not-being) are considered tauto`n kou tauto`n (the
same and not the same). By means of the verb nomzo (consider) related
to nomos (custom, habit, among other things), Parmenides stresses the
habitual character of this human opinion that is based on a wandering
intellect (plagkto`s noos).559 It is long habit (ethos polupeiron) (7.3) that leads
men to create (plattontai) (6.5) a method that mixes up being and not-being,
and therefore, to state that they are both the same and different at once.
The expression we nd in 6.89 simply means that mortals attribute
being to non-being and non-being to being. To put it another way, they confuse
(mix up, combine, advocate) that which is and that which is not. The verb
that Simplicius uses to refer to this operation is sunpherein (bring together), and he says that this operation takes place in thought (Phys.
78.2). It is interesting to note that when Plato refers to the predication or
attribution of being to non-being and vice versa, that is, when he quotes
line 7.1 of Parmenides in order to refute it, he also uses a series of verbs
synonymous with sunpherein, which all refer to language: prosggnesthai
(Soph. 238a4, 6), prosarmotten (238c), and prosaptein (239b). Finally, when he
searches for an epithet to describe those who say that to` me` on, enai pos
(that which is not, in a certain way is)namely, the way that for Parmenides is the way of error, but which Plato wishes to defendhe uses the
term polukephalos (many-headed) (240c).
For all these reasons, I do not share the opinions of researchers who
see in 6.89 a simultaneous allusion to the two ways of investigation stated
in fragment 2 (a conjunction that would amount to a third way). This
hypothesis ignores the fact that the formula tauto`n kou tauto`n (the same
and not the same) must be taken as a whole. The point of this phrase is
to stress the confusion between two elements, the failure to differentiate
between them by not knowing whether they are the same or not the same.
A parallel example can be found in a treatise by Hippocrates: panta tauta`
ka ou tauta (it is all the same and not the same); phaos Zen, skotos Aidei
558

559

Like most scholars, I consider that to` pelein (6.8) is synonymous with enai. Contra,
Somigliana considers that to` pelein is a verb of movement meaning to turn, which produces this curious translation of the passage: That which turns and non-being are the
same and not the same (Somigliana, A., I versi 4 ss. del fr. B 6 di Parmenide e la
supposta polemica antieraclitea, Rivista di losoa neoscolastica 67 [1975] 333). However,
the author does not tell us how the second part of the statement (sc., that which turns
and non-being [ . . . ] are not the same) could be false, given that Parmenides expressly
says that to` eon is immobile (8.26), i.e., it does not turn. The use of pelein in 8.11 and in
8.19 as an obvious synonym of enai refutes Somiglianas hypothesis.
Instead of noos, Simplicius codex E proposes nomos, and a reviser of codex P also added
nomos beside noos. This is probably an erroneous reading, but the closeness of nomzo
might be an invitation to reect on the question.

132

No Distinction Between Being and Not-being

(light, for Zeus; darkness, for Hades); phaos Adei, skotos Zen (light,
for Hades; darkness, for Zeus) (De victu I.5). This phrase does not contain
a statement corresponding to tauta (the same) and another to ou tauta
(not the same), one of which is true and the other false; there is a mixture
of four statements, in which the same notions are attributed to two different
gods, from which it can be deduced that those who express themselves thus
do not know what light or darkness is or to which god they correspond.560
From what I have said, the following conclusion can be drawn: lines
6.89 do not present a new way of investigation because, among other reasons, to propose a new way after having proclaimed that there are only
two possibilities would be equivalent to saying that Parmenides is not
keeping to his plan, fundamentally because, given the message the Goddess
transmits to her disciple, everything can be summed up in one essential
alternative. The main fault that this passage seeks to highlight is the admission of a conjunction. Neither of these two requirements are present when
there are three possibilities: either one or the other, or one and the other. To
maintain that both that which is and that which is not exist is equivalent
to maintaining that that which is, both is and is not at once. As I said supra,
Parmenides bitterest enemy, Gorgias, grasped the meaning of the way of
error with great subtlety. Indeed, when the Sophist claims to demonstratein order to go on to refute itthat that which is not, exists, he
states: If that which is not, is [viz., the negation of the thesis, corresponding to the second way, according to my interpretation], then it will be and
at the same time it will not be; since, to the extent that it is thought of as
non-existent, it does not exist; but to the extent that it is something that is
not, it will be (fr. 3, 67).561 As S. Karsten says, those who state that nonbeing is, deny being and, at the same time afrm it.562 Furthermore, when
we examined the content of the second way, we said then that as this consists of the afrmation of a negation and of the negation of an afrmation,
we found in it a mixture of being and not-being in which each notion was
attributed to its contrary. Here in 6.89, we nd an attribution (which according to the parameters of Parmenides thesis must be established in
analogous terms) made with regard to contradictory terms. But, fundamentally, we nd (1) the negation of the principle of non-contradiction and (2) the
postulation of difference.
We must remember that the principle of non-contradiction was the
foundation of the way of truth. To state now that that which is and that
560
561
562

The same happens with Heraclitus statement emen te ka ouk emen (we are and we
are not) (fr. 49a). This is not an alternative but a conjunction.
Wiesner, who supports nding a third way, interprets this passage of Gorgias in a
radically different way (Wiesner [1996], 101).
Karsten (1835), 152.

The Negation of the Thesis, Opinions, and the Nonexistent Third Way

133

which is not are the same and not the same assumes that two contradictory
judgments are possible at the same time. And this is so because mortals do
not respect this principle and believe that being and that which is not are
different; that is why they have established (8.39) different names to talk
about them. But as they have been unable to choose, they allow contradictory names simultaneously: enai te ka oukh (being and not-being) (8.40).
Men belong to the akrita fula (6.7) that is incapable of separating (etymologically deciding or, as G. Germani didactically translates, disgiungere)563 one thing from another, and therefore they construct a theory
of reality founded upon the simultaneous presence of contraries. For them,
those contraries necessarily (khreon) (8.54) constitute two forms, instead
of one, according to which everything is full of light and dark night at the
same time (9.3). They forget that the only possibility, namely, that which
respects the principle of non-contradiction, is an alternative: pelenai e` oukh
(being or not [being]) (8.11), a separation: estin e` ouk estin (being or not
being) (8.16). That is why, although they assume that being and not being
are different,564 we may deduce from what they say that they act as if there
were no difference between them, and assume that that which is not, is,
and that which is, is not.565
Finally, we may note that Plato comes to my aid to conrm what I am
saying. Indeed, if Parmenides had already admitted a way in which being
563
564

565

Germani, G., Per uninterpretazione delle vie parmenidee, Annali del Dipartimento di
Filosoa, Universita` di Firenze, Vol. II (1986) 23.
According to Jantzen, mortals do not maintain the existence of being and non-being,
but in their statements the categories of being and not being approach one another and,
consequently, their statements are false (Jantzen, J., Parmenides zum Verhaltnis von Sprache
und Wirklichkeit, Zetemata 63 [Munich: Beck, 1976], 110).
The version of 6.89 proposed by Reinhardt, which consists of putting a comma after
enai (Reinhardt [1916], 87), might give rise to an interesting interpretation of the text, but
not in the sense the author imagines. According to this scholar, tauton kou tauton are
not the predicates of the two innitives, but two terms related to the rst group, to` pelein
te ka ouk enai. It is difcult to grasp the meaning of this version, since Reinhardt does
not offer a translation of it. Nevertheless, given the value he ascribes to nomzein (Etwas sich zu seinem Nomos machen), I believe that Cornford comes fairly close to his
thinking when, following the same syntax as Reinhardt, he translates: Who have determined to believe that it is and that it is not, the same and not the same (Cornford [1939],
32). For his part, Beaufret, who says he is adopting Reinhardts version, proposes this
text: . . . pour qui letre et aussi bien le non-etre, le meme et ce qui nest pas le meme,
font loi (Beaufret, J., and Rinieri, J. J., Le Poe`me de Parmenide [Paris: Presses Universitaires
de France, 1955], 81). Later, his translation changes: . . . dont le lot est de dire aussi bien
etre que netre pas, etre meme et ne letre pas (Beaufret, J., Parmenide, Le Poe`me, 1982
[Paris: M. Chandeigne, 1986] 13). We said that the syntax proposed by Reinhardt could
give rise, despite himself, to an interesting reading of the text, since if the formula tauton
kou tauton were independent, it could become an explanation of the previous clause,
depending on a possible hos esti (with esti understood): for whom (hos) [there is] both
being and non-being; they conceive (nenomistai) both the same and the different (tauton
kou tauton).

134

Logos as the Criterion by Which to Judge the Critique of the Way

and not-being combine and, in some way, being is not and non-being is,
why did Plato decide to write the Sophist to refute Parmenides and maintain,
polemically, that a certain union exists between being and non-being
(240bc), and that it is necessary to force non-being, in certain conditions,
to be and being, in its turn and according to a certain modality, not to be?
(241d). If the mortals of Parmenides had admitted the simultaneous reality
of that which is and that which is not, difference and otherness, whose discovery is the basis of the Sophist, would have preceded Plato by almost a
century.

(b) Logos as the Criterion by Which to Judge

the Critique of the Way Made by Men
Fragment 7 ends in a rather unusual way: the Goddess addresses her disciple and exhorts him to judge by means of logos what she has just said:
krnai de logoi poluderin elegkhon ex emethen rhethenta. This request denitively refutes any interpretation that claims to nd a sort of revelation in
the Poem. For D. Furley, these lines are the most astonishing in the
Poem.566 There can be no doubt that Parmenides Goddess is an extraordinary character. Traditional divinities require reverent acceptance of everything they say; they do not ask for their speech to be judged, much less
that this judgment should be carried out by means of a logos (I shall come
back to this term). Parmenides Goddess is a professor of philosophy567 who
asks the student to analyze her speech. Why does she want this? The text
gives us the answer: what the Goddess has just expounded (rhethenta) is
really a proof (elegkhos) against something that was held to be certain, something habitual, and therefore the speech that contains this proof has a polemical (poluderin) aspect. As Verdenius observes, the participle rhethenta
(perfect aorist) indicates unequivocally that the Goddess is alluding to
something already expounded.568 This already expounded something is an
elegkhos.569
As we know, this term originally had the meaning object of shame,
even derision,570 but already in Parmenides time it was beginning to refer
566
567
568
569
570

Furley, D., Truth as What Survives the elenchos (1987), reprinted in Cosmic Problems
(Cambridge, England: Cambridge University Press, 1989), 38.
Cf. Cordero, N. L., La deesse de Parmenide, matresse de philosophie, in La naissance
de la raison en Gre`ce, ed. Mattei, J. F. (Paris: P.U.F., 1990).
Verdenius (1942), 64. Cf. also Taran (1965), 81; Mourelatos (1970), 91, note 46.
An exhaustive analysis of this notion can be found in Furley, Truth as What Survives.
Cf. Homer Il. 23.342: If you destroy your chariot, you will cover yourself with shame,
says Nestor to his son.

The Negation of the Thesis, Opinions, and the Nonexistent Third Way

135

to a combination of arguments used, generally, to accuse (and, in Platos

time, the meaning of refutation will become predominant). In fact, elegkhos is a collective term, a singular with plural value, since it combines a
series of elements, but these have to be very precise: data, proofs, and arguments. These elements can be used not only to refute, as will happen later
on with Socrates, but also to demonstrate, prove, or justify something, especially in the legal sphere, where it is practically synonymous with proof
(of a crime, for example). An accusation is a series of conrmed facts or
data, and this combination adds up to an elegkhos. As M. Dixsaut subtly
observes, this elegkhos is the proof that [Parmenides] has adduced, if we
may so put it, against the false witnesses of custom, the senses and language.571 This is the meaning of the term in Parmenides.572
These proofs must be judged. The Goddess uses the imperative krnai. This verb (innitive: krnnai) means discern, separate, decide,
and like elegkhos, is often found in the legal sphere with the meaning interpret the proofs, or, directly, judge.573 Parmenides uses forms of this verb
on three occasions and the related noun krsis once. In line 15 of fragment
8, the term krsis is used to make a sort of assessment after having presented
a long series of proofs of the thesis according to which that which is being,
is. This is the thesis that constitutes the true way: the krsis (decision,
judgment) on these things lies in this: it is or it is not. And in the following line, Parmenides says that, in fact, the question has already been decided: kekritai [verb krnnai, it has already been decided] of necessity
that one exists . . . Finally, the verb krnnai is used once more in 8.55, this
time in relation to the world of opinions, in which men decided [ekkrnanto] the existence of separate forms. Before I give my view on the meaning of krnnai here in fragment 7.5, let us look at why the Poem is full of
references to the sphere of law.
We saw in the rst chapters of this work that Parmenides is a philosopher who tries to demonstrate his statements and that, therefore, he holds
that thought must follow a way, that is, a method. This method leads
to knowledge, but even in the rst fragment the sphere of knowledge is
presented as a home (do, 1.25) whose gates are guarded by Dike, symbol
of justice, prodigal in punishments (1.14). The way of the Goddess
abounds in signs, messages (it is poluphemos, rich in news),574 and
571
572
573
574

Dixsaut, M., Platon et le logos de Parmenide, in Etudes sur Parmenide, Vol. II, ed. Aubenque, P. (Paris: Vrin, 1987), 243.
Cf., contra, Furley, for whom the term means refutation (Furley, Truth as What Survives).
Remember that judge in Greek is krites and that one of the terms for court is kriterion.
Leszl, Parmenide, 146.

136

(c) The Meaning of logos in Parmenides

when the moment comes to demonstrate the foundation of the thesis,

Parmenides offers a series of proofs, semata, in fragment 8. In fragment
8 we nd proof or evidence of the absolute, total, and unique character of
the fact of being. In the group made up of fragments 6 and 7, this combination of proofs goes to show that the opinions of mortals are based on sensations and a wandering intellect that strays from its objective. Here we have
an elegkhos, and this must be judged by means of logos, as if it were on
trial. So what is the role of logos at this trial?

(c) The Meaning of logos in Parmenides

Given that Parmenides criticizes a way that is based on the senses, the
temptation exists to translate logos as reason and to see the alternative
feeling vs. intellect in the text. However, we must not fall into this temptation. A detailed study of the term logos from Homer onward shows
that in all its uses there are allusions to a certain reasoning (either in the
criterion that must be applied to combine separated elements to form a
collection, a term that keeps the root leg- [= lec-]; or to speak: legein, in
a combination of words; or to argue, etc.), but it is practically impossible to
know at what moment the meaning of reason arose as the dominant
meaning of the term logos. With regard to Parmenides, although Guthrie
exclaims enthusiastically that in 7.1 for the rst time, the senses and reason
are contrasted and that reason alone is to be trusted,575 I am inclined
toward Verdeniuss interpretation, according to which logos never meant
reason [Vernunft] in the Presocratics.576 As in the vast majority of Greek
authors writing before Stoicism, the term logos in Parmenides fundamentally has the meaning of speech. However, there are various kinds of
speech, and perhaps Parmenides originality lies in his attempt to establish
a hierarchy among them.
The term appears on three occasions in the passages of the Poem that
have come down to us, and I do not think I am exaggerating if I say that,
as with Hesiod, the notion acquires a different meaning if it is used in the
singular or the plural. In the plural, as was already the case with Homer577
(and Hesiod does the same),578 logos refers to a speech (that is, a combination of words) that may be supercial or directly false (as is the case in
575
576
577
578

Guthrie (1965), 25.

Verdenius, W. J., Der Logosbegriff bei Heraklit und Parmenides, Phronesis 12 (1967)
100.
Cf. Homer, Il. 15.393: consoled him with words . . . ; Od.1.55: lulled him with caressing
words.
Cf. Hesiod, Theogony 890 and Works 789: caressing words; Works 78: deceiving words.

The Negation of the Thesis, Opinions, and the Nonexistent Third Way

137

Hesiod).579 For example, in the rst fragment of Parmenides, the Daughters

of the Sun convince Dike with gentle words [logoi] (1.15) to open the
heaven gates of the Goddesss realm. On the other hand, when the word is
used in the singular, it refers to a speech, but to a very special speech.
As we said, Parmenides followsperhaps without knowing he doesthe
example of Hesiod, who uses logos to refer to an account, which is presumed to be true, since it describes the history of humanity.580 Indeed, in
fragment 8, after expounding the proofs or evidence that conrm the
true thesis, Parmenides says: Here I end for you the trustworthy logos and
thought about the truth (8.5051). Once again we have a speech, but the
speech reproduces a thought (which is, in its turn, a reasoning, i.e., a series
of arguments) and it concerns something true: so it is to be presumed that
this logos is also true.
What happens in our fragment 7? The term is found in the singular,
and in the dative. Once we have set aside the meaning reason (there is
no need to say judge by means of reason, that is, rationally), I do not
share the opinion of Verdenius, who gives the dative the value of a modal
dative: judge by arguing (since, according to him, logos has the meaning
argument, debate).581 I believe that as in fragment 2 onward, the Goddess has presented an argumentative speech (and will continue to do so in
fragment 8, where the term logos will reappear); the moment has come to
tell the disciple to judgeby means of (instrumental dative) [this] speech
(logos)the polemical582 proof against the foundation of opinions that she
has just expounded. The alternative sensation vs. reason does not make
sense, because Parmenides also criticizes the activity of the intellect when
it wanders astray. Of course, the senses must be set aside (since they only
apprehend apparent changes and do not grasp the profound reality of the
fact of being), but an intellect (noos) must also be rejected that ultimately
only depends on the mixture of limbs (krasis meleon) (fr. 16) of mortals.583 And as for the faculty of judging, even though it acts positively in
fragment 8 (where it eliminates the second way and only keeps the rst) it
can also make mistakes and judge (ekrnanto, 8.55) that there are opposite
forms. The Goddess asks the listener to judge by means of (or in accordance with) true reasoning the value of the way made by men who know

579
580
581
582
583

Cf. Hesiod, Theogony 226: lying words.

Cf. Hesiod, Works 107: I will end my account (logon) by another . . .; this is an account
of the ages of humanity, which claims to be relating real facts.
Verdenius, Logosbegriff, 99.
Polemical (poluderin) has an active meaning. Cf. Mourelatos (1970), 91, note 46.
According to von Fritz, the intellect wanders in search of eon (Fritz, K. von, Nous, noen
and Its Derivatives in Presocratic Philosophy [Excluding Anaxagoras], I, From the Beginnings to Parmenides, Classical Philology 40 [1945] 239).

138

(d) The Hypothetical Third Way

nothing. The sentence following the judgment will be proclaimed in fragment 8: it has already been decided that the wrong way should be abandoned.

(d) The Hypothetical Third Way

According to my interpretation, in the combined passages 6.49 and 7.15,
I nd a new presentation of the second way, now completed with the viewpoint of its makers and its users. This way is clearly formulated in 7.1
(enai me` eonta), and 7.2 says it is to be rejected. Nevertheless, the vast majority of scholars today nd in this passage the presentation of a third way of
investigation. I stress the word today because for twenty-one centuries nobody ever found three ways, routes, or paths in Parmenides thought. The
strong disjunctions between day (or light) and night in the Proem, between
the only two ways proposed in fragment 2, and between Truth and Opinion, which form part of all that is to be known, were always elements that
excluded any compromise a priori. In the philosophy of the inventor of
the principle of identity and non-contradiction, there is no third possibility
(= tertium non datur).
So let us see how people came to speak of a third way of investigation
in Parmenides. First of all, we may say that today this third way is vox
populi: indeed, 99 percent of works on Parmenides speak of a third way.
According to my interpretation, for this to be possible, rst it is necessary
to relativize or distort the text of the Poem. When the text is distorted, this
automatically nullies the interpretation. This is the case with scholars who
say that when Parmenides presents ways of investigation, only two of them
are thinkable, but there may well be others that are unthinkable. This viewpoint distorts the text of fragment 2.2, in which Parmenides wrote noesai
(to think: nal innitive with instrumental value) and not noeta
(thinkable: passive participle). Moreover, if the latter were right, how
could Parmenides have presented ways characterized by unthinkability?
One of the only two ways, once examined, will be shown to be unthinkable, but a posteriori (fr. 8.17), and rigorously speaking, it could also be
called unthought. And then, if one of the ways turns out to be unthinkable (which is authentically Parmenidean), how could it be differentiated
from other ways that were also unthinkable?
There remains the case of those who have relativized Parmenides
words. For these interpreters it would be possible to admit other ways, but
these would not be ways of investigation. For example, L. Couloubaritsis
shows no doubt in saying that in the presentation of the ways there is a
mythical way of speaking on Parmenides part, and that it is necessary to

The Negation of the Thesis, Opinions, and the Nonexistent Third Way

139

distinguish between the way and what is dealt with on it.584 Nevertheless,
we can reply to this that the whole of fragment 8 is the fullest possible
exposition of the intrinsic value (dramatized by Parmenides, since his contemporaries do not appear to grasp it) of the single term estin, which is,
at the same time, the single word (muthos) that remains of the way (there
remains one single word of the way: is, fr. 8.12). And the semata that
follow are proofs of the word (estin) and are on the way.
Despite what is usually believed,585 K. Reinhardt was not the rst to
nd three possibilities in Parmenides. It was H. Stein, although his three
possibilities do not agree with the classical three ways that were to be
systematized denitively by Reinhardt. Stein recognizes that there is a way
expounded in 6.12, another from 6.4 onward, and nally, another from
8.1 on.586 This position, difcult enough in itself to maintain (indeed, what
difference could there be between the way presented in 6.12 and that expounded in fr. 8?), becomes even more obscure because of Steins very
personal restructuring of the original text of Parmenides, leaving it unrecognizable.
On the other hand, K. Reinhardt scientically systematized three possibilities, which he presents in two different ways, but whose content is the
same: (a) being is, (b) being is not, (c) being is and is not; or (A) being is,
(B) non-being is, (C) both being and non-being are.587 It is impossible to
know the reasons that led Reinhardt to construct this tripartite scheme (although, as we shall see, there are material elements in the Poem that might
justify this process). Once the scheme has been proposed, the three ways
have to be located. There can be no doubt that the rst two, (a) and (b), as
well as (A) and (B) are the two classical ways, the ones in fragment 2. As
for (c) and (C), there only remains a part of fragment 6, from line 4 onward
to try to place them. This then would be the way of mortals (or of Heraclitus, as some have interpreted it) who mix being and not being.
The great majority of cases then go on to say that this way reappears
from line 51 of fragment 8, since this, they hold, is about the way of doxa,
a description of appearances, and presents a probable (fr. 8.60)588 speech
about them. In this assimilation, M. Heidegger, a disciple of Reinhardt,
584
585

586
587
588

Couloubaritsis (1990), 182.

Cf., for example, Kahn, C. H., review of Taran, Gnomon 20 (1968) 126; and Couloubaritsis,
L., Les multiples chemins de Parmenide, in Etudes sur Parmenide, Vol. II, ed. Aubenque,
P. (Paris: Vrin, 1987), 26.
Stein, H., Die Fragmente des Parmenides Per Phuseos, in Symbola philologrum Bonnensium in honorem F. Ritschelii (Leipzig: Teubner, 186467), 780.
Reinhardt (1916) 36.
Even if this were so, it would be the diakosmos, not the world, that was probable (Finkelberg, A., Being, Truth and Opinion in Parmenides, Archiv fur Geschichte der Philosophie
81 [1999] 237). For this reason, I prefer to translate eoikota as apparent in opinions.

140

(d) The Hypothetical Third Way

played a salient part: for this scholar, the third way has the same aspect as
the rst, but it does not lead to being: it is the way of doxa, in the sense of
appearance.589 On the basis of this interpretation, a great majority of scholars nd in the second part of the Poem a plausible speech about appearances, and various Anglo-Saxon scholars call the section of the Poem devoted to opinions the Way of Seeming.
What relation does all this have to the philosophy of Parmenides? Little
or none. Let us begin at the end. In Parmenides, doxa never means appearance. Parmenides is not Plato. In the next chapter we shall analyze
this notion, but we can say immediately that Parmenides says nothing
about appearances because for him there is only one subject of study:
that which is being. On this subject, either the truth can be told or opinions
can be ventured. Therefore, the student must assimilate the truth; that is
what will decide whether everything is full of opinions or not. Lastly, we
may say that, as we shall see, the term dokounta (in 1.31) does not introduce
a new dimension of study; it is simply synonymous with doxa.
The Goddesss speech on opinions in fragment 8 (henceforward learn
the opinions of mortals . . . , line 51 ff.) is the deceitful exposition of a probable (eoikota) cosmic order (diakosmon) (8.60). For Parmenidean logic, the
probable is synonymous with the erroneous, the untrue (and, had he used
the term, the false). The Goddess is clear and precise: when she announces that she is going to present this speech, she says that it will be a
deceitful (apatelon) combination of words (kosmon epeon) (8.52). A probable speech is not a speech about appearances; it is the speech that is apparent, not the object of the speech. And moreover, how can it be believed that
a speech described a priori as deceitful could be probable? As De Rijk
observes, by revealing the basic error in human opinions about the world
(8.5159), the Goddess had fullled what she promised in lines 301 of
fragment 1,590 where she said that all true conviction is absent from opinions. As Hesiods Muses had already said, on the one hand, there is true
speech, and on the other, there are falsehoods (pseudea) that appear like
reality (Theogony 2728).
This is the case with Parmenides speech about opinions. Its purpose is
to tell the listener what to hold on to when listening to a combination of
mere words claiming to be true. When we look in detail at this passage
about opinions, I shall come back to this point.

589
590

Heidegger, M., Einfuhrung in die Metaphysik (Frankfurt am Main: Max Niemeyer, 1983),
120.
Rijk, L. M. de, Did Parmenides Reject the Sensible World? in Graceful Reason, ed. Gerson, L. P. (Toronto: Pontical Institute of Mediaeval Studies, 1983), 47.

The Negation of the Thesis, Opinions, and the Nonexistent Third Way

141

Let us now return to the nucleus of the so-called third way. It would
mix together being and non-being by saying that there is non-being and
that being is not. But why propose a third way, if, as we saw, the second way
says exactly the same? Thus we come to the root of the problem. Those who
maintain that there is a third way of investigation in Parmenides believe
that the second way is the way of not being. If this were so, what does
this way assert? That there is non-being, or that being is not. In both cases it
is assumed that there is being, that is, the content of the rst way, either
because the quality of being is ascribed to non-being (and in order to
state that non-being is, it is necessary to possess the notion of being), or
because it is said about being that it is not. Being is omnipresent, and it
could not be otherwise, since being is a primary, necessary, and absolute
certainty, at least for Parmenides.
The so-called formula of the third way is none other than the formula of the
second way, mixing being and not-being. As there is no way of non-being,
we could state with a certain irony that we accept the existence of a third
way provided that it is admitted that there is no second way; then there
would be two ways: the rst and the third. Those who nd the expression
of a second way in line 2.5, different from that which is found in passages
where being is predicated of not being, which they would call the third
way, do not take into account the fact that the statement of each way (that
is, the thesis and its negation) in fragment 2 is double: the second hemistich
of each formula claries the rst. J. Wiesner wrongly says that the second
way maintains ouk estin,591 that is, just the negation of the verb, and that
when a subject is added there is a mixture of being and not being, or a
third way. This second statement is totally correct, but the second hemistich
of line 2.5 shows clearly that ouk estin already presupposes a subject,
whose necessity is stated in 2.5b: it is necessary not to be. To say that is
already a mixture of being and not being, and the concise ouk estin already
presented the same idea by stating that there is not.
From line 4 of fragment 6 onward, Parmenides presents the foundation
of the second way, that is, the negation of the thesis. How is it possible to
offer a foundation for saying that that which is being is not? It is very
simple: mortals who know nothing do not know the basic thesis (there
is being because it is not possible not to be), and so being and not being
are considered as the same and not the same (6.89). Parmenides stresses
the habitual character of this human opinion, which is supported by a
wandering intellect (plagktos noos, 6.6). It is long habit (ethos polupeiron,
7.3) that leads men to create (plattontai, 6.5) a method that mixes being
591

Wiesner (1996), 84 ff.

142

(d) The Hypothetical Third Way

and not being and, therefore, to state that these are both identical and different at once. But, as we saw, the text of this fragment 6 came down to us
containing a very unfortunate conjecture that allows the introduction of a new
possibility. Indeed, if the Goddess orders the disciple to withdraw from two
different ways (protes ga`r, epeita de) (for from this rst . . . and then from
. . .) it is inconceivable that one of these should be the rst, true way, since
the Goddess would not order the disciple to withdraw from that; therefore,
there must be three ways: the true one, and the two ways that are rejected.592 But an essential fact eludes all these interpreters: it is not Parmenides who speaks of rejecting two ways, but the editor of the Aldine edition
of Simplicius, followed by H. Diels, whose text unfortunately nobody now
discusses. F. Fronterotta conrms that Reinhardts interpretation is based
exclusively upon the completion conjectured by Diels for the end of line 3
of fragment 6.593 A reading of fragment 6 with the gap as it stands, that is,
without any conjecture at all, clearly shows that in this fragment Parmenides presents only two possibilities: one in 6.13 and another in 6.49. If the
gap is lled with a verb that is valid for both ways, as I proposed on the
above pages, it is not necessary to imagine a possible third way. This possibility only appears if we assume that Parmenides used a verb that required
a rejection, and in that case, as there are two ways to be rejected, we have
to imagine that there remains a third way to be adopted, making three
ways in all.
From line 4 of fragment 6 onward we nd what Plato had called a
elegkhos per` tou me` ontos (refutation of that which is not) in the passage
already quoted from Sophist 239b. Its full scope is grasped with the quotation from the rst line of fragment 7, in which the particle gar has a strong
explanatory-causal value:594 mortals mix that which is and that which is not
and nd themselves on a way that is a dead end because it is equivalent to
stating that there are things that are not. This occurs because they are incapable of grasping the scope of the thesis, that is, the absolute and necessary

592

593
594

Heitsch is a good example of this way of thinking. According to this scholar, the two
ways presented in 6.3 ff. cannot be the same as those in fragment 2, since fragment 6
clearly says that both must be avoided. (Heitsch, E., Gegenwart und Evidenz bei Parmenides
[Wiesbaden: Franz Steiner Verlag, 1970], 39). So Heitsch imagines a third way and then
sets out on a long march to avoid the conictas he calls itbetween the dichotomous
scheme of the whole Poem, and the trichotomy of the three ways (Heitsch, Gegenwart
und Evidenz, 4053).
Fronterotta, Essere, tempo, 839.
Cf. Denniston, Greek Particles, 6067. This line refers to the content of fragment 6. Both
Untersteiner and Diels maintained that the text that is called fragment 7 today was a
continuation of fragment 6: it can be considered as a denite fact that B7 and 8 come
immediately after B6 (Untersteiner [1958], cxxvi); 7.1 immediately follows 6.9 (Diels
[1897], 73).

The Negation of the Thesis, Opinions, and the Nonexistent Third Way

143

character of the fact of being. When the Goddesss teaching has been assimilated, then nal judgment can be brought (that is, the moment of the
decision [krsis] arrives) (cf. 8.16), and this is when one single way is eliminated and one single way is kept.

(e) Conrmation of the Existence of Only

Two Ways of Investigation
A third way does not form part of this way of thinking. Indeed, in lines
8.1618 we read that it has been decided, of necessity, that one remains
unthinkable and unnamable [i.e., anonymous: it cannot be expressed,
since it is not the true way], and that the other exists and is genuine. There
is not even the slightest reference to a third way. One of the two ways that
were presented, a priori, as capable of directing thought (2.2), is now595 found
to be deprived of the thought (from noema: it is a-noetos) of its possible
user, who must withdraw his thought from it (7.2). That conrms the Goddesss prognosis: this way is panapeuthea, totally unknowable (2.6), since
one cannot know what one cannot think. Moreover, if we take into account
what is said in the passage 8.3536 about the relationships between thinking, being, and saying, it is obvious that following an unthinkable way we
do not nd any reference to that which is, and for that reason the way is
anonumos, anonymous, that is, etymologically, nameless.596 Finally, we
may say that in the passage 8.17 one single way is eliminated, the one that
is not true (cf. the clause in brackets). This means that if Parmenides had
presented three ways, and if one had been rejected in 8.17, that would still
leave two; but the beginning of fragment 8 is explicit: one single word of
the way [singular] remains.
My rejection of a possible third way is based on the impossibility
of nding a middle term between being and not being, according to the
foundations of Parmenides philosophy. And I have stressed that the Parmenidean schemes are essentially dichotomous, since his philosophy assumesperhaps without being conscious of his discoverythe principles
of non-contradiction and the excluded middle. With this way of seeing
things, it is not possible to admit a tertium between the two contradictory
elements that constitute his system:597 it is or it is not (8.15); it is neces595

596
597

That is, after having been described as a way based on the senses and wandering intellect.
The Goddess speaks in the past: kekritai; its impossibility has already been decided upon.
Already, that is, before fragment 8.
On this term cf. Aristotle, Eth. Eud., 1221a.
Seligman observed that an intermediate position between being and not-being is not allowable (Seligman, P., Being and Not-being: An introduction to Platos Sophist [The Hague:
Martinus Nijhoff, 1978], 5).

144

(e) Confirmation of the Existence of Only Two Ways of Investigation

sary to be absolutely, or not (8.11). Moreover, both being and not being
are absolute598 in themselves (which will not escape criticism by Plato and
Aristotle).
Although the traditional interpretations of Parmenides, apart from very
rare exceptions, nd three ways of investigation in his philosophy, there
are certain, somewhat timid, attempts to consider that in fragment 6 there
is only the presentation of a new possibility, which does not actually
become a way. This is a minor possibility, adopted by some pre-Parmenidean thinkers or by people in general, in an almost intuitive way. A. Pasquinelli was the rst to expound this viewpoint. According to this scholar,
both fragment 2 and fragment 8 are completely explicit on the existence of
only two ways of investigation, since there is no tertium between being and
non-being. But in 6.4 ff. there is a false possibility: the thesis implicit in the
world of doxa, which constitutes a new position.599 When we come to analyze the content of so-called doxa we shall see that here we do not have
a separate possibility from the original dichotomy; and as for Pasquinelli,
although ultimately there are not three ways, there are three possibilities, his position belongs to the traditional schemes.
For his part, L. Taran explicitly states that there is no third way of
investigation in Parmenides,600 against scholars who think they can nd
one, but in his analysis of 6.49, he admits that here there is a doctrine
that is incapable of distinguishing being from non-being, a doctrine that
might refer to Heraclitus (that is, which did not accept the Parmenidean
dichotomy and proposed something different). For this reason, when he
refers to the rejection by the Goddess, Taran states that that rejection relates
to the second way of investigation and the impossible path along which
mortals wander who know nothing.601 That is to say, there is a positive
way and rejection of a negative way and of an impossible path. Finally, A.
P. D. Mourelatos makes a distinction between a negative way and that of
mortals, although he recognizes that the latter lapses into the former.602
To date I have only found a position similar to my own in L. Robin
and E. Loew, although neither of these two scholars drew the nal conclusions that result from admitting the fundamental dichotomy in Parmenides
thought. These two scholarswho, however, did not criticize Dielss conjecture for line 6.3vehemently maintain that there are only two ways of
598
599
600
601
602

Parmenides does not make any difference at all between relative being and absolute
being (Verdenius [1942], 54).
Pasquinelli, A., I Presocratici (Turin: Einaudi, 1958), 397.
Taran, L., (1965), 208.
Taran, L., (1965), 81.
Mourelatos (1970), 78, note 7; and Mourelatos (1970), 91. A similar position can be deduced from the words of Loenen (1959), 94.

The Negation of the Thesis, Opinions, and the Nonexistent Third Way

145

investigation in Parmenides, but, as we shall see, they nd certain specic

nuances within the second way and so, we might say, they also offer three
possibilities, though these are reducible to a fundamental alternative.
For L. Robin, who in 1923 expounded a most subtle interpretation of
Parmenides thought in barely nine pages, the principle of contradiction
obliging thought to opt for an absolute yes or no is thus stated for the
rst time and rests on the ontological impossibility of something claiming
to be different from being it is not nevertheless non-being.603 This rigorous
dichotomy leads Parmenides to work out his system around two ways of
investigation, a true one and a false one; nevertheless, this false way assumes a double direction (fr. 6), which appears to correspond to different
degrees of rationality:604 on the one hand, a way that was perhaps followed
by the Heracliteans, in which being and non-being are considered as identical, and on the other, a way in which they get confused and lost and which
contributes to the creation of a physics based on opinion.605
For his part, E. Loew expounded a similar viewpoint in a series of
works published between 1917 and 1935: he said that in Parmenides, there
are only two ways of investigation, the true and the false,606 between which
tertium non datur. And as there is no not-being, neither is there the combination being and not-being. For this reason, the Goddess warns rst
against the way of not-being, and then, even more vigorously, against the
way of being and not-being, along which men wander.607 It is worth noting
that, for Loew as well, the Goddess rejects two wrong ways.608
Be that as it may, in the work of these two scholars we nd an element
essential to our search: the conviction that the two ways correspond to
aletheia and to doxa, that is, to the two parts that have always been recognized in Parmenides Poem. Thus Robin states that, in Parmenides, there
are two routes or methods: one, that of immutable and perfect truth, to
which logical thought belongs; the other, that of opinion and its different
603
604
605

606
607
608

Robin, L., La pensee grecque et les origines de lesprit scientique, 3rd ed. (Paris: La Renaissance du Livre, 1963), 104.
Robin, La pensee grecque.
Wahl accepts this thesis of Robins: Parmenides only ever speaks to us about two ways;
one, which is inevitable, which is the way of being; and another, which is inaccessible,
which cannot be used, and is the way of non-being (Wahl, J., Vers la n de lontologie
[Paris: Societe de lEnseignement Superieur, 1956], 118).
Loew, E., Das Verhaltnis von Logik und Leben bei Parmenides, Wiener Studien 54 (1935)
3.
Loew, Verhaltnis von Logik, 10.
Loew, E., Das Lehrgedicht des Parmenides: Gliederung und Gedankengang, Rheinisches
Museum 78 (1929) 153. For Guazzoni Foa, there are only two ways in fragment 6, but then
he states that doxa, which has a positive aspect, cannot be identied with either of the
two ways (Guazzoni Foa, V., Il problema delle vie di ricerca in Parmenide [Bergamo: Arti
Grache Mariani & Monti, 1979], 2359, 5567).

146

(e) Confirmation of the Existence of Only Two Ways of Investigation

and changing appearances, condemned by custom and the confused experience of the senses.609 For his part, Loew characterizes the two ways in the
following manner: the rst is the logical-critical way (that is, the way of
truth), whereas the second is the empirical-physical way that corresponds
to the opinions of mortals.610
I accept this position of Robin and Loew, but at the same time, I hold
that the two possibilities offered by Parmenides are radical and do not
allow for any intermediate nuance. In other words, the alternatives Parmenides presents throughout his Poem are always different aspects of the same
main alternative. It makes no sense to nd in one passage of the Poem, for
example, an alternative between a rst true way and a second wrong way,
and in another passage a choice between a rst true way and a third way
represented by opinions. There is one single alternative, since the second
way corresponds to the opinions of mortals, and against this stands its contradictory pole, the true way. Consequently, and before drawing the conclusions that have to be drawn, we must look further at this equivalence between the way of error and the way of opinions, on which I based my ideas
when proposing a solution for the gap in 6.3. Now we need to study a
series of passages that show this equivalence, which are diametrically opposed to the content of the way of truth.
Even in the incomplete form in which it has come down to us, the
passage 6.4 ff. diametrically opposes a way created by mortals who know
nothing, to the one that states that being is possible and nothing[ness] does
not exist. The way of men maintains that that which is not being, is (7.1),
which was the thesis of the second way of investigation. The rst way
stands at the antipode of this dead-end way. Already in fragment 1 the
Goddess had said that her way (that is, the way of truth) was to be found
far from the path of men (1.27), who before they become men who know
are in the realm of night. That is, even the rst fragment speaks of two
ways of investigation.611 The alternative is clear: on the one hand, the rst
way, the true way, that is, the way of the Goddess; on the other, the way
of mortals. The rst way is related to truth; the second depends on mens
way of thinking, that is, opinion.
This same alternative reappears in other passages, in which Parmenides
refers again to two ways of investigation, which set truth against non-truth.
For example, in fragment 2.4 Parmenides tells us that the rst way accompanies612 truth, and for that reason, this way is genuine (etetumos) (8.18).
Its content is trustworthy (piston) (8.50), since knowledge of the heart of
609
610
611
612

Robin, La pensee grecque, 104.

Robin, La pensee grecque, 104.
Cf. Leszl, Parmenide, 158.
According to Ballew, this verb means to follow after, as a servant (Ballew, Straight
and Circular, 192).

The Negation of the Thesis, Opinions, and the Nonexistent Third Way

147

truth is the essential content of the teaching the Goddess offers her disciple
(1.28). In contrast, the second way belongs to the sphere of the not-true.613
This way is completely unknowable (panapeuthea) (2.4), since it is not
possible to say or to think that which is not being. Consequently, this way
is described as unthinkable (anoetos), anonymous (anonumos, i.e., nameless, unnamable), and fundamentally not true. No faith can be placed in
this way, and for this reason, the Goddess had already announced in fragment 1 that no trust can be put in opinions (ouk eni . . . pstis alethes) (1.30).
They amount to a deceitful combination of words (kosmos . . . epeon apatelon)
(8.52). There can be no doubt that the wrong, untrue way is the way of the
opinions of mortals. So everything leads to the basic alternative: truth vs.
opinions, or, if you prefer, the rst way (or rst thesis) vs. the second way
(or negation of the thesis).
In the Poems introduction, the Goddess announces to the young man
who has decided to set out on a hazardous journey to nd her that, rst,
he must learn the heart of truth and then the opinions of mortals. Then in
fragment 2, the disciple, who has become an attentive and, we may suppose, obedient pupil, must listen to what are the only two ways offered to
thinking, and these are none other than the way of truth (2.3) and the
wrong way (2.5). Finally, in fragment 8 the Goddess says that she has just
expounded her reasoning and thought about the truth and that from now
on it will be the turn of the opinions of mortals (8.52). Let us not forget
that in 8.1718 the Goddess returns to the scheme in fragment 2 and expresses value judgments about her teaching: one way is true and the other
is false.
Finally, we may say that the whole of classical antiquity found only
two possibilities in Parmenides Poem: aletheia and doxa. For example, Alexander of Aphrodisias states that Parmenides marched along two ways614
(epamphoteras elthe ta`s hodous), since his doctrine stated that, according to
truth (kataletheian), everything is one, whereas, according to opinion (kata
doxan), principles had been made up to explain phenomena (Met. A.3.984b
= DK 28 A 7). Diogenes Laertius shares this opinion since, according to him,
the philosophy of Parmenides can be divided into two parts: one, according
to truth; the other, according to opinion (IX.22.4 = DK 28 A 1). As I have
already said, this interpretation remained in force until the end of the nineteenth century.615
613
614
615

In the quotations from Parmenides Poem that have come down to us, the word false
(pseudes) is absent.
Somewhat unusually, Untersteiner states that Alexander made a mistake: perche le vie
sono tre (Untersteiner [1958], 34).
In Parmenides, wrote Brucker, philosophia duplex est, vel secundum opionem, vel secundum veritatem (Brucker, I., Historia criticae philosophiae, I [Leipzig: B. C. Breitkopf,
1742], 1158).

148

(e) Confirmation of the Existence of Only Two Ways of Investigation

I have stated my completely dichotomous interpretation of Parmenides

philosophy (according to which there is no imaginable third possibility between the thesis and antithesis) from my 1971 doctoral thesis onward.616 An
article published in Phronesis617 widened the readership that might encounter and possibly adopt my viewpoint. The most direct result of this publication was the adoption, two years later, of my verb arkhesthai, albeit in the
active voice, arkho, by the Greek-American scholar A. Nehamas, who kindly
admits that I had already preceded him618 in his choice.619 Even so, my position is radically different from that of Nehamas, and the fact that this
scholar proposed the same verb as myself does not mean we share the same
point of view. Indeed, I proposed the verb in the middle voice arkhomai,
and, for metrical reasons, I decided upon arxei, the second person aorist. In
contrast, Nehamass conjecture proposed the verb should be in the active
voice and in the rst person future: arxo. Given that it is the Goddess
who is speaking, this conjecture appears more coherent than mine, but it is
not. The reason is very simple: Nehamas explains that the use of arkhein
with apo occurs in L.S.J. (see arkho, I.2),620 but in all the examples given, the
verb is found in the middle voice, and so Nehamas cannot adduce a single
passage in support of a possible use of arkho + apo, either in L.S.J. or in
any classical source. In order to maintain the coherence of his interpretation, which requires a rst person, Nehamas should have proposed arkhomai, which is impossible from the metrical point of view; on the other hand,
arxei, though not an ideal solution, does respect the meter.621
G. Germani adopted my hypothesis in his work Per uninterpretazione
delle vie di Parmenide,622 and without ostensibly basing himself on my

616
617
618

619
620
621

622

Cordero, N. L., Letre et le non etre dans la philosophie de Parmenide, Universite de Paris IV,
directeur: Pierre-Maxime Schuhl, December 1971, passim.
Cordero, N. L., Les deux chemins de Parmenide dans les fragments 6 et 7, Phronesis 24
(1979).
This admission by Nehamas has not prevented certain scholars (for example, OBrien)
from becoming a source of disinformation when they frequently speak of the conjecture
of Nehamas and Cordero, a phrase suggesting that I was inspired by a work that was
published two years later than my own (cf. OBrien, D., Etudes sur Parmenide, Vol. I, ed.
Aubenque, P. [Paris: Vrin, 1987], 22225, who makes eight references to my hypothesis,
and inverts the relevant chronology in six of them. Even alphabetical order is in my
favor.). Moreover, given that my conjectures only have a verb in common (imagined by
us and attested also by Nehamas), whose meaning differs considerably when it is used
in the active or middle voice, it does not make sense to speak of the conjecture of
Cordero and Nehamas.
Nehamas, A., On Parmenides Three Ways of Inquiry, Deucalion 3334 (1981) 110.
Nehamas, On Parmenides, 105.
Inexplicably, Couloubaritsis does not take this syntactical evidence into account and believes that the difference operates [only] on the level of the status of the Goddesss
speech (Couloubaritsis, Les multiples chemins, 27).
Germani, G., Annali del Dipartimento di Filosoa, Vol. II (Firenze: Universita` di Firenze,
1986), 23 ff.

The Negation of the Thesis, Opinions, and the Nonexistent Third Way

149

solution to the gap in 6.3, G. Giannantoni concludes that in B6 not three

but two ways are proclaimed, in a form that is coherent with fragments 2
and 8.623 For her part, D. de Cecco624 bases herself upon my new integration (which she considers extremely fruitful) to propose a very subtle
and original interpretation of the last sentence of fragment 6: the way of
all of them returns to the starting point, in which the Goddess makes a
correction, since the journey towards truth is circular. Finally, we may
say that perhaps with excessive optimism and after mentioning my interpretation, E. Berti states that whatever the correct addition that should be
added to the text (which certainly has a gap, as is proved by the meter),
present-day commentators hold that the ways mentioned by the Goddess
are only two.625
If my interpretation is valid, in the nal lines of the combined whole
made up of fragments 6 and 7 (since, as Untersteiner stated, it is a fatto
acquisto dalla critica that fragment 7 follows immediately on from fragment 6),626 we nd an exhortation to judge by reasoning the polemical proof
that the Goddess has given (7.56) about the way created by mortals. The
basic thesis (that is, the rst way) does not need to be submitted to any
test; as soon as the Goddess states it, at the beginning of fragment 6, she
orders it to be proclaimed (6.2) and tells her listener that later on, he will
begin to study it. In contrast, the opposite way harbors a pernicious virus
beneath the apparent familiarity of long habit. The senses and the wandering, wavering intellect have worked out a tentative but false worldview,
whose foundation was set forth by the Goddess. This way of conceiving
reality through proofs that only someone situated outside this daily life
could adduce has been criticized. The totality of these proofs form a polemical, combative elegkhos. The hearer of the Goddess could blindly accept the
arguments that his teacher dictates to him, but she urges him to judge
these proofs. There will be no appeal from the result of the judgment, the
verdict: one single way remains, and this way can be reduced to the single
word is. The rst lines of fragment 8 contain this verdict.

623
624
625
626

Giannantoni, G., Le due vie di Parmenide, La parola del passato 43 (1988) 226.
De Cecco, D., Parmenide 28 B 6,89 DK, in Esercizi Filosoci (1992) (Trieste: Edizioni
Lint, 1993), 14.
Berti, E., Parmenide, in Le savoir grec, ed. Brunschwig, J., and Lloyd, G. (Paris: Frammarion, 1996), 725.
Untersteiner (1958), xxxvi.

This page has been intentionally left blank.

Chapter VIII: The Meaning of the

Opinions of Mortals
Parmenides Poem is presented as the development of the two possibilities
the Goddess has offered as a program of studies toward the end of fragment 1. Anyone wanting to advance into the realm of knowledge (some
time later this desire to know will be called philosophy) must take into
account that there is one basic and irrefutable truth, but human beings may
not know it. In that case, they will be condemned to wander aimlessly.
Thought must either follow the way of truth (aletheia), which is founded
upon a necessary and irrefutable thesis, or it will be reduced to empty and
contradictory opinions (doxai, plural of doxa). This conceptual dichotomy
(which culminates in a gnoseological monism, since only one word of the
way remains, 8.1), excludes any middle term, and therefore excludes any
third possibility that might result from a combination of both positions.
The Parmenidean problem, as Fulleborn maintained, only allows for two
possibilities: either things are, they have reality; or they are not and they
have no reality.627 This is the alternative corresponding to the two parts that
make up the structure of the Poem. The exposition of the fact that there is that
which is, and that it is impossible that it should not be, is called Aletheia.
The presentation of the hypothetical possibility that argues as if that which
is being did not exist is called Doxa. This attempt to propose an explanation
of reality in terms of a radical alternative had already had certain antecedents in pre-Parmenidean cosmogonies, especially in Hesiod,628 for whom
the muses usually express plausible lies but, when they wish, they proclaim
the truth (Theogony 2728). Parmenides inherits this schema and states it in
terms of philosophical canons and then, if speech captures the reality of
that which is being, it can only be true. From Parmenides on, the notions
of logos (reasoning, argued speech) and of aletheia begin to fuse, to the point
where later philosophy, especially in Platos time, will become involved in
titanic efforts to justify a logos that can be erroneous, mistaken, or directly
false (i.e., pseudes).629 We saw earlier that the trustworthy logos was the ex627
628
629

Fulleborn (1795), 95.

Cf. Schwabl, H., Hesiod und Parmenides, Rheinisches Museum 106 (1963) 139.
Probably Antisthenes is claiming to be the faithful heir of Parmenides when he states that
every logos says the truth (aletheuei), because anyone who speaks says something (ti),

(a) Doxa Is Not Appearance

152

pression of a thought (noema) about the truth (8.50) and that this speech
constitutes the positive aspect of Parmenides thesis.
But we also saw that, as any genuine teacher must, Parmenides,
through the intermediary of the Goddess, warns his disciple about the danger of being seduced by a deceitful (or deceptive) order of words (8.52).
Knowing the error of error is part of the truth, and for that reason it is
necessary for the disciple also to inform himself about the opinions of
mortals, as the last lines of the rst fragment say. The exposition of this
negative part of the thesis is fundamental, since Parmenides shows the hidden aspect of the virus that is apt to contaminate philosophical thought. I
say hidden because, obviously, no one (except perhaps Gorgias) openly
admits that there is nothing, that that which is being, is not. Nevertheless,
long habit leads us to relativize the fact of being, to believe that it exhausts
itself in things (beings, in Greek). If this is so, how can we justify the
absolute and necessary character of the fact of being? Relativizing the fact
of being is equivalent to contaminating it with its negation and ignoring
the fundamental alternative: it is or it is not (8.15). Those who are incapable
of making this choice are a-kritoi (people with no capacity for discernment), and they can only create opinions.
We have seen already in our commentary on fragments 6 and 7 that
opinions (the patrimony of mortals) constitute the way made by men
who know nothing (or know nothing[ness]). Opinions are mere fantasies,
combinations of empty words claiming to replace the truth. There is no
true conviction in them (1.30); they form part of a deceitful speech, certainly persuasive, against which one has to take precautions, as Odysseus
had to tie himself to the mast of his ship so as not to allow himself to be
seduced by the song of the sirens. All interpreters who have decided to
defend the positive value of opinions have had to relativize, or even distort, Parmenides words.630

(a) Doxa Is Not Appearance

Despite Parmenides insistence on always relating the notion of doxa to the
sphere of speech, that is, of knowledge, a large number of interpreters of

630

and anyone who says something says something that is being (to` on), and anyone who
says something that is being, says the truth (aletheuei) (text transmitted by Proclus in his
commentary on Cratylus, 37).
Curd, for example, asks why, if Doxa is false, it constitutes the best possible explanation, but she admits that she has taken the word best from the classic work of Long
(Long, A. A., The Principles of Parmenides Cosmogony, Phronesis 8 [1963] 90107)
(Kenig Curd, P., Deception and Belief in Parmenides Doxa, Apeiron 25 [1992] 112).
Then she goes on to ask whether all human opinions must be rejected, since the kouros

The Meaning of the Opinions of Mortals

153

the Eleateans thought tend to give the term an ontological value, as a synonym of appearance. This error is most common among Anglo-Saxon
scholars, who are apt to describe the second way (and hence the second
part of the Poem) as the way of seeming.631 Here we have a grave sin of
anachronism. As I said in the previous chapter, Parmenides is not Plato.
The term doxa appears three times in the Poem, and twice it is accompanied
by the subjective genitive of mortals (1.30, 8.51). Mortals (subject) have
opinions, that is, viewpoints, assessments, conjectures. There is never a
question of the look, that is, the appearance, of mortals. The third case is
even clearer, since men established opinions in order to name things
(19.1). Parmenides says nothing about appearances, since he knows that
on that subject it would be possible to say something and also say the
opposite. In Parmenides there is no theory, not even a plausible theory, of
appearances. If they exist (and given Parmenides concept of being, the way
of being that might correspond to them would have to be justied, something that Parmenides does not do), nothing true can be said about them.
Some interpreters base their ideas on the term dokounta in fragment
1.31 in order to state that Parmenides took appearances into account. I
said earlier that dokounta is synonymous with doxa (whose root it shares).
It is a question of what seems to mortals, not what appears to them.
Parmenides (who, I repeat, is not Plato) does not make any distinction between being and appearing, simply because this would be contradictory in relation to his philosophy. For Parmenides, thingsta` onta, in
Greekare beings, are particularizations, presentations of the fact of
being, not appearances of it. The difference is enormous. The philosopher
grasps it; mortals believe that there are only these presentations, which
they call things (which for one who has had access to the truth are apparent forms morphas: cf. fr. 8.5) and to which they give a name in order
to recognize them (8.3841), but this has nothing to do with appearances.
The dokounta in fragment 1.31 might have existed (khren: unreal imperfect) if truth had not made itself present. Dokounta, like truth, belongs to the
realm of knowledge.
The fact of always associating the term doxai632 with the subjective genitive of mortals shows clearly and distinctly that when Parmenides de-

631
632

who listens to the Goddess is also a mortal. There should be no need to point out that in
Parmenides, mortal does not have a biological meaning; it is synonymous with the
masses, who believe what is said. The kouros is a mortal who has become a man who
knows (1.3), and therefore does not belong to the bewildered masses.
Cf., for example, Gallop, who speaks of the so-called Way of Seeming (Gallop [1984],
5), but who makes this formula the title of his Chapter V (Gallop [1984], 21).
An exception would appear to be 19.1, where the term doxa, in the singular, forms part
of the modal expression according to opinion; however, it is not an exception, since
line 8.39 states that it was mortals who established these names to designate things.

154

(b) The Object of Opinions

scribes the human condition, he is suggesting that it can only create opinions. If we take into account that the most complete description of this
situation is found in the passage that runs from 6.4 to 7.5, we see that the
way from which it is necessary to withdraw in 7.1 is, without any doubt
whatsoever, the way of doxa, whose formulation matches word for word
the impossible way in fragment 2: that which is not being, is. Let us
remember that already in antiquity the opinions of mortals were considered
as an exemplication of the second way of investigation, that is, the negation of the thesis. When Plato quotes lines 7.12 in the Sophist, he does so,
as we have seen, in order to try to refute the proposition that there is no
not-being (i.e., the thesis, logos, of Parmenides, 237a). Sextus Empiricus is
even more explicit, since he quotes the same lines as a demonstration of
the statement contained in line 1.30, according to which there is no true
conviction in the opinions of mortals (Adv. Math. VII.111). That is, the way
condemned in 7.2 is without any doubt the one containing the opinions of
mortals mentioned in 1.30.

(b) The Object of Opinions

If, as my interpretation states, opinions are mere empty words created by
mortals who know nothing, what do these words claim to reect? To put
it another way: as I do not agree that we nd in Parmenides a speech
about appearances, what is the object of these opinions? What are these
opinions about?
To answer this question, we must be aware of the abyss that separates
Parmenides from Plato. Indeed, for Plato, anyone who opines (doxazei) has
an opinion on something, but this object is neither that which is nor that
which is not (Rep. 478c), but a combination of both (479d) (and for that
reason opinion is halfway between ignorance and knowledge, and in certain dialogues this fact allows him to speak of true opinion or false opinion). Parmenides stands at the antipode of this view. For Parmenides, the
object of opinions is that which is. But on this object opinions present an
empty, illusory speech, that is, they have to be wrong. As W. Leszl observes, without any doubt Parmenides attributes to doxa a character not
only of absence of truth, but also of deceit, error or falsehood.633 Although
Parmenides never uses the term false (pseudes) (at least, it is not found in
any of the quotations of the Poem that have come down to us to this day),
for him, opinion is always false, untrue. A doxa alethes would be inconceivable.
633

Leszl, Parmenide, 223.

The Meaning of the Opinions of Mortals

155

That is, the genuine philosopher (the one who advances methodically
from the basic, indisputable thesis) and mortals who know nothing (among
whom previous philosophers are certainly to be foundas I shall try to
show) share the same object of study. And it could not be otherwise. Since
philosophical thought began (and as I have said above, in the passage on
Parmenides theora) everything became amazing, and everything means all
that is being, which, as we know, is expressed in Greek as ta` onta. But
the same object can be looked at in a deeper or shallower way. When we
speak of water, for example, only the scientist grasps its inner structure, its
logos, as we might say, and expresses it in a formula: H2O. The two molecules of hydrogen, added to one of oxygen, are not visible or audible, and
a wandering intellect would not even know that they existed. A lay, nonscientic viewpoint believes that water is just a liquid, colorless, tasteless,
and odorless element. But this opinion is a description of the way in
which a manifestation of water presents itself to the witness attempting to
dene it.
Exactly the same happens with reality as a whole, with the totality of
everything that is being. The philosopher who has set out on the right way
grasps its truth, that without which there would be no reality: the fact of
being, because by being, reality is. Traditional philosophers (we may say)
or mortals guided by habit opine, for example, that the being of things
is exhausted in themselves, that they are what they are, and as Plato would
say later in the Sophist, the central problem for them will then come down
to poa ka` posa are ta` onta (242c), that is, what and how many beings
there are. The essential question, What is the fact of being that allows there
to be beings? remained unmentioned.
Both precedent philosophers and those amazed mortals who want to
know consider (nenomistai, 6.8) that the principle or principles that they
have found exist. But if this is so, the fact of being and the reality of the
principles coincide, even though such principles may, as tends to happen,
be contrary. On the other hand, for a philosopher who knows that point
of view will be only a combination of words, since anyone reasoning thus
does not know what being is and what not-being is, or to put it another
way, believes that being and not-being are the same and not the same. As
we know, this combination is the nucleus of the negation of the thesis. The
opinions of mortals set out upon the second way of investigation, and for
that reason, they are condemned to fail. Instead of recognizing the alternative it is or it is not, they maintain the combination it is and it is not.
The things that are being are born, die, changethat is, cease to be.
The principles that are being transform themselves, unite, or separate
that is, they become that which they were not, they deny themselves. This
relativism only leads to viewing reality as if it were composed of names.

(c) Doxa and Names

156

(c) Doxa and Names

As soon as Parmenides begins his exposition of the reality based on opinions, he states their arbitrary character: they [mortals] established two
viewpoints to name external forms (8.53). This phrase, which has been
much discussed, deserves detailed analysis. We must not forget that it is
not mortals who are speaking, but the Goddess, and therefore she expounds in her terms: that which is real for mortals is a simple verbal
form, and consequently external. For this reason I have translated morphas as external forms, and if I have related this term with speech it is
because that is its meaning in Homer. Indeed, on the two single occasions
when the term morphe appears in the Odyssey, 8.170 and 11.367, it is accompanied by the genitive epos (word). In our passage, the verb name,
(onomazein) picks up this meaning.
I said that this passage is susceptible to various interpretations because
the numeral adjective duo (two) may refer to forms or to viewpoints (gnomas). So does it mean establishing (katethento, a verb already
used in 8.39 when the Goddess conrms that everything mortals establish
is just name) two viewpoints in order to name external forms, or establishing viewpoints in order to name two external forms? Most interpreters
have followed H. Dielss634 line, for whom gnome katatthestai constitutes a
habitual formula, and consequently, the two terms should not be separated;
therefore, the number must refer to forms. I support associating the number with viewpoints (gnomas) for two reasons: (1) because the meaning
of the term is associated with a sort of criterion (it reappears with this
meaning in the last line of fr. 8: so that no viewpoint [gnome] of mortals
will prevail over you, 8.61), and a criterion is precisely what Parmenides
philosophy establishes (cf. supra the analysis of the verb krnai and the noun
krsis) for judging arguments. But, of course, he only proposes one single
criterion, one single viewpoint: that of the disjunction: either . . . or . . .
The failure of mortals or previous philosophers consists in holding to a
double viewpoint, to the conjunction . . . and . . . , that is, two viewpoints.
This is one of the reasons leading us to link viewpoint with two. The
other reason (2) is that Parmenides probably wanted to contrast his two
viewpoints with the proverbial expression one viewpoint (ma gnome)
in unanimous form.635
Mortals error consists precisely in not having thought it necessary to
propose one single viewpoint: which they did not necessarily bring together (and in that they are mistaken) (8.54). This phrase inherits the dif634
635

Diels (1897), 92.

On this cliche, cf. Thucydides, 1.122 and 6.17; Demosthenes, 10.59; Isocrates, 4.139.

The Meaning of the Opinions of Mortals

157

culties of the preceding one, since once more it is a case of relating a numeral adjective, in this case one (man) (since a literal translation of the
passage would read with which it is not necessary [for them to make]
one). One (that is in the feminine) may refer either to viewpoint
(which is feminine in Greek) or to external form. According to my interpretation, this number conrms what I have been saying: instead of adopting a single viewpoint (ma gnome), mortals established two viewpoints.
That is, there can be no doubt that the number one in 8.54 refers to
gnome and not to external forms.636 And for this reason the Goddess is
able to comment that in that they are mistaken (8.54b). The subject of the
sentence is denitely mortals, the relative (en hoi) refers to the mistake they
have made, and the participle (peplanemenoi, corresponding to the verb planao) describes their attitude: planao means both stray and make a mistake and even lose the occasion (kairos: the opportune moment).637 Mortals have made a mistake, because as the Goddess had already said in
fragment 6, lack of resources drives the wandering intellect in their breasts
(6.56).638 As G. Germani says, mortals do not know where to go, they are
disorientated, they proceed at random in a movement full of waverings.639
And we have already seen that P. Destree had compared this erratic gait to
that of drunks.640
The following lines show that not only have mortals established two
viewpoints, but that, consequently, this choice has led them to distinguish
[ekrnanto]641 a form [the term here is demas, which alludes to form in the
sense of external aspect, a sort of conguration] contrary to itself [anta].
Fortunately, Parmenides expresses this idea in a phrase coordinated with
the preceding phrase by means of the conjunction ka (and), in which
he says that mortals also set [ethento] proofs [semata] [of it] separated [khors] from each other (8.55b56a). This passage is very important because
Parmenides begins to identify mortals. They are people who have offered proofs of something, but instead of these proofs being based on the
conjunction that, as we have seen, is the nucleus of Parmenides thesis, they
are victims of double-thinking that separates two universes. As we shall
see, Parmenides will also offer proofs (semata) of his thesis, but these
proofs are coherent: they are not separated from one another, but derive from one another. As mortals decided to express themselves by means
636
637
638
639
640
641

Cf., contra, Coxon: to name only one Form is not right (Coxon [1986], 220).
Cf. Pindar, Nem. 8.4.
Straying (plakton) is a form of the verb plazo, which is synonymous with planao.
Germani, G., Aletheie in Parmenide, La parola del passato 43 (1988) 202.
Destree, P., La communaute de letre (Parmenide, fr. B 5), Revue de philosophie ancienne
1 (2000) 12.
Coxon translates as choose (Coxon [1986], 221).

158

(d) The Opinion-Makers

of a form that is contrary to itself, logically they must offer contrary

proofs that are separate, even contradictory. So what kind of proofs could
these be?
In the nal lines of our current fragment 8, Parmenides gives various
examples. The fragments that follow, up to the so-called fragment 19, in
which he says that thus these things arose, according to opinion, and thus
they exist now, propose a sort of apparent (therefore, deceptive, because
it is like truth) description of reality. This is an explanation based on
opinions that, as we saw, depend on two opposing viewpoints and even
offer proofs of these.

(d) The Opinion-Makers

Who held the worldview that Parmenides unfolds from 8.56 onward? The
candidates are so numerous that the most sensible position is not to take
them into account. Faithful to his discovery, Parmenides could have said,
as another important person would say centuries later, that he who is not
with me is against me.642 Every philosophical system, ideology, or way of
thinking that ignores that by being, it is will be a victim of opinions,
because it will be condemned to relativize the fact of being and make it
cohabit with its opposite, that which is not-being. To illustrate this position,
in the part of his Poem called Doxa, Parmenides works out a ctitious
philosophical system, a true summing up both of what is said and of the
previous philosophical schools. The question: who is Parmenides arguing
against here?643 It makes no sense. Parmenides is opposing any philosopher,
layperson, or other who has been incapable of recognizing the necessary
and absolute character of the fact of being, and who, consequently, has left
a space open to admitting that that which is not (and even that a certain
way of not-being) can enter into a way of conceiving things.
Most scholars of Parmenides thought have seen in the philosopher a
criticism of Pythagorism. It is possible, but the Pythagoreans are not the
only target of Parmenides criticisms. It is normal that when one is criticizing ideas that one does not share, one should begin by settling accounts
with the predominant ideology of the time. Plato did this with the Sophists,
the Stoics did it with Aristotelianism and Platonism, and perhaps so did
Parmenides with Pythagorism. Let us not forget that at the end of the sixth
642
643

Words attributed to Jesus Christ by Luke (11.23).

There are even some scholars who have maintained that Doxa expounds a theory held
ber Parmenides, Verhandl. d. 52 Vers.
by Parmenides in his youth (cf. Rustow, A., U
deutsch. Philol. u. Schulm. zu Marburg [1913] 164).

The Meaning of the Opinions of Mortals

159

and beginning of the fth century B.C., southern Italy was a region under
strong Pythagorean inuence, and the rst philosopher Parmenides heard
was the Pythagorean Ameinias. Perhaps for this reason, we can detect a
certain abundance of Pythagorean644 notions in Parmenides criticism, but
it is also clear that in the apparent explanation of reality that he expounds
(and whose deceptive character he does not fail to point out) there are
elements that could be found in other philosophical movements.
The didactic example chosen by Parmenides, light (that is, the light of
a ame) and darkness, which reappears in fragment 9 as light and night,645
does not have any special signicance. Fundamentally they are opposites,
irreconcilable opposites. As is explained in lines 8.5758, each is what it is
and, consequently, is different from the other, but the previous lines have
said that there is a difference in the sense of an opposition. However, it is
interesting to point out that a century (or perhaps less) before Plato, Parmenides is aware of the notions of sameness (touton) and otherness
(heteroi) (8.5758) and of the fact that the one assumes the other. The use of
such notions is clearly different (although the philosophers coincide on one
point: both notions dene the identity of every thing), since for Plato, as
everything shares in both these notions, each thing can cohabit with its
opposite; whereas for Parmenides, this would be valid for everything except for the principle that, for him, is the existence of that which is being.646
It is obvious that, apart from certainties that belong to common sense
(and are usually expressed in all kinds of proverbs) some previous philosophical systems, which Parmenides was sure to know of, could be included among the opinions criticized in the Poem.
The indenite (to` apeiron) of Anaximander, which is considered to be
eternal (if the opinion transmitted by Hippolytus, Adv. haer. I.6.1 [= A 11
DK], is authentic) and that then originates everything that is from contrary
germs, might belong to the two-headed (6.5), as well as Anaximenes
evolutionary cosmogony.647 And regarding the falsehood of opinions, it is
644

645

646

647

We shall see below that light and darkness are presented as examples of external forms,
and we know that, according to Aristotle, phos and skotos gured among the ten opposing
principles admitted by the Pythagoreans (Met. A.5.986a22). Cf. Parmenides 8.5659 (pur
. . . nuktos), 14.3 (phaos ka` nuktos). In 8.41 Parmenides speaks of khros, which, according
to Aristotle (De sens., 439a31), was a specically Pythagorean term, which was conrmed
by Aetius: the Pythagoreans called the surface of the body khros (I.15.2).
Earth as a principle is, as we know, only mentioned in doxographical texts (cf. Aristotle, Gen. and corr., II.3.330b13). Aristotle also speaks of two causes (aitas), the hot
and the cold, as for example re and earth (Met., A.5986b3334).
Even so, when Plato nds at least ve forms that are more important than the others (cf.
Sophist, 254c), he does not propose a form that is opposite of, contrary to, or simply
different from that of being, which is why these important forms are only ve (being,
rest, movement, identity, and difference) and not six.
Reinhardt accepted this hypothesis (Reinhardt [1916], 50).

160

(e) The Content of Opinions

probable that Parmenides was thinking of Xenophanes, for whom sometimes that which is held as an opinion is close to the real (fr. 35).648 For
Parmenides, there is nothing more dangerous than the plausible, because it
is similar to the true, although it is not true.

(e) The Content of Opinions

Even at the risk of disappointing possible readers of this book, I shall not
go further with the analysis of Parmenidean doxa. My decision rests on
two assumptions. The rst is that Parmenides himself says that the combination of words (epeon) that he allows himself to present is deceptive (apatelon, 8.52). This term more than any other in the Parmenidean Poem leads
us to think of an antithesis between aletheia and falsehood.649 As I have
repeated in the above pages, Parmenides would have found it difcult to
call a speech deceptive that proposes his own thesis. Therefore, contrary to
certain scholars,650 I think it makes no sense to speak of a Parmenidean
cosmology. My second assumption is this: even if the deceptive character
of this part of the Goddesss speech is admitted, as in my case, if we want
to understand what type of cosmology is contained in the Parmenidean
doxa, we have to resort, almost exclusively, to doxographical witnesses.
Hardly anything can be drawn from the quotations of passages of the Poem
that have come down to us to this day. A clear example of this dependence
on the doxographies is the work of J. Bollack mentioned in note 650, which
is nearly 90 percent made up of texts by Aetius and Cicero, grouped together by Diels and Kranz in testimony A37. Nevertheless, if there is a
Presocratic author about whom the interpreter must completely disregard the
opinion of ancient commentators, that author is Parmenides. The doxographer nearest in time to the real Parmenides, Plato, admits with sincerity
and courage: I am afraid that we may not understand [suniomen] his
words [ta` legomena] and I am even more [polu` pleon] [afraid] that what he
was thinking of when he said them goes quite beyond us (Thaeat. 184a).
So what can be expected of more recent witnesses, especially when they do
not provide textual quotations to accompany their commentaries? Little or
nothing.651 Parmenides (and perhaps the same applies to Heraclitus) is a
648
649
650

651

According to Taran, in this text Xenophanes is saying that, on certain subjects, the only
guide for men is opinion (Taran [1965], 207, note 15).
Germani, Aletheie, 199.
Especially Bollack, author of an extensive work (Bollack, J., La cosmologie parmenidienne de Parmenide, in Hermeneutique et Ontologie. Hommage a` Pierre Aubenque, ed.
Brague, R., and Courtine, J. F. [Paris: P.U.F., 1990] 1753).
In the chapter dedicated to the study of fragment 2, I already displayed my surprise at
the fact that a decisive text like this had not drawn the attention of ancient commentators,

The Meaning of the Opinions of Mortals

161

philosopher who must be interpreted from what remains of his writings,

and ancient testimonies should only be listened to if they are commenting
on texts that we possess and know what they are saying.
There are two types of content among the texts making up the Parmenidean doxa. Some describe a cosmology that perhaps some philosophers followed, but, given the principles established by the Poem, they come within
the sphere of ignorance of the thesis that by being, it is, and therefore lack
any kind of truth. We will concern ourselves with these texts very briey.
Other texts offer a sort of reection on such a cosmology, and we shall also
look at those. Finally, there is a text which is difcult to place, because
although it deals with a physical problem (the material foundations of
thought), its consequences are felt in the sphere of thought: fragment 16.
We shall also look at that text.
Let us begin with the group of texts describing a cosmology that
Parmenides, as I have said, not only does not share, but considers deceptive. The group includes fragments 10, 11, 12, 13, 14 and 15.652 Fragments
17 and 18653 have a content we could describe as biological, but that nevertheless falls within a cosmogonic sphere. A brief reading of the context
of the passages in which various commentators (Plato, Clement, Simplicius,
Stobaeus, Plutarch, Sextus Empiricus) have quoted fragments 1015 shows
us that in every case the authors appear to know of a Parmenides text that
we do not, and that they have extracted certain lines from a precise context
to illustrate topics that interested them. Needless to say, we shall never
know what the original Parmenidean context was in which each quotation
was to be found.
Fragments 17 and 18 simply conrm that Parmenides was interested in
biological, gynecological, or directly medical problems. However, these
texts pose a very serious problem. Indeed, do they expound the same Parmenidean ideas? If they do, we would have to reconsider the value of Parmenides doxa. I believe that, once again, Parmenides is expounding what
is said about topic A or B; in this case, conception and birth.
The text preserved today as fragment 16 gathers, I believe, genuinely
Parmenidean ideas about mortals way of thinking. We could say it presents the material foundation of erroneous thinking, since it says that what
thinks (hoper phroneei) is the nature (phusis) of bodily limbs. Like everything
belonging to the sphere of doxa, we suppose that these limbs are made up
of opposing principles and that they determine the way of thinking since,

652
653

and that for more than a thousand years (until Proclus and Simplicius decided to quote
it) they appear to have ignored it (unless possible quotations from it in works dating
before the sixth century A.D. have been lost).
Cf. a translation of these fragments at the end of this book.
See previous note.

162

(e) The Content of Opinions

ultimately, feeling (aisthanesthai) is thinking (phronen).654 B. Cassin and M.

Narcy lucidly state that, in the absence of the authentic Parmenidean context, it is practically impossible to pronounce upon the real problems that
the text presents.655 Even so, I venture the following hypothesis. In Parmenides, that which is is always the basis of thinking, since it is thanks to that
which is that thinking occurs (cf. supra the analysis of lines 8.3336). This
principle is also valid for others, but in that case it is the conception of
being that others (mortals, philosophers) have that will be the foundation
of their thinking, and as we know, from a double, conjunctive base nothing can come except double, conjunctive thinking.
There remain the texts that propose a sort of reection on this cosmology. In this set we can include lines 3841656 of fragment 8 and fragments
9 and 19. The beginning of the speech on the doxa has already conrmed
that mortals established viewpoints to name (onomazein) external forms.
For Parmenides, opinions are a question of words. But these words are just
names (onomata) (8.38, 19.3), mere sound signals (epeon) (8.52), and talking about reality consists merely in naming (8.53, 9.1). But as mortals
are victims of long habit, they believe that something corresponds to these
names, and that that something is real. It is here that Parmenides stands at
the opposite pole to mortals, because only that which is can be said, and
only that which is is real. Birth and dying, changing appearance, changing
position (changing places), even being and not-being (8.3841) are mere
names; the real, the true cannot be reduced to these labels stuck onto
things that are, but are not real.
The listing of names by Parmenides is not random. We saw that
among them are being and not-being (eina te ka` oukh) (8.40). This example luminously shows that, as I said above, the object of study of opinions
and of the thesis is the same. The difference consists in the way of confronting these objects. The positive way, which is the development of the thesis,
grasps the truth of both and judges (krsis) that a disjunction is required:
either it is completely, or it is not at all (8.11). Opinions of mortals, who
do not know what being is (and therefore do not know why that which is
not should be rejected) accept both being and not-being at once. They do
not know that the decision is to choose between it is or it is not (8.15).
Fragment 9, which belongs to the texts offering reasons in favor of the
choice by mortals of a dualist cosmology, conrms that everything [panta]
has been given the names [onomastai] of light and night and that in virtue
654
655
656

Cf. the sources of this text: Aristotle (Met., 1009b21) and Theophrastus (De sens., 3).
Cassin, B., and Narcy, M., Parmenide sophiste, Etudes sur Parmenide, Vol. II (Paris: Vrin,
1987), 291.
For Ebert as well, this passage forms part of the content of the Doxa (Ebert, T., Wo
beginnt der Weg der Doxa? Phronesis 34 [1989] 123).

The Meaning of the Opinions of Mortals

163

of both possibilities,657 that name is applied658 to this or that. But as the only
dunameis are light and night, it can be said that everything is full of both at
the same time (homou). And as only they exist, between them there is nothing
(meden). If the whole is made up of two principles, and besides them nothing exists, these principles represent that which is; but in that case, they
could not be opposites. Plato, who doubtless takes his inspiration from Parmenides, says that when a philosopher states the existence of two contrary
principles, he assumes, without daring to confess it, that being is a third
thing that provides its existence to the two principles (cf. Sophist 243d).
Finally, fragment 19 appears to conclude the presentation of doxa, since
it refers to a current (nun) state of affairs, which has already been (ephu)
produced. As it is talking about the explanation of reality proposed by
mortals, this current state of affairs is a consequence of opinion: things are
like this according to opinion (kata` doxan) (19.1). Each thing is represented
by its own onoma, established by men. In accordance with his circular
method, Parmenides repeats his starting point of 8.38: men have established
names and things are like that now.

657

658

The term that I translate as possibilities is dunamis. This notion will be present in
Plato with a meaning it surely already has in Parmenides time: the power to do or suffer
an action. My translation possibility tries to recover this double value of dunamis, which
is not only active, as is usually believed, but also passive.
On the idea according to which a name rests on (ep) the thing, cf. supra, my analysis
of line 8.35.

This page has been intentionally left blank.

Chapter IX: The Foundation of the Thesis:

The Way of Truth
After showing in fragments 6 and 7 that the negation of the thesis leads to
a dead end (and that, therefore, thought must withdraw from it; 7.2), it only
remains to bring proofs in favor of the only way that remains. Fragment 8
begins with precisely these words: So there remains one single word659 of
the way. Through the use of the term muthos, this formula returns to the
beginning of the Goddesss speech, where she had announced (in fr. 2) that
her muthos would consist in presenting the only ways of investigation
there are to think (2.2). We know that both possibilities were critically
examined, and in lines 17 and 18 of fragment 8 the Goddess conrms that
one of the ways has been abandoned. Therefore, we suppose, there remains
one single way, and that is represented by a single word, the same that
characterized it in fragment 2: estin.

(a) The Only Way That Remains

Only two authors have quoted these rst lines of fragment 8, Sextus and
Simplicius. On the other hand, from line 3 onward, the citations are superabundant. This shows that, for a good part of the classical tradition, it is in
this fragment 8 that the nucleus of Parmenides thought is to be found. Let
us return for a moment to the rst two lines. I said that only Sextus and
Simplicius had quoted them, but it is important to point out that both authors saw clearly that the way presented in them is opposed to the possibility of the existence of that which is not, criticized in fragments 6 and 7, that
is, the negation of the basic thesis. The way that exposition begins in fragment 8 is, without any doubt, the opposite way to that of opinions which
ignored the principal and only thesis. The text I have proposed for the
beginning of fragment 8 corresponds to the version transcribed by Simplicius on page 142 of his Commentary on Aristotles Physics (in other passages, on pages 78 and 145, there is a slightly different text), which makes
clear that these lines follow immediately after the text preserved as fragment 7: after rejecting the way that investigates non-being, withdraw your
659

On the translation of muthos as word, cf. Chapter III (a).

166

(a) The Only Way That Remains

thought from this way of investigation [fr. 7.2], [Parmenides] continues:

so there remains one single . . . Phys. 78). Further on, Simplicius conrms
this assessment: here there is what he says after the exclusion [anaresis] of
non-being: so there remains one single word of the way . . . (Phys. 146).660
There can be no doubt that for Simplicius the way that now begins is the
negation of the possibility that that which is not-being, is (7.1), which was
the second way. Here, as in the rest of the Poem, there is no vestige whatever of a third way.
The same happens with the quotation of these rst two lines by Sextus.
Here we nd a curious fact, since Sextus quotes in continuous form the last
line of fragment 7 (which consists of a hemistich) and the rst line of fragment 8 (which also consists of just one hemistich).661 The anomaly is the
following: Sextus only quotes the last ve lines of fragment 7, and this
quotation comes immediately after line 30 of fragment 1. This means that
Sextus lacks the sentence for this shall never prevail: that things that are
not being, are [enai me` eonta] (7.1). That is to say, for him, the way from
which it is necessary to withdraw mentioned in 7.2 is the way stated immediately beforehand, which, in his version, is not our text 7.1, but the opinions of mortals in which there is no true conviction (1.30). In the most
natural way possible, for Sextus, opinions are the way that states that
things that are not being, are. Nevertheless, it is odd that the last two
lines of fragment 1 should be missing from Sextus quotation, and that his
text runs on immediately into fragment 7.2. This is not the moment to venture a hypothesis about this anomaly, but we might imagine that Sextus
had an abbreviated version of the Poem in which, now as plainly as possible, the way presented in 8.1 follows immediately upon the suppression of
the way that afrms the existence of not-being presented in 7.2.662 It is worth
pointing out that both Sextus and Simplicius give us to understand that a
way has been eliminated in fragment 6 and that another has been retained,
and this is what the text they quote from the beginning of fragment 8 is
concerned with. D. Furley sums up these stages of Parmenides reasoning
thus: Either p, or q, or both p and q; but not q; and if not q, then not both
p and q; therefore only p remains.663

660
661
662

663

On the context of Simplicius quotations, cf. Stevens, A., Posterite de lEtre: Simplicius interpre`te de Parmenide (Brussels: Ousia, 1990).
Furley nds in 8.1 the reply to the judgment carried out by the logos in 7.56 (Furley,
Truth as What Survives, 44).
Sextus text is different from that of Simplicius. In it we read de ti (a reading that is
also found in Simplicius, Phys., 78) thumos. Vitali follows this version, but unexpectedly
translates thumos as conoscenza (Vitali [1977], 35).
Furley, Truth as What Survives, 39. Despite the clarity of this scheme, the weight of
prejudice leads Furley to say that Parmenides rejects all the ways, except one (Furley,

The Foundation of the Thesis: The Way of Truth

167

So the only way that remains is the fundamental thesis. The other possibility presented in fragment 2 was abandoned because a precise analysis of
its content showed that, in reality, it does not exist. Those who claim to
follow this nonexistent way are merely playing with words: they believe
they are nding the meaning of reality, and only nd what they themselves
have created: words, names. This way is a vicious circle, and therefore has
already been abandoned. This the Goddess proclaims when, in the middle
of the arguments she presents in fragment 8, she engages in a sort of recapitulation. Her use of the perfect tense shows that the exclusion of one way
(just one; so just one remains, since at rst there were only two) has already
taken place and has been necessary: it has already been decided [kekritai],
of necessity [anagke], that one remains unthinkable [anoeton] and unnamable
[anonumon] [since it is not the true way] and the other exists [pelein] and is
genuine [etetumon (8.1618). The exclusion occurred when, faced with a
judgment in fragments 6 and 7, the wrong way could not defend itself
against the polemical proof (7.5) to which it was subjected.
These three lines (8.1618) give us complementary information of great
importance. Although Parmenides does not use the term false (pseudes)
(at least in the passages of the Poem that have come down to us), we now
know that one of the ways is not true. It is worth pointing out that fragment 8 completes the polar opposition between the two ways already established in fragment 2: one way was accompanied by the truth and the other
was completely unknowable. Now we know that, in addition, this second
way is not true. Let us remember that as the negation of the thesis, this way
was self-contradictory, since it claimed to state a term with respect to its
negation: of that which is being, it said that it did not exist, and that it was
necessary that it should not exist (2.5), and already in fragment 2 Parmenides had said that it was not possible to know or to utter that which is
not. Fragment 8 conrms both impossibilities: the way was abandoned
because it is unthinkable and unnamable. If we take into account that, as
fragment 3 states, thinking and being are the same we understand that
that which is not is unthinkable. Although the epithet anonumon is also
negative, it is more difcult to interpret. I prefer to translate it as unnamable, with the meaning impossible to utter, basing my ideas on the trilogy be-think-say that was discussed in Chapter V. Nevertheless, the term
could also mean without name, as in a passage of Aristotles Nicomachean
Ethics,664 an allusion, perhaps, to the human habit of naming everything.

664

Truth as What Survives, 38). If this scholar believed in his own interpretation, he would
have to recognize that only two ways are possible, p and q, and that one is rejected and
one is retained.
Passage 1107b2, which says that various states of mind do not have a name of their own.

(b) The semata of estin

168

(b) The semata of estin

At the beginning of fragment 8, the Goddess says that one single word
remains as the content of the way: is. Thanks to this single term, an assimilation is produced between is and the way, to the point where the
explanation of the meaning of is is found in a series of semata which are
upon (ep) the way. These semata are multiple (polla), which means that
the single word is possesses unsuspected riches. This was brought out
by Parmenides by means of the opposition between one single word and
multiple semata. As I said, these semata are upon (or on) the way. Up
to this point, Parmenides has presented his ideas directly, and it was only
regarding the wrong way that he gave a sort of justication that was both
theoretical and practical: theoretical when in fragment 2 he said that that
which is not cannot be the object either of thought or of utterance; practical when he showed, in fragments 6 and 7, that this wrong way is untravelable, since it is a ction created by mortals. So now proofs have to be
brought in favor of the thesis, that is, the way accompanied by truth. This
way maintained that there is being because it is not possible not to be,
and given the necessary character of this thesis, the fact of being (whose
impossibility has proved unimaginable) acquires a necessary and absolute
character.
The semata of fragment 8 are indications, even proofs (not to say demonstrations), of the necessary and unique character of the fact of being.
They are on the way because the way is a route to travel, on which there
are stages, milestones, and signposts that indicate that thought is going the
right way. This argumentative moment in Parmenides thought is found
between lines 2 and 49 of fragment 8, which constitute a trustworthy reasoning [logos] and thought [noema] about the truth (8.50). But what does
semata mean in this context? The singular term is sema, and its meaning
is very broad, but in every case there is a reference to a certain type of sign,
signal, or pointer. This must be the meaning of the word in 8.55 in relation
to the external forms (which we have already discussed above). At the beginning of fragment 8, as Parmenides is speaking of a way, we may consider that the semata are signposts to be found along the way. But what is
the purpose of these signposts? They are witnesses to the necessary and
absolute character of the presence signied by estin, and all that that presence implies. A sema is a semeon: a sign or proof.665 Melissus of Samos, who
without any doubt knew Parmenides Poem, says that the principal semeon
665

Cf. L. S. J.: semeon . . . = sema, in all senses; . . . in reasoning, a sign or proof. This position was to be found before that in Albertelli (Albertelli [1939], 240, note 2), and Simplicius had already stated that to` ga`r o`n . . . a`lla ekhei semea (Phys., 77.30).

The Foundation of the Thesis: The Way of Truth

169

of his philosophy is the argument according to which only one being exists,
but that, apart from that, there are also other semea (fr. 8.1). These are signs
pointing to proofs in favor of a doctrine.
Let us return to the beginning of fragment 8. The single word that
remains as a subject of study, reection, and investigation is, nally, estin,
standing alone. Now we understand why Parmenidean truth is circular
(eukukleos) (1.29) and why the succinct and enigmatic text preserved as fragment 5 says that it is common for me that where I begin, there I shall
return again. The starting point coincides with the arrival point. After a
long journey, the beginning of fragment 8 says exactly the same as line 3
of fragment 2, estin. And just as in the second hemistich of this line 2.3 it
was demonstrated that is is, because it is not possible not to be, in
the semata of fragment 8 it will be demonstrated with proofs that is is
because the fact of being is necessary, absolute, and unique.
Parmenides says that there are many proofs (polla) of (or upon, ep)
the single term that remains as a valid way. This statement conrms that
the way and its content have fused: the proofs of the existence of the way
will be the proofs of the existence of estin. Then an unusual fact conrms
that estin is inseparable from the subject that it itself has produced
(since, as I have said on a number of occasions, is can only be said of
that which is being, since the proofs of the absolute, necessary, and
unique character of is refer to being [eon]: that which is being represents is, since only that which is being, is). Hence all the semata of
fragment 8 are sorts of predicates, attributes, or properties of eon, with
which Parmenides very rarely uses the article to` to turn it into a noun,
since the participle, without an article, which I have translated as being,
captures more precisely the dynamic character of the presence denoted by
estin, because by being, it is.
Fragment 8 will present the foundations (proofs) of the thesis stated
rather dogmatically in line 2.3: [that which is being] is, and [or because]
it is not possible not to be. I said at the time that this thesis assumes (or
postulates) the necessary and absolute (and therefore unique) character of
the fact of being. This thesis is set against its negation as a contradiction,
and this opposition occurs between absolute possibilities, excluding an intermediate term. In the middle of the presentation of the proofs in support
of the necessary character of the thesis, in fragment 8, Parmenides reminds
us that the fundamental decision (krsis) (8.15) is between estin e` ouk estin
(is or is not), but this alternative is between absolute terms: It is necessary to be absolutely [pampan pelenai] or not [be absolutely] (8.11). But why
does the fact of being that characterizes that which is being have this absolute character? Lines 36a of fragment 8 list a series of proofs that lead
to this conclusion.

170

(c) The Field in Which the semata Operate

(c) The Field in Which the semata Operate

It is not my intention to go more deeply into the study of the semata in
fragment 8.666 I will only refer to a few points and, in particular, we shall
constantly remind ourselves that, as these semata concern the fact of being,
that is, a sort of activity, potency, even energy, any analysis of them in terms
of spatial coordinates makes no sense. It is Melissus, Parmenides illegitimate
heir, who will study this fact of being in terms of time and space and,
hence, will deduce that as the void does not exist, being has nowhere (hokei)
to go (!) and therefore is one; and as it was neither born nor will die, it is,
always (ae ) was (!) and always will be (!) (fr. 2). Both space and the three
temporal tenses are inapplicable to the fact of being, which, at most, can be
recognized as a permanent presence.667 Parmenides does not forget that already in fragment 1 he used allegorical images to present his ideas, and he
does the same in this fragment 8, in which Dike and Moira will reappear
and Ananke will enter on the scene. References to limits, chains, and even
a sphere clearly have an allegorical value, as was the case with light and
darkness in fragment 1, since who can imagine that being can be chained,
unless it is metaphorically?

(d) The First sema: That which Is Being Is Everlasting

Parmenides begins the presentation of the semata of is with a sort of summing up of the characteristics that will be developed from line 6b onward:
that which is being is unbegotten and incorruptible, whole [oulon], unique
[mounogenes], unshakable [atremes], and nished [teleston]. The way of
presenting these rst semata by Parmenides is valid for those that will come
later. The Greek text says: ageneton eo`n, ka` anolethron estin, literally, being unbegotten, it is also incorruptible. The subject, which once more is
assumed, is none other than the fact of being, and will only reappear in
line 19 as that which is being, to` eon. Lines 319 develop an authentic
demonstration, in which argument from the absurd plays an essential part,
without what is said therefore referring to a concrete subject, which shows
brilliance in arguing.

666

667

On this subject I recommend the excellent analyses of Taran (Taran [1965], 82160); Bormann (Bormann [1971], 15079); and, with respect to lines 521, Wiesner (Wiesner, J.,
Die Negation der Entstehung des Seienden: Studien zu Parmenides B 8.521, Archiv fur
Geschichte der Philosophie 52 [1970] 134).
Ruggiu speaks of a timeless presence (Ruggiu [1975], 251).

The Foundation of the Thesis: The Way of Truth

171

The Parmenidean machine sets out to prove that that which is being is
everlasting (without temporal origin or ending), and the pseudo-questioning to which the Goddess subjects her disciple (What origin will you seek
for it? How and when might it have increased? [8.67]) is purely formal:
if there is only that which is being, this cannot beget itself or cease to be.
The secondary consequence of this everlastingness of the fact of being is its
permanent presence. The present of estin is not the verbal present tense.
We would need to imagine a sort of temporal presence that endures, whose
intensity is constant and which cannot be controlled through temporal parameters. L. Taran speaks of the timeless present, aloof from time and its
structures.668 Although F. Fronterotta does not share this viewpoint, he asks
a basic question, which I answered in my own way when I looked at the
context in which the semata occur: why, in a word, must being belong to
time?669 According to my interpretation, that which is being has nothing to
do with time that is structured in temporal moments. Parmenides characterizes estin with a present-tense verb because, in Greek grammar, it is the
tense that allows him to show the presence proper to the now (nun), but
that does not mean that this present comes after a past or before a future.
If we keep the category present, as I have already said, here it means a
permanent present.
The everlasting character of that which is, is assured because Dike, who
does not loosen or untie the links, does not permit it either to be born or to
die (8.1314). In line 30, another divinity, Ananke, and in line 37, Moira,
will also exercise similar coercive force over it: the former will keep it in
the bonds or chains that hold it, and the latter will force it to remain unalterable. These three divinities replace the Daughters of the Sun at this
higher stage of the philosophy course. Indeed, the Heliades led the traveler to the heavy gates kept by Dike. Once it had been conrmed, allegorically, that the future philosopher had the right to continue on his journey,
the way led inexorably toward the truth. And as we saw in the previous
chapters, the way of truth is rigorous: it starts from axioms, covers stages,
overcomes problems.670 Randomness and digression belong to opinions.
The true way follows a necessary course. Thought is chained to it and no
straying is allowed. This is the meaning of the links, ties, even chains that
668

669
670

Taran (1965), 180, note 2. Fronterotta criticizes this position of Tarans (Fronterotta, Essere, tempo, 85658). For this scholar, even if being is located in an instantaneous present, it belongs to time, and this condition is necessary to explain the process of knowledge, which implies contemporaneity between subject and object (Fronterotta, Essere,
tempo, 867).
Fronterotta, Essere, tempo, 866.
Remember that problema in Greek means obstacle, barrier.

172

(d) The First sema: That which Is Being Is Everlasting

hold that which is being. The beautiful image of the chains of Necessity
means without any possible doubt the impossibility that being should not
be.671
When he expounds the rst semata (according to which that which is
being is unbegotten and incorruptible), Parmenides says clearly that the
power of Dike is none other than the force of conviction (pstios iskhus)
(8.12). Conviction, which was absent from opinions (1.30), reappears in line
8.28 to exile (that is, send far away) generation and corruption once
again, and if Parmenides stresses this aspect it is because all philosophical
systems have followed a sort of genetic scheme in which a certain principle,
dened a priori as eternal, then becomes this or that (i.e., elements, things)
and thus beings come to take the place of being. This leads to confusion,
and then it is not known what is that which is. For there to be things (beings) that are, there has to be a force that is not subject to the vicissitudes
of generation, destruction, and change. That genuine principle (the others
are illusory) is the fact of being.
The stability and solidity of this necessary and absolute force of being
is the object of two more semata, also given at the beginning of fragment 8:
unshakable (atremes) and nished (teleston). It is impossible not to relate the term unshakable with the same epithet applied to the heart of
truth in line 1.29. When we analyzed that passage, I suggested that truth
possesses a content, a nucleus: metaphorically, a heart. This nucleus is
the total presence of that which is being. And as that which is being is
already that which is (it was not born, will not be born, was not corrupted,
and will not be corrupted), it is nished; it is complete, perfect (etymologically, per- [wholly]; -factum [made]).
For the Greek mentality, the nished is the symbol of perfection, since
it lacks nothing; Parmenides himself says so in line 32: it is not permitted
[ouk . . . themis] that that which is being should be imperfect [ateleuteton].
Parmenides resorts once more to legal terminology: just as Dike did not
allow it to be born or to die, now it is not allowed to remain unnished,
and this prohibition, which in line 32 has an impersonal origin, is in fact an
imposition of powerful necessity (1.30). As I have already said, this coercion
derives from the necessary linking of the argument, which advances by
stages and follows a rigorous method. The links that it establishes tie that
which is being to itself; they are ob-ligatory.672 This is how remaining
identical in the same, it abides in itself (8.29). Unshakable, it resists and
671
672

Reale, G., in Zeller, E., and Mondolfo, R., La losoa dei Greci, Part I, Vol. 3, (Florence: La
Nuova Italia, 1967), 217.
An echo of the relation there is between perfection and the coherent unity between the
parts of the whole will be found in Plato, when he states that it is the good (to` agathon),
i.e., that which unites (deon), that maintains (sunekhein) (Fedon, 99c).

The Foundation of the Thesis: The Way of Truth

173

remains constant (empedon) (8.30). As it does not have to face polemical

proofs, as was the case with the wrong way, it remains wholly intact (asulon) (8.48).673 B. Cassin nds a parallel between these lines and a passage
in the Odyssey, in which Odysseus resists the sirens song and says that
by its self-limitation being escapes the coercion of events, birth and death
[ . . . ] Being, which is, but which never was and never will be is only
immobilized in an identity present to itself if it leaves time and becoming
beyond its limits.674

(e) Immobility
Line 37 says that Moira675 forces it to remain whole (oulon) and immobile
(akneton). This immobility has already appeared in line 26: immobile
within the bonds of mighty chains . . . We have already seen that these
chains, links, and ties are the requirements imposed by the argument that
ties the reasonings; if you like, it is a question of logical necessity.
There can be no doubt that, for inexplicable reasons, immobility, together
with oneness, is the sema that even in antiquity achieved most popularity
and marked Parmenides forever as the philosopher who denied the reality
of movement. Zenos demonstration of the impossibility of a rational explanation for movement (from which its nonexistence was erroneously deduced) has nothing to do with Parmenides. Neither does Melissus demonstration of the impossibility that being should move, as a consequence of
the nonexistence of the void (fr. 7 [7]). Probably, Platos676 invention of the
Eleatic school contaminated Parmenides with these ideas of his fellow citizen Zeno and his reader Melissus (a distant reader, since he lived on
Samos). Given what we have said up until now, the interpretation of this
immobility does not present any difculty whatsoever. As Parmenides
says nothing about the beings who constitute the eld that Plato calls the
sensitive eld, any negation (or afrmation) of the movement of things
673
674

675

676

Inviolable is the translation of Couloubaritsis (Couloubaritsis [1990], 372).

Cassin, B., Le chant des syre`nes dans le Poe`me de Parmenide (Quelques rermarques sur
le fr. VIII, 2633), in Etudes sur Parmenide, Vol. II, ed. Aubenque, P. (Paris: Vrin, 1987),
16869.
Parmenides had already introduced Dike, and also Ananke, if we consider that necessity is also a divinity. According to Reale, they are three names for the same divinity
(Reale, La losoa dei Greci, I.3, 217).
On Platos invention of the Eleatic school, cf. Cordero, N. L., Simplicius et l ecole
eleate, in Simplicius, sa vie, son oeuvre, sa survie, ed. Hadot, I. (Berlin/New York: Walther
de Gruyter, 1987), 16682; and Cordero, N. L., Linvention de lecole eleatique (Platon,
Sophiste, 242d), in Etudes sur le Sophiste de Platon, ed. Aubenque, P. (Naples: Bibliopolis,
1991) 91124.

174

(f) Homogeneity

is excluded. Given the analysis of the fact of being in fragment 8, the relevant question is not, why does being not move? but, why should being
move? Indeed, what sense does it make to apply the category of movement to a necessary and absolute notion? Like most of the semata in fragment 8, movement has an allegorical value, like the limits, chains, sphere
(which we will look at below), and so forth. History has been stricter with
Parmenides than with Plato, whom it has forgiven for saying the opposite:
that existing reality (ousa), that is, Form, moves (kinesthai) (Sophist 248e);
nobody took the expression literally,677 although there can be no doubt that
Plato and Parmenides were trying to say the same thing: kinesthai means
shake, alter. By being known, Form in Plato becomes an object of
knowledge, a noema; now it is not what it was, although the change it has
undergone only consists of acquiring a new property, that of being known.
In Parmenides, that which is being does not alter or change, since any
change would mean becoming what it is not, or acquiring that which it
lacks, and in that case it would have to be admitted that something exists
as well as that which is, which was denied a priori by the other semata.
As it is perfect, nished (tetelesmenon) (8.42), it lacks nothing. With a
touch of irony, Parmenides says that, as it is whole, if it lacked anything,
it would lack everything (8.33).

(f) Homogeneity
Indeed, that which is being is homogeneous (homon) (8.47), everywhere
equal to itself (son) (8.49). There are no degrees of being: it is (absolutely)
or it is not (absolutely).
Who could hold that it is possible to half-exist? The fact of being exists
now (nun) in a wholly homogeneous way (homou pan) (8.5); it is itself
wholly homogeneous (pan estin homoon) (8.22) and wholly continuous (xunekhe`s pan estin) (8.25), so that it presents no gap whatsoever: that which
is being touches [pelazei] that which is being (8.25).
To illustrate this homogeneity, Parmenides resorts to the image of the
perfect solid, the sphere, whose sphericity depends on the homogeneity
of its mass. Indeed, if the mass of a sphere has different densities at some
points, then its surface will be bound to reect this anomaly. A perfect (that
is, well-made) sphere cannot present a rough surface. Everything depends
on the homogeneity of its mass. For this reason, Parmenides states that
677

Cf. the classic article of De Vogel (De Vogel, C., Platon a-t-il ou na-t-il pas introduit le
mouvement dans son monde intelligible? Actes du XIe Congre`s International de Philosophie
XII, Brussels [1953] 6167).

The Foundation of the Thesis: The Way of Truth

175

that which is being is not as if it had a greater quantity of that which is,
either here or there (8.48). That which is cannot be quantied: either it is
or it is not; it is like [enalgkios] the mass of a well-rounded sphere,678 completely equidistant from the center, since it is not possible that it should be
a bit stronger or a bit weaker, here or there (8.4345). Likeness does not
mean identity. To encourage Diomedes, Athena made to spring from her
helmet an indefatigable re, like [enalgkios] the autumn star [Sirius] (Il.
5.5); that does not mean that the star itself sprang from her helmet. Menelaus rises from his bed like [enalgkios] a god, but that does not mean he
is a god. In Parmenides, the fact of being is like the mass of a well-rounded
sphere, because just like the mass that makes this kind of sphere possible,
it is homogeneous, everywhere equally dense. The image is didactic: the
perfection of the curve of the sphere is claried by not a bit stronger or a
bit weaker: it is the negation of all difference in the power or intensity of
being, since any difference of this kind in the manifestation of being would
reinforce the illusions of mortals.679

(g) Oneness
As there are no degrees of intensity in being, it is homogeneous, continuous, one (8.5). Together with immobility, the sema of one (hen) made
Parmenides become the greatest representative of Eleatic oneness. As in
the case of the rst immobility, this does not apply to the object of Parmenides reection either. As we know, only two sources of this line 8.5
have come down to us: Simplicius and Asclepius, and the term hen appears only in Simplicius (Phys. 78, 145). Instead of hen, sunekhes, Asclepius gives us oulophues (Met. 42), which M. Untersteiner, who adopts this
version, translates as is a whole in its nature.680 As we shall see, Platos
criticisms of Parmenides are largely based on the notion of wholeness
that appears to abound in the Poem and that would be contradictory to the
oneness, and in the text offered by Asclepius there is a new reference to the
whole, so it is very probable that Plato knew Parmenides text through this
version of Asclepius.681 Even so, we do not believe there are reasons to
distrust Simplicius text; it is simply a question of interpreting it.
678
679
680
681

Dixsaut translates of a ball (boule) (Dixsaut, M., Platon et le logos de Parmenide, in

Etudes sur Parmenide, Vol. II, ed. Aubenque, P. [Paris: Vrin, 1987], 233.
Dixsaut, Platon et le logos, 23435.
Untersteiner (1958), 145. On the basis of this reading, this scholar maintains that lessere
di Parmenide e` oulon, non hen (Untersteiner [1958], xxvii1).
Reale thinks it possible that both versions are authentic and correspond to different periods in Parmenides life (Reale, G., Melissus: Testimonianze e frammenti [Florence: La Nuova

176

(g) Oneness

What does the statement that that which is being is one mean? Let
us see (1) what can be said about this characteristic; and (2) what must not
be said about it.
(1) As we have already said with respect to movement and other notions assuming spatial-temporal parameters, any reference to a physical
ergo quantiable universe must be excluded. If a philosopher asks a question about the quantity of beings, it is reasonable to expect he will reply
with numeral adjectives: one or more than one. But, as J. Barnes points out:
as far as we know, the question of how many items the universe contains
did not concern him [Parmenides].682 K. Reinhardt had already said that
the predicate of oneness was almost marginal (nebensachlicher) in Parmenides.683 Hen means that that which is being is a total presence that, tautologically, monopolizes the fact of being: Being is the only thing there is.684
In this sense, being is a unique, singular fact. And for this reason, for
the rst time in the terminology of Greek philosophy (unless he was preceded by texts now lost to us), Parmenides, who like all philosophers reects upon ta` onta (things), discovers that if these exist it is because
they have something in common, which is unique, and for that reason
they are considered to be to` on, that which is being. The only oneness
detectable in Parmenides is linguistic; the singular replaces the plural; reection upon on replaces reection upon onta. Just as the life studied by a
biologist is one, although it manifests itself differently in every kind of
living thing, the fact of being that Parmenides discovered is also one,
since there cannot be various kinds of being: it is or it is not (8.15).
(2) In my commentary on opinions (cf. Chapter VIII), I said with
reference to a possible Parmenidean cosmology that the commentaries of
doxographers should not be trusted. The same thing happensI may add
nowwith the opinions of some philosophers about others. In the case of
Parmenides, I cited the confession of Plato (who was both a doxographer
and a philosopher): I am afraid I do not understand his words nor what
he was thinking of when he said them (Theaetetus 184a). This does not prevent him from commenting on and criticizing the Eleatean, as Parmenides
himself had also criticized previous philosophers. The same thing happened with Aristotle with respect to Plato, with the Stoics with respect to
Aristotle, and so on. Philosophy is a perpetual, salutary, ongoing dialogue,
and thanks to this dynamic it will never end. When we have the good
fortune to possess authentic texts, these must be analyzed themselves, and

682
683
684

Italia, 1970], 111). For Barnes, this position is romantic (Barnes, J., Parmenides and the
Eleatic One, Archiv fur Geschichte der Philosophie 61 [1979] 11, note 35).
Barnes, Parmenides, 21.
Reinhardt (1916), 108.
Taran (1965), 190.

The Foundation of the Thesis: The Way of Truth

177

commentators can be listened to when they are based on these texts. Nothing in what remains of Parmenides Poem supports the sentence with
which Plato sums up the credo of the Eleatic group: all things [ton panton] are considered to be one single thing [heno`s ontos] (Sophist 242d). The
sentence makes sense only if it is interpreted in the way I have suggested:
Parmenides is referring to everything as if it were one single thing, because
it, the fact of being, is a common denominator of all things. But we do not
know whether Plato would have agreed with our interpretation. In short,
Parmenides says that the fact of being is unique, not that everything is one,
and certainly not that being is The One. It is true that in the above-mentioned passage of the Sophist, Parmenides is not mentioned, but it is from
this text that the Eleatic group, whose most distinguished representative
was Parmenides, was established. From this to attributing the aforementioned creed to Parmenides is a mere step, which commentators did not
hesitate to take.
In the Sophist, Plato is interested in overcoming a conception of being
that, according to him, led philosophy down a blind alley. An absolute
conception of being does not enable crucial problems to be resolved, such
as the justication of predication, false speech, images. Philosophers contemporary with Plato who were considered to be Parmenides heirs are the
causes of this situation, especially Melissus and Antisthenes. Parmenides is
innocent, but Plato, who is not a historian of philosophy but a philosopher,
battles against a system of ideas as this system had come down to him.
And the one-being, which prevents any justication of multiplicity, the
sensitive universe, or change, is the main enemy.
However, Plato is not tilting at windmills. A philosophical current defends this one-being, and its origins are to be found in Melissus. Perhaps
Plato wanted to suggest to readers of the Sophist that Elea itself was not the
cradle of this conception when he says that these pernicious ideas arose
later starting from Elea.685 But it is certain that Melissus does clearly proclaim the unity of being. Melissus is the only Eleatic who promoted the
theme of the hen to the level of critical knowledge, and who offers a rigorous demonstration of this attribute of the eon.686 To carry out this demonstration, as we know, Melissus distorts Parmenides philosophy because he
makes the fact of being a spatial-temporal Being, whichunlike the dynamic force of that which is being, which is perfect, that is, nishedis
685

686

The Greek text of the passage does not say parhemin (among us), as is read by those
who follow a reviser of the manuscript Parisinus 1808, but parhemon (starting from
us), as the whole of the manuscript tradition attests. Cf. my translations: Platon, Dialogos,
El sosta, Vol. V (Madrid: Gredos, 1988), 403, note 165; and Platon, Le Sophiste (Paris:
GF-Flammarion, 1993), 242, note 193.
Reale, Melissus, 121.

178

(h) Truth

characterized by being innite (apeiron, i.e., without limits) in size (megethos)! The oneness of being is the consequence of this unlimitedness: if
being was not one, it would be limited by something else (Melissus, fr. 5).
There can be no doubt that Melissus could go down in posterity, in the
history books, as the inventor of real monism.687
If Plato combats this conception of the one-being, why does he not
criticize Melissus directly? For two reasons: (1) Plato tends to trust the philosophical culture of the reader (especially if the reader has been his pupil
in the Academy) and he knows his reader cannot fail to be aware that in a
passage of his book Melissus had stated that only one thing exists (hen
monon esti) (fr. 8.1). So when in the Sophist the criticism of the monists
begins, the protagonist of the dialogue asks the anonymous monist: Do
you say that only one thing exists [hen . . . monon enai (in direct speech,
esti)]? (244b). No one can doubt that this is Melissus. (2) It is usual for
Plato to blame the originators of a system for the developments to be found
in those who claim to be heirs of the system, as if the germs of the danger
were already to be found in its origin. This is the case with Heraclitus, who
never wrote the phrase everything is in ux (which, moreover, would be
contradictory to the eternal law of the logos), even though the phrase is
attributed to the Heracliteans (Cratylus 43839) and also exaggeratedly
ascribed to their founder. Doubtless Plato believes that Parmenides absolute conception of the fact of being was responsible for the developments
of philosophers such as Antisthenes, who stated that all speech (logos) is
true (cf. Proclus testimony In Crat. 37), which produced unacceptable secondary consequences for Platos system, since if lying, falsehood, and illusion do not exist, what difference is there between the sophist and the philosopher? Plato wrote the Sophist in order to answer this question, and the
gure to be eliminated was not Melissus or Antisthenes, but Parmenides.

(h) Truth
After repeating that that which is being persists in homogeneous form to
its limits (i.e., it remains protected in its identity), Parmenides indicates
that those words mark the ending of the trustworthy reasoning (piston
logon) and the thought (noema) about (or around: amphs) the truth (alethees)
(8.5051). The pair thought-reasoning deserves an explanation. The noema
has a content, and that content is expressed in speech. As the speech took
the form of an argument, I have preferred to translate logos as reasoning, but speech would also have been a correct translation, since speech
687

Barnes, Parmenides, 21.

The Foundation of the Thesis: The Way of Truth

179

and reasoning fuse. The reasoning was made up of the presentation of a

series of proofs, and as these revolve around the truth, the reasoning
was trustworthy. Here we nd the same schema as in fragment 2, when
Parmenides stated that the way of persuasion accompanied the truth. Persuasion and trustworthiness go together. Sophistry has not yet been
born, and the persuasive logos is not deceptive (as it will be in the case of
Gorgias, cf. Encomium of Helen [fr. 11], 8, logos ho pesas . . . apatesas, the
speech that persuades . . . deceives). If persuasion accompanies the truth,
it is because only the truth is trustworthy, and already in fragment 2, truth
accompanies the way that expounds the thesis. Moreover, in fragment 8 it
will be said that the negation of this way is not a true way, from which
it can be deduced that the other way is true. So can it be deduced that
the way is true because its content is the truth? Yes, but with reservations.
If the content of the way is the truth, it is because the way says something
true. That means that truth will be a prerogative of a logos (speech) that is
presented by a way, and in presenting a true speech, through a sort of
perhaps illegitimate generalization, Parmenides says that the way itself is
true. It is for this reason that I have said that I hesitate to apply the epithet
true directly to the way (that is to say, I do so with reservations).
The problem of truth in Parmenides is very complex. Scholars who
blindly accept the Heideggerian thesis, according to which, at the start,
truth is an ontological category (which is still far from being proved), show
no doubt in stating that as being is true, then a speech that speaks about
being must also be true. Alethes is absent from the semata of that which is
being, which is wholly coherent with the reasoning of fragment 8, in which
Parmenides conrms that the thought expressed up to line 49 was a trustworthy logos that revolved around (amphs) the truth. As always, from
Homer on, truth lies in a speech that, if valid, has the privilege of being
accompanied by the truth, and in that case it can be said that the speech
revolves around the truth, as in 8.49. But in Parmenides, truth is still the
truth of speech. As this speech is a speech about being, Antisthenes (perhaps without thinking of Parmenides) will have no doubt in stating that
that which is, is true. We do not know whether Parmenides reached this
level. Remember that even at the end of fragment 1, Parmenides admits
that truth has a heart, and, according to my interpretation, that heart is
a content: there is being. But to state that there is being is a speech. So
what is true is the speech.

This page has been intentionally left blank.

Epilogue
Any respectable work ends with an epilogue. I would be ashamed to
break this rule, although I feel sad at the idea of writing an epilogue to the
philosophy of Parmenides, since that would mean that his ideas might be
exhausted at any particular time. But an epilogue can also bear witness to
the permanence of certain ideas, and can analyze why an innovative and
revolutionary author could have been marginalized and misunderstood
even by his immediate successors. Several times in this work I have quoted
Platos confession in the Theaetetus: I am afraid I do not understand his
words or what he was thinking when he said them (184a). Nevertheless,
Plato goes on to comment upon (and criticize) Parmenides, and there can
be no doubt that Platos interpretation of the Eleatean is indebted to the
philosophers who presented themselves as the masters heirs. So are there
any philosophers before Plato who explicitly refer to Parmenides? The answer is negative, but this fact is normal. If we leave aside Heraclitus, who
alludes to some of his predecessors,688 treatises by Presocratic philosophers
present themselves as oracular texts without references to the past. Nevertheless, there are two philosophers who, even from antiquity, although they
are not named, are usually associated with Parmenides. They are Zeno of
Elea and Melissus of Samos. Let us begin with Zeno. Can a Parmenidean
legacy be detected in this philosopher? My reply is negative. It is not
enough to be a citizen of Elea, and to possibly have heard Parmenides
speak, in order to share his ideas. If we carefully read Platos testimony on
the relationship between the two philosophers, we can state that Plato himself invites us to distrust Zenos Eleaticism.689 Indeed, if we leave out the
reference to be found at the beginning of Parmenides (128ab) and the Sophist (216a),690 all Platos allusions to Zeno present him as a debater (i.e., an
eristic) and even as a sophist.
688
689

690

Cf. fragment 40, which refers to Pythagoras (who appears again in fr. 129), Xenophanes,
and Hecateus, and fragment 39, in which there is a eulogy of Bias de Priene.
Regarding the Eleatic school, in various passages of this work I have referred the reader
to my articles (Cordero, N. L., Simplicius et lecole eleate, in Simplicius, sa vie, son
oeuvre, sa survie, ed. Hadot, I. [Berlin/New York: Walther de Gruyter, 1987], 16682; and
Cordero, N. L., Linvention de lecole eleatique (Platon, Sophiste 242d), in Etudes sur le
Sophiste de Platon, ed. Aubenque, P. [Naples: Bibliopolis, 1991], 91124).
Cf. my translations of the Sophist, in which I show that even at the beginning of the
dialogue, Plato says that the protagonist, the Stranger of Elea, is different from the
followers of Parmenides and Zeno (Platon, Dialogos, Vol. V, El sosta [Madrid: Gredos,
1988], 332, note 5; and Platon, Le Sophiste [Paris: GF-Flammarion, 1993], 212, note 5).

182

Epilogue

In his Parmenides, Plato makes Zeno say that he wrote his defense
of Parmenides for the pleasure of discussion (philonika) (128d), and the
ctitious691 Parmenides of the dialogue calls Zenos arguments a gymnastic
exercise (an exercise that he recommends to Socrates, an inexperienced
and somewhat dogmatic youth). These speeches by Zeno refer both to a
hypothesis and to its negation (136a), which coincides with the image of
the philosopher presented in the Phaedrus (if it is Zeno who hides beneath
Palamedes of Elea, as most researchers agree). This character spoke in
such a way that he was able to make it appear to his listeners that the
same things were equal and different, single and multiple, at rest and in
movement (261d). And nally, in Alcibiades I, the panorama is completed,
since Plato says that Pythodorus and Callias through having paid a hundred mines to this Zeno, acquired talent and enlightenment (119a). That
is, for Plato, Zeno is a mere sophist.
Various post-Platonic commentators share this viewpoint. Aristotle
makes an enigmatic reference from which it can be deduced that Zeno was
capable of making two different things692 coincide at the same time, and
this skill is also attested to by Isocrates (on what is both at once possible
and impossible, Hel. 3) and for Proclus (on the equal and unequal, In Parm.
620.1 Cousin). We should not be surprised that Pseudo-Galen (Hist. phil. 3
Diels = Dox. gr. 601.89) and Epiphanius (Adv. haer. III.11 Diels = Dox. gr.
590.20) considered that Zeno was an eristic philosopher and that for the
Souda (s.v.) and Diogenes Laertius (VIII.57), Zeno was the inventor of dialectic.
The conclusion is obvious: Zeno did not have his own philosophical
system.693 The testimonies that present him as a disciple of Parmenides depend exclusively upon Plato,694 but only in the Parmenides. In 1971, in a
revolutionary article, F. Solmsen demonstrated conclusively that Plato arbitrarily combined the ideas of Parmenides and Zeno. Although I do not fully
share the viewpoint of this scholar, for whom Zeno represents a modied
version of Eleatism,695 a rigorous reading of the Parmenides (128a) shows
that Plato appears to discover the link between Zeno and Parmenides in
certain ideas that, for Plato, are similar. That means that the similarity between the two philosophers was not something obvious. For example, Aris-

691
692
693
694
695

Parmenides in the Parmenides is an archetype: the philosopher as such. No Parmenidean

idea is expressed by the Parmenides of the Parmenides.
As if it were possible, as in Zeno, to have revenge on both father and mother at once
(Rhetoric, A.12.1372b).
Cf. Cordero, N.L. Zenon dElee, moniste ou nihiliste? La parola del passato 43 (1988)
100126.
Cf. Cordero, Simplicius et lecole and Cordero, Linvention de lecole.
Solmsen, F., The Tradition About Zeno of Elea Re-examined, Phronesis 16 (1971) 140.

Epilogue

183

totle, who devotes a few pages of his Physics to Zeno, never links his name
with that of Parmenides, that is to say Zeno, a citizen of Elea, may have
listened to Parmenides without becoming his disciple. Furthermoreand
fortunatelytexts of Zeno have come down to us, and these show that he
and Parmenides were not talking about the same thing. Parmenides estin
is not an object; it is an inexhaustible, complete, perfect force, which cannot be regarded as either one or multiple, as either divisible to the innite
or indivisible in one part.
From Plato on, posterity has associated the name of Melissus with Parmenides, although this philosopher does not make a single concrete reference to the Eleatean. In the Theaetetus (180e, 183e), Plato mentions both
philosophers as representatives of the tendency that maintains that there is
an immobile One-Being, and from then on anything found in Melissus was
attributed to Parmenides. In the chapter on the semata of the fact of being
(the passage on oneness), we already saw that Melissus is the creator of
the One-Being, thanks to his refutation of the void, and I gave my own
viewpoint there: Parmenides has nothing to do with these ideas.
With or without heirs, it is clear that Parmenides Poem has immediate
repercussions. In Empedocles, practically contemporary with Melissus, and
perhaps even a little earlier, there are echoes not only of the problem but
also of the terminologyincluding grammatical expressions696 that are
found in Parmenides. These details show that Empedocles read Parmenides text, and so did Plato, Aristotle, Theophrastus, Plutarch, and Clement,
since all these authors quoted passages from it. The last direct witness to
the Poem seems to be Simplicius (sixth century A.D.), who allows himself
those are his wordsto quote an extensive passage of it, given the rarity
of the work.697 From then on, no new quotation from Parmenides appears.698 Already-known texts are quoted again, and there is nothing to
show that these sources directly used Parmenides book; they may be indirect quotations. From this series of textual references going from Plato to
Simplicius, the attempt was made to reestablish the lost text of the Poem,
and in Chapter I we looked at the stages of this long process, thanks to
which we can know about Parmenides thought today.
There is no need to say that we shall never know if our knowledge of
the text is precise. And of course that means we shall never know if we
have succeeded in interpreting its content. By way of consolation, I have
696

697
698

Cf. the clear parallelism between line 8.52 of Parmenides (learn the opinions of mortals,
listening to the deceitful order of my words) and passage 17.2627 of Empedocles (Listen to the undeceiving order of the speech . . .).
Simplicius, Phys., 144.
For a detailed analysis of the question, cf. Cordero, N. L., Lhistoire du texte de Parmenide, in Etudes sur Parmenide, ed. Aubenque, P. (Paris: Vrin, 1987), 324.

184

Epilogue

several times quoted the text from the Theaetetus, in which Plato, barely a
century after Parmenides, said exactly the same thing. Be that as it may,
whether faithful to its author or to an approximate image of him, today we
possess something of his thought, which both makes us think and forces
us to converse with him and, if the verb were not too solemn, we might
add, invites us to philosophize. The rest is silence . . .

Appendix 1
Parmenides Poem
(a) Text
The text of the Poem I present here is based on a direct revision of the
manuscript tradition. It differs in certain places from the last orthodox
version by Hermann Diels, as given by him in Die Fragmente der Vorsokratiker. In the footnotes I tell the reader about any changes I have made. A
complete critical apparatus can be found in my work Les deux chemins de
Parmenide.

Fragment 1 o
1

10

15

699

700

o
o o
o
` o`  o,
o,
`
o o` o o } o
.
oo, { ` 699 < >700



o, `
o
o
o
o


o, o
o o`  oo.
}
o <
>
o


oo (oo `

o o
o
o
), o

oo
o,
 o
,
oo

o,

} o
`

} o ` } o
,


o
` } ` o o
o.

`

o

o
` o oo }
oo .

` `
o
o oo

o` o



,


} } o. ` `



` o
o o
} o


oo`

Conjecture. All Sextus Empiricus manuscripts, the only source for this passage, present a
corrupt text: KATAIIANTATH. On this conjecture, cf. Cordero, Le vers 1.3 de Parmenide
(La Deesse conduit a` legard de tout), La Revue Philosophique 107(2) (1982) 158179.
< > Idem previous note.

186

(a) Text

20

25

30

oo `
o


o



` } o o

o`

`
o.
`
o
 o,
`

& } o o
o

`
,
% o

o o
o  oo,

o
o  
o
,

,
` o} o
`
o}

o` o
),
o o (%




`

` .




o

` % o

o, o
}
.

`
o

} `  ` oo
.

o % ` o`

Fragment 2

`

, o ` ` o
o

}



o o` o o
o
 ` o
} `  o

` %,
o
o (
`
o
)

` %,
  o

` 

` o
}

o
o} `
o o `
o` (o
`

o)
o

.

Fragment 3
o` `

o` o
` %.

Fragment 4

o
o

o o
o
`

o o`
o` o
oo }
o} o ` oo
o} o.

Fragment 5
o o

o o }
o o

o % .

Appendix 1

187

Fragment 6

` ` o`701 o
o` }
`
%,
`
}
` o
} 702

703


o o o <}
>704,



`
}
o` , { `
oo`
o o

o,
o
`




o` oo o ` o
o .
o , o, }
,
o` o
o& o` ` o
%
o` o
o

o `
oo
o.

Fragment 7

o
`
o o o % `
o.

` `
o o o %
o
} o o
o o ` ` ,
} oo o} `

o

` o o
} o
`



 .

Fragment 8

10

701
702
703
704

oo } o o oo
 }
` }

o

o` , 
o
o` `

o% o oo `

`
` o
o
o % o
} ,
` } o o ,

, `

o ;
o
; o}
`
oo

o
` o o
`
o` o
` oo
%

} o
o
} . } `
o


o
o, o o`

o, ;

o
.
o


o
` o
`
oo
o



o o
o}
o} o}
,

o` is found throughout the manuscript tradition. Cf. Chapter V (a).

() Manuscript D. Cf. Chapter VI (b).
() Manuscripts B, C. Cf. Chapter VI (j).
< > Conjecture. All Simplicius manuscripts, the only source for this passage, have a gap.
On this conjecture, cf. Chapter VI (g).

188

(a) Text

15

20

25

30

35

40

45

50

705

}  `

` o
}



,
} o
}
o% ,
o (o
`


` `

oo


`
o %.
} o o), `
} o o`
o ;
} oo ;

}
}
}
`







, o (), o o } .
` `
` } o o}
o

`

,
`

o
` o, o }
o ,
o

o
o, } o

oo.
`

o` `

o

`

o


} }
o } o, ` ` o}
o
`
.

,


o

o  o
o} } o %

` `



o
o } , o
`

.
o
o

o o`
o` %.
} `
o

[` ]
o` o`
o.

o`
` o ` o
} o.
o
`
} o
oo,
705& o
,

o` o o
% `
< > } }
} o
o
oo,
` o o


% o
o } } o} o }


oo` o oo %
,
` o} , % ` o,
` oo

o o`
.

`

`
o, o

o,
o

o o} ,
o
o` o`
o} o
o} o
o
o
.
o} `
o

o` } , o o 

o o, o}
o` } o
}
oo
o & o,
`
} o



.
o& `
o %o, o

o ` oo
` o

`
o
o` o
o


o`
o .
oo

o
o
o
` `
o o

-
& o
o

(=
) Proclus;
Simplicius. On my preference, cf. Chapter V (a).

Appendix 1

55

60

189

o ` } o

`

o, ` oo`

o
,
} o o} , [

o]

o,  o
o,

`
o
`

o
o


, o`

.
` oo
oo ,
o o


.

 o
o
o

Fragment 9

`

` o ` ` o

` ` `

` o ` o,
o
` o o o ` o`
o
}
o
,
` o

.

Fragment 10

}





`

o
oo
o }

`  o
oo,
}
o
o
` ,
` ` o

o`
` } o

} o()


} [` `
] } `

} }
.

Fragment 11
`
o
`

o` o

o ` o} o
} o
}

o` o 

.

Fragment 12

 `


`

,

` , ` ` `
%

` {

`
<{ >
` }

}

%
}

.

Fragment 13
` }E

.

190

(a) Text

Fragment 14



.
`
`

Fragment 15

`
`
`
.

Fragment 16
 `

}
,
`



` `

}






` ` ` `

` .

Fragment 17

`
, `
.

Fragment 18
Femina virque simul Veneris cum germina miscent,
venis informans diverso ex sanguine virtus
temperiem servans bene condita corpora ngit.
Nam si virtutes permixto semine pugnent
nec faciant unam permixto in corpore, dirae
nascentem gemino vexabunt semine sexum.

Fragment 19

` } ` }
`
`

} }

 .

Fragment 2

191

(b) Translation
Fragment 1
1 The mares that lead me carry me as far as my will wishes to go, for,
guiding me, they brought me toward the way, full of signs, of the Goddess,
who leads <there>, about <everything>, the man who knows.
4 There I was carried, since the wise mares brought me, drawing my chariot, while the maidens showed the way.
6 The axle, which struck sparks in the hubs, whistled like a pipe (as it was
pressed on at both ends by round wheels) when the Daughters of the Sun,
who abandoned the home of night, hastened to drive me toward the light,
with their hands pushing back the veils from their heads.
11 There stand the gates of the ways of night and day, framed by a lintel
and a threshold of stone. High in the air, both have great double-doors,
whose keys, that alternate, belong to Dike, prodigal in punishments.
15 Coaxing her, the maidens skillfully persuaded her with caressing words
at once to draw back from the gates the bolts that barred them. When the
doors were opened, they made a wide gap, causing the bronze axles to spin
one after another in the hubs, fastened with pins and rivets. There through
the middle of them, the maidens guided the chariot and the mares, straight
along the great way.
22 The Goddess greeted me kindly, took my right hand in hers, and addressing me, spoke these words:
24 Oh youth, accompanied by immortal guides and the mares that bring
you to reach my home, welcome! For it is no sad fate that has impelled you
to take this way (which, indeed, lies far distant from the path of men), but
Themis and Dike. So it is necessary for you to be abreast of everything; on
the one hand, the unshakable heart of well-rounded truth, and, on the
other, the opinions of mortals, in which there is no true conviction.
31 But, nevertheless, you will also learn this: how it might have been necessary that things that appear in opinions really existed, ranging over everything incessantly.

Fragment 2
1 Well then, I will tell youand you who listen, receive my wordwhat
are the only ways of investigation there are to think:
3 one, on the one hand, [to think] that is, and that it is not possible not
to be; this is the way of persuasion, since it accompanies the truth;
5 another, on the other hand, [to think] that is not, and that it is necessary
not to be; I tell you that this path is completely unknowable, since you will
not know that which is not (as it is not possible) or utter it.

192

(b) Translation

Fragment 3
since it is the same to think and to be

Fragment 4
Observe how the absent is rmly present to the intellect; since it is impossible to force that which is not to be connected with that which is, neither
scattering it completely in regular order, nor gathering it.

Fragment 5
. . . it is common for me that where I begin, there I shall return again.

Fragment 6
1 It is necessary to say and to think that by being, it is, since it is possible
to be, and nothing[ness] does not exist. This I order to proclaim since you
<will begin> with this rst way of investigation, but then with that made
by mortals who know nothing, two-headed, since their lack of resources
drives the wandering intellect in their breasts. They are carried along, blind
and deaf, amazed, people with no capacity for discernment, who consider
that being and not being are the same and not the same; the way of all of
them returns to the starting point.

Fragment 7
1 For this shall never prevail: that there are things that are not.
2 But you, withdraw thought from this way of investigation and let not
long habit force you along this way, to use the eye that does not see, the
echoing ear, and the tongue. Judge by reasoning the polemical proof I have
stated to you.

Fragment 8
1 So there remains one single word of the way: is. About it, there are
many proofs that that which is being is unbegotten and incorruptible,
whole, unique, unshakable and nished.
5 It neither was nor will be, but is now, wholly homogenous, one, continuous. What origin will you seek for it? How and when might it have increased? I do not allow you to say or to think that it [came] from that which
is not being, since it is not sayable or thinkable that it is not. What necessity
could have made it grow before or afterwards, beginning from nothing[ness]?

Fragment 8

193

11 Therefore, it is necessary to be absolutely, or not. The force of conviction

will not permit, from that which is not being, something else to be born
beside it, since Dike does not allow it to be born or to die, loosening the
ties, for she holds it. The decision on these things lies in this: it is or it is
not. It has already been decided, of necessity, that one remains unthinkable
and unnamable (since it is not the true way) and that the other exists and
is genuine.
19 How could that which is being perish? From what could it have been
born? For if it was born, it is not, just as if it will be some day, it is not.
Origin is extinguished and destruction is unknown. Neither is it divisible,
since it is wholly homogenous.
23 There is not anything to a greater degree, which would prevent its cohesion, neither is there anything to a lesser degree: it is wholly lled with that
which is being. It is wholly continuous: that which is being touches that
which is being.
26 Immobile within the bonds of mighty chains, it is without beginning and
without end, since origin and destruction remain far away: true conviction
rejected them.
29 Remaining identical in the same, it abides in itself, and, in this way, it
remains constant, since powerful necessity keeps it within the chains of the
limit that holds it, for it is not permitted that that which is being should be
imperfect. Indeed, it lacks nothing at all; if it did, it would lack everything.
34 Thinking and that because of which there is thinking are the same, since
without that which is being, thanks to which it is expressed, you will not
nd thinking, for there is not and there will not be anything else apart from
that which is being, given that Moira forces it to remain whole and immobile. Therefore they are mere names that men have established, believing
that they were true things: birth and dying, being and not-being, changing
place and altering the outer color.
42 But as there is a supreme limit, it is everywhere nished, like the mass
of a well-rounded sphere, completely equidistant from the center, since it
is not possible that it should be a bit stronger or a bit weaker, here or there.
46 Since that which is not-being, which would prevent it attaining homogeneity, does not exist; and that which is being is not as if it had a greater
quantity of that which is, either here or there, for it is wholly intact. Everywhere equal to itself, it remains in homogenous form to its limits.
50 Here I end for you my trustworthy reasoning and thought about the
truth. Henceforward learn the opinions of mortals, listening to the deceitful
order of my words.
53 They established two viewpoints to name external forms, which they
did not necessarily bring togetherand in that they are mistaken. They
distinguish a form contrary to itself and offer separate proofs for the one
and the other; on the one hand, the ethereal re of the delicate, nimble

194

(b) Translation

ame, wholly identical with itself, but not the same as the other; and on
the other hand, that which is in itself its opposite, dark night, which is thick
and heavy.
60 I tell you of this probable cosmic order so that no viewpoint of mortals
will prevail over you.

Fragment 9
But as everything has been given the names of light and night, and that
which has its own powers was named thanks to these or those, everything
is full at the same time of light and dark night, the one the same as the
other, since, apart from them, there is nothing.

Fragment 10
You will know ethereal nature and all the signs that are in the ether, and
the works destructive of the bright suns pure ame, and whence all this
comes; and you will learn the works of the turning moons rotation, and
its nature and you will also know the surrounding sky, whence it was born,
and how the necessity that governs it anchors it to hold the limits of the
stars.

Fragment 11
. . . how the earth, the sun, the moon, the common ether, the Milky Way,
high Olympus and the burning power of the stars came to be.

Fragment 12
1 The tightest [rings?] are full of pure re; the next, of night; but between
them a lick of ame escapes. In the middle of these [rings?] is the Goddess
who governs everything. She rules over fearful childbirth and coupling,
driving the female to go with the male, and, likewise, the male with the
female.

Fragment 13
. . . She conceived Eros, the very rst of the gods.

Fragment 14
Shining by night, wandering round the earth, with borrowed light . . .

Fragment 19

195

Fragment 15
. . . always turned toward the beams of the sun . . .

Fragment 16
Just as on every occasion there is a mixture of prodigious limbs in movement, so the intellect is present in men. Since, for men, both in general and
in particular, the nature of the limbs is the same that thinks; since thought
is the full.

Fragment 17
On the right the boys, on the left the girls.

Fragment 18
1 When the woman and the man mix together the seeds of Venus, the
power that, in the veins, should form bodies with different blood, creates
them well-shaped if it keeps proportion;
4 but if the seed-powers conict and do not unite in the body that results
from them, by their double seed they disturb the sex that is to be born.

Fragment 19
Thus these things arose according to opinion, and thus they exist now. And
then, once they have developed, they will die. To each thing men have
given a particular name.

This page has been intentionally left blank.

Appendix 2
Note on the Transliteration of the Greek Alphabet

To transliterate the Greek alphabet I have adopted the following model:

a
b
g
d
e
z
e
th
i
k
l
m
n
x
o
p
r
s
t
u
ph
kh
ps
o

Smooth breathing: not reproduced

Rough breathing: h (e.g.,
= hule)
Accents: the same (e.g., e, e`, e )
The type of transliteration adopted will allow the reader to reconstruct the
original Greek term exactly. Underlining the letters eta ( = e) and omega (
= o) makes it possible easily to reproduce the cases in which the letter in
question has an accent (e.g., = sema).

This page has been intentionally left blank.

Bibliography

The bibliography given here includes all books and papers cited in this
work. For additional titles, I refer the reader to the new edition of my book
Les deux chemins de Parmenide, 2nd edition (Paris: Vrin, 1997), in which I
have listed 817 items dealing exclusively with Parmenides. I am currently
compiling a comprehensive and up-to-date list of all the material on Parmenides I have collected during my years of teaching and research on the
subject. I look forward to making this available online by the end of 2004
at www.parmenides.com.

Aalto, P. Studien zur Geschichte des Innitivs im Griechischen (Helsinki: P. Katara,

1953).
Albertelli, P. Gli Eleati (Bari: Laterza, 1939).
Allen, R. E., trans. Plato, Parmenides (Oxford: Blackwell, 1983).
Arrighetti, G. Lheredita` dellepica in Parmenide, Festschrift fur Robert Muth zum
65. Geburtstag, ed. Haendel, P., and Meid, W. (Innsbruck: Institut fur Sprachwissenschaft der Universitat Innsbruck, 1983).
Aubenque, P. Syntaxe et semantique de letre dans le Poe`me de Parmenide,
Etudes sur Parmenide, Vol. II, ed. Aubenque, P. (Paris: Vrin, 1987).
, ed. Etudes Etudes sur Parmenide, Vol. II (Paris: Vrin, 1987).
Austin, S. Parmenides: Being, Bounds and Logic (Yale, CT: Yale University Press,
1986).
Baldwin, B. Parmenides in Byzantium, Liverpool Classical Monthly (1990) 115116.
Ballew, L. Straight and Circular in Parmenides and the Timaeus, Phronesis 19
(1974) 189209.
Barnes, J. Parmenides and the Eleatic One, Archiv fur Geschichte der Philosophie 61
(1979) 121.
Basson, A. H. The Way of Truth, Proceedings of the Aristotelian Society 61 (196061)
7386.
Baumann, A. Formen der Argumentation bei den vorsokratischen Philosophen, dissertation (Wurzburg, 1906).
Beaufret, J. and Rinieri, J. J. Le Poe`me de Parmenide (Paris: P.U.F., 1955).
. Parmenide, Le Poe`me, 1982 (Paris: M. Chandeigne, 1986).
Becker, O. Das Bild des Weges und verwandte Vorstellungen im fruhgriechischen
Denken, Hermes Einzelschriften 4 (1937) 1223.
. Drei Abhandlungen [but Bemerkungen in the contents] zum Lehrgedicht des Parmenides, Kant-Studien 55 (1964) 255259.
Benveniste, E. Proble`mes de linguistique generale (Paris: Gallimard, reprinted 1966).
Beraud, J. La colonisation grecque de lItalie meridionale et de la Sicile dans lAntiquite
(Paris: P.U.F., 1957).

200

Bibliography

Bergk, T. Commentatio de Empedoclis Proemio, Kleine philologische Schriften II

(Halle: Buchhandlung des Waisenhauses, 1886).
Bergson, H. Levolution creatrice, 3rd ed. (Paris: Alcan, 1907).
Bernabe, A. De Tales a Democrito (Madrid: Alianza, 1988).
Berti, E. Parmenide, Le savoir grec, ed. Brunschwig, J., and Lloyd, G. (Paris: Frammarion, 1996).
Bicknell, P. J. Dating the Eleatics, For Service of Classical Studies, Essays in Honor
of F. Letters, ed. Kelly, M. (Melbourne: Cheshire, 1966), 114.
. A New Arrangement of Some Parmenidean Verses, Symbolae Osloensis 42
(1968) 4450.
Boardman, J. The Greeks Overseas (Baltimore: Penguin Books, 1964).
Bohme, R. Die verkannte Muse. Dichtersprache und geistige Tradition des Parmenides
(Berne: Francke, 1986).
Bollack, J. La cosmologie parmenidienne de Parmenide, Hermeneutique et ontologie, Hommage a` Pierre Aubenque, ed. Brague, R., and Courtine, J. F. (Paris: P.U.F.,
1990), 1753.
Bormann, K. Parmenides: Untersuchungen zu den Fragmenten (Hamburg: Felix
Meiner, 1971).
Bowra, C. M. The Poem of Parmenides, Classical Quarterly 32 (1937) 97112.
Brague, R. La vraisemblance du faux, Etudes sur Parmenide, Vol. II, ed. Aubenque,
P. (Paris: Vrin, 1987) 4468.
Brandis, C. A. Commentationum Eleaticarum. Pars prima: Xenophanis, Parmenidis et
Melisi doctrina a propriis philosophorum reliquiis exposita (Altona: J. F. Hammerich, 1813).
. Handbuch der griechisch-romischen Philosophie (Berlin: G. Reiner, 1835).
Brocker, W. Die Geschichte der griechischen Philosophie vor Sokrates (Frankfurt: Klostermann, 1965).
Brucker, I. Historia criticae philosophiae, I (Leipzig: B. C. Breitkopf, 1742).
Brugmann, K. Griechische Grammatik (Munich: C. H. Beck, 1913).
Brumbaugh, R. S. Review of Taran, International Philosophical Quarterly 4 (1966).
Burkert, W. Das Proomion des Parmenides und die Katabasis des Pythagoras,
Phronesis 14 (1961) 130.
Burnet, J. Early Greek Philosophy, 4th ed. (London: A. & C. Black, 1930).
Buroni, B. Dellessere e del cognoscere: Studii su Parmenide, Platone e Rosmini,
Memorie della Reale Accademia delle Scienze di Torino, Serie II (Scienze morali,
storiche e lologiche) 29 (1878), 287528.
Buxton, R. G. A. Persuasion in Greek Tragedy. A Study of Peitho (Cambridge, England:
Cambridge University Press, 1982).
Calogero, G. Studi sullEleatismo, 2nd ed. (1932; Firenze: La Nuova Italia, 1977).
. Parmenide e la gene`si della logica classica, Annali della Reale Scuola Normale Superiore di Pisa, Classe di Lettere, Storia e Filosoa, ser. II, 5 (1936), 143
185.
Capizzi, A. Introduzzione a Parmenide (Bari: Laterza, 1975).
. La porta di Parmenide (Rome: Edizioni dellAteneo, 1975).
Casertano, G. Parmenide: Il metodo, la scienza, lesperienza (Naples: Guida, 1978).
Cassin, B. Si Parmenide (Lille: Presses Universitaires de Lille, 1980).
. Le chant des syre`nes dans le Poe`me de Parmenide (Quelques rermarques
sur le fr. VIII, 2633), Etudes sur Parmenide, Vol. II, ed. Aubenque, P. (Paris:
Vrin, 1987).

Bibliography

201

. Leffet Sophistique (Paris: Gallimard, 1995).

. Parmenide: Sur la nature ou sur letant (Paris: Ed. du Seuil, 1998).
Cassin, B., and Narcy, M. Parmenide sophiste, Etudes sur Parmenide, Vol. II (Paris:
Vrin, 1987).
de Cecco, D. Parmenide 28 B 6,89 DK, Esercizi Filosoci (1992) (Trieste: Edizioni
Lint, 1993).
Cerri, G. Elea, Senofane e Leucothea, Forme di religiosita` e tradizioni sapienziali in
Magna Grecia, ed. Casio, A. C., and Poccetti, P. (Pisa/Rome: Istituti Editoriali e
Poligraci Internazionali, 1993).
. Il v. 1.3 di Parmenide: la ricognizione dellesperienza, Mousa, Scritti in
onore di Giuseppe Morelli (Bologna: Patron, 1997).
. Parmenide di Elea: Poema sulla natura (Milan: Rizzoli, 1999).
Chalmers, W. R. Parmenides and the Belief of Mortals, Phronesis 5 (1960) 522.
Chantraine, P. Grammaire homerique, II: Syntaxe (Paris: Kliencksieck, 1953).
. Dictionnaire etymologique de la langue grecque (Paris: Klincksieck, 1968).
Chiereghin, F. Implicazioni etiche della storiograa losoca di Platone (Padua: Liviana,
1976).
Ciaceri, E. Storia della Magna Grecia (Milan: Albrighi, Segati & Co., 1927).
Collobert, C. Letre de Parmenide ou le refus du temps (Paris: Kime, 1993).
Colombo, A. Il primato del nulla e le origini della metasica (Milan: Publicazioni della
Universita Cattolica Sacro Cuore, 1972).
Conche, M. Parmenide: Le Poe`me: Fragments (Paris: P.U.F., 1996).
Constantineau, P. La question de la verite chez Parmenide, Phoenix 41 [1987]
217240.
Corbato, C. Studi Senofanei, Annali Triestini XXII (1952).
Cordero, N. L. Letre et le non etre dans la philosophie de Parmenide, Universite de Paris
IV, directeur: Pierre-Maxime Schuhl, December 1971.
. Acerca de tres pasajes del Poema de Parmenides, Revista latinoamericana
de losofa 1 (1975), 237243.
. Analyse de ledition Aldine du Commentaire de Simplicius a` la Physique
dAristote, Hermes 105 (1977).
. Les deux chemins de Parmenide dans les fragments 6 et 7, Phronesis 24
(1979) 132.
. Le vers 1.3 de Parmenide, La revue philosophique 107(2) (1982) 159179.
. La version de Joseph Scaliger du Poe`me de Parmenide, Hermes 110 (1982)
391398.
. Les deux chemins de Parmenide (Paris/Brussels: Vrin/Ousia, 1984; second edition, augmented and corrected, 1997).
. Les sources venitiennes de ledition Aldine du Livre I du Commentaire de
Simplicius de la Physique dAristote, Scriptorium 38(2) (1984), 7088.
. Lhistoire du texte de Parmenide, Etudes sur Parmenide, Vol. II, ed. Aubenque, P. (Paris: Vrin, 1987), 324.
. Simplicius et l ecole eleate, in Simplicius, sa vie, son oeuvre, sa survie, ed.
Hadot, I. (Berlin/New York: Walther de Gruyter, 1987), 166182.
. Zenon dElee, moniste ou nihiliste? La parola del passato 43 (1988) 100126.
. La deesse de Parmenide, matresse de philosophie, La naissance de la raison
en Gre`ce, ed. Mattei, J. F. (Paris: P.U.F., 1990).
. Linvention de lecole eleatique (Platon, Sophiste, 242d), Etudes sur le
Sophiste de Platon, ed. Aubenque, P. (Naples: Bibliopolis, 1991) 91124.

202

Bibliography

. Les deux chemins de Parmenide, 2nd ed. (Paris/Brussels: Vrin/Ousia, 1997).

. Parmenide platonise, Revue de philosophie ancienne 18(1) (2000) 1524.
Cornford, F. M. Parmenides Two Ways, Classical Quarterly 27 (1933), 97111.
. Platos Theory of Knowledge (London-New York: K. Paul-Harcourt, 1935).
. Parmenides Way of Truth. Plato and Parmenides (London: Routledge &
Kegan Paul, 1939), 3857.
Cosgrove, M. R. The kouros Motif in Parmenides, Phronesis 19 [1974] 8194.
Couloubaritsis, L. Les multiples chemins de Parmenide, Etudes sur Parmenide,
Vol. II, ed. Aubenque, P. (Paris: Vrin, 1987).
. Mythe et philosophie chez Parmenide, 2nd ed. (Brussels: Ousia, 1990).
Coxon, A. H. The Philosophy of Parmenides, Classical Quarterly 28 (1934) 134144.
. The Text of Parmenides Fr. 1.3, Classical Quarterly 18 (1968) 69.
. The Fragments of Parmenides (Assen: Van Gorcum, 1986).
. Deception and Belief in Parmenides Doxa, Apeiron 25 (1992) 109133.
Curtius, G. Grundzuge der griechischen Etymologie (Leipzig: B. G. Teubner, 18581862).
Das, A. C. Negative Fact, Negation and Truth (Calcutta: University of Calcutta, 1942).
Dehon, P. J. Les recommendations de la deesse. Parmenide, fr. 1.2832, Revue de
philosophie ancienne 6(2) (1988).
Deichgraber, K. Parmenides Auffahrt zur Gottin des Rechts (Wiesbaden: Akademie
der Wissenschaften und der Literatur in Mainz, 1959).
Denniston, J. D. The Greek Particles (Oxford: Clarendon Press, 1934).
Destree, P. La communaute de letre (Parmenide, fr. B 5), Revue de philosophie
ancienne 18(1) (2000).
Detienne, M. Les matres de verite dans la Gre`ce archaque (Paris: Maspero, 1967).
. Par la bouche et par loreille, Linvention de la mythologie (Paris: Gallimard,
1981).
Diels, H. Chronologische Untersuchungen uber Apollodors Chronik, Rheinisches
Museum 31 (1876) 154.
. Preface, in Simplicii In Aristotelis Physicorum [. . .] Commentaria (Berlin: G.
Reimer, 1895).
. Parmenides Lehrgedicht (Berlin: G. Reimer, 1897).
Diels, H., and Kranz, W. Die Fragmente der Vorsokratiker, Vol. I (Zurich-Berlin: Weidmannsche Verlagsbuchhandlung, 1956).
Dixsaut, M. Platon et le logos de Parmenide, Etudes sur Parmenide, Vol. II, ed.
Aubenque, P. (Paris: Vrin, 1987).
Dorion, L. A. Euthyphro (Paris: GF-Flammarion, 1997).
Dumont, J. P., et al. Les Presocratiques (Paris: Gallimard, 1988).
Ebeling, H. Lexicon Homericum (Leipzig: B. G. Teubner, 1885).
Ebert, T. Wo beginnt der Weg der Doxa? Phronesis 34 (1989) 121138.
Eggers Lan, C. Los losofos Presocraticos, I (Madrid: Gredos, 1978).
Estienne, H. Poiesis philosophica (Geneva: 1573).
Falcon Martnez, C., Fernandez-Galiano, E., and Lopez Melero, R. Diccionario de
mitologa clasica, Vol. I (Madrid: Alianza, 1980).
Falus, R. Parmenides-interpretationen, Acta Antiqua Academiae Scientiarum Hungaricae 8 (1960), 267294.
Finkelberg, A. Parmenides Foundation of the Way of Truth, Oxford Studies in
Ancient Philosophy 6 (1988) 3967.
. Being, Truth and Opinion in Parmenides, Archiv fur Geschichte der Philosophie 81 [1999], 233248.

Bibliography

203

Finley, M. I. The Ancient Greeks (London: Viking Press, 1963).

Floyd, E. D. Why Parmenides wrote in verse Ancient Philosophy 12 [1992] 251265.
Frankel, H. Parmenidesstudien [1930]. Reprinted in Wege und Formen des fruhgriechischen Denkens (Munich: Beck, 1955).
. Dichtung und Philosophie des fruhen Griechentums (New York: American Philological Association, 1951).
von Fritz, K. Nous, noen and Its Derivatives in Presocratic Philosophy (Excluding
Anaxagoras), I, From the Beginnings to Parmenides, Classical Philology 40
(1945) 223242.
Fronterotta, F. Essere, tempo e pensiero: Parmenide et lorigine dellontologia,
Annali della Scuola Normale Superiore di Pisa, Classe di Lettere e Filosoa, Serie
II, Vol. 24 (1994), 835871.
Fulleborn, G. G. Parmendou tou Eleatou lepsana: Fragmente des Parmenides (Zullichau, 1795).
Furley, D. Truth as What Survives the elenchos (1987), reprinted in Cosmic Problems (Cambridge, England: Cambridge University Press, 1989).
Furth, M. Elements of Eleatic Ontology, Journal of the History of Philosophy 6 (1968)
111132.
Gallop, D. A. Parmenides: Fragments (Toronto: Phoenix, 1984).
Garca Calvo, A. Lecturas Presocraticas (Madrid: Lucina, 1981).
Germani, G. Per uninterpretazione delle vie parmenidee, Annali del Dipartimento
di Filosoa, Universita` di Firenze, Vol. II (1986).
. Annali del Dipartimento di Filosoa, Vol. II (Firenze: Universita` di Firenze,
1986).
. Aletheie in Parmenide, La parola del passato 43 (1988).
Giannantoni, G. Le due vie di Parmenide, La parola del passato 43 (1988).
Gigante, M. Velina gens, La parola del passato 19 (1964).
Gigon, O. Der Ursprung der Griechischen Philosophie von Hesiod bis Parmenides (Basel:
Schwabe & Co., 1945).
Gomez-Lobo, A. Parmenides. Las puertas de la noche y del da, Revista latinoamericana de losofa 3 (1977) 185188.
. Parmenides (Buenos Aires: Editorial Charcas, 1985).
Gomperz, H. Psychologische Beobachtungen an griechischen Philosophen, Imago
10 (1924), 234.
Guazzoni Foa`, V. Attualita` dellontologia eleatica (Turin: Societa` Editrice Internazionale, 1961).
. Il problema delle vie di ricerca in Parmenide (Bergamo: Arti Grache Mariani
& Monti, 1979).
Gunther, H. C. Der Satz des Parmenides von der Identitat von Denken und Sein,
Studi Italiani di lologia classica 15(2) (1997), 125175.
. Aletheia und Doxa, Das Proomium des Gedichtes des Parmenides (Berlin: Dunker
& Humblot, 1998).
Guthrie, W. K. C. A History of Greek Philosophy, Vol. I (Cambridge, England: Cambridge University Press, 1965).
. A History of Greek Philosophy, Vol. II (Cambridge, England: Cambridge University Press, 1965).
Hadot, I. ed. Simplicius, sa vie, son oeuvre, sa survie (Berlin/New York: Walter de
Gruyter, 1987).
. Quest-ce que la philosophie antique? (Paris: Gallimard, 1995).

204

Bibliography

Hegel, G. F. Wissenschaft der Logik, 3rd ed. (Leipzig: Felix Meiner, 1929).
Heidegger, M. Was ist Metaphysik? (Bonn: Cohen, 1929).
. Sein und Zeit, 9th ed. (Tubingen: Max Niemeyer, 1960).
. Parmenides Gesamtausgabe II. Abteilung, Vorlesungen 19231944, Vol. 54 (Frankfurt am Main: Klostermann, 1982).
. Einfuhrung in die Metaphysik (Frankfurt am Main: Max Niemeyer, 1983).
Heitsch, E. Gegenwart und Evidenz bei Parmenides (Wiesbaden: Franz Steiner Verlag,
1970).
. Sein und Gegenwart im fruhgriechischen Denken, Gymnasium 18 (1971)
221237.
. Parmenides: Die Anfange der Ontologie, Logik und Naturwissenschaft (Munich:
Tusculum, 1974).
. Parmenides und die Anfange der Erkenntniskritik und Logik (Donauworth: L.
Auer, 1979).
Hobbes, T. Elementorum philosophiae, Vol. II (London: A. Crook, 1655).
Hoffding, H. La base psychologique des jugements logiques, La revue philosophique 26 (1901) 374).
Hoffmann, E. Die Sprache und die archaische Logik (Tubingen: 1925).
Holscher, U. Grammatisches zu Parmenides, Hermes 84 (1956) 385397.
. Anfangliche Fragen (Gottingen: Vandenhoeck & Ruprecht, 1968).
. Parmenides: Vom Wesen des Seienden (Frankfurt am Main: Suhrkamp Verlag,
1969).
Huxley, G. L. The Early Ionians (New York: Faber, 1966).
Imbraguglia, C. Teoria e mito in Parmenide (Genoa: Studio Editoriale di Cultura,
1979).
Jacoby, F. Apollodoros Chronik, Eine Sammlung der Fragmente (Berlin: Weidmann, 1902).
Jaeger, W. The Theology of Early Greek Philosophers, English translation (1936; Oxford:
Clarendon Press, 1947).
Jantzen, J. Parmenides zum Verhaltnis von Sprache und Wirklichkeit (Munich: C. H.
Beck, 1976).
Kahl-Furthmann, G. Das Problem des Nicht, 2nd ed. (Meisenheim/Glan: A. Haim,
1968).
Kahn, C. H. review of Taran, Gnomon 40 (1968), 123133.
. The Thesis of Parmenides, Review of Metaphysics 22 (1968/69), 700724.
. The Verb Be in Ancient Greek. Part 6 of The Verb Be and Its Synonyms:
Philosophical and Grammatical Studies. Ed. John W. M. Verhaar. Foundations of
Language: Suppl. Series. Vol. 16. (Boston: D. Reidel, 1973).
. The Art and Thought of Heraclitus (Cambridge, England: Cambridge University Press, 1978).
. Being in Parmenides and Plato, La parola del passato 43 (1988) 237261.
Karsten, S. Philosophorum graecorum veterum praesertim qui ante Platonem oruerunt
operum reliquiae, Vol. I, Parte 2: Parmenidis (Amsterdam: J. Muller & Soc., 1835).
Kent Sprague, R. Parmenides: A Suggested Rearrangement of Fragments in the
Way of Truth, Classical Philology 50 (1955) 124126.
Kern, O. Zu Parmenides, Archiv fur Geschichte der Philosophie 3 (1890) 13176.
Ketchum, R. J. Parmenides on What There Is, Canadian Journal of Philosophy 20
(1990) 177190.
Kirk, G. S. Heraclitus: The Cosmic Fragments (Cambridge, England: Cambridge University Press, 1954).

Bibliography

205

Kirk, G. S., Raven, J. E., and Schoeld, M. The Presocratic Philosophers, 2nd ed.
(Cambridge, England: Cambridge University Press, 1983).
Klibansky, R., and Labowsky, L., eds., Plato Latinus, Vol. III: Parmenides nec non
Procli Commentarium in Parmenides (London-Leide: In Aedibus Instituti
Warburgiane-Brill, 1953).
Klowski, J. Die Konstitution der Begriffe Sein und Nichts durch Parmenides,
Kant-Studien 60 (1969) 404416.
. review of Lloyd, Archiv fur Geschichte der Philosophie 51 (1969).
. Parmenides: Grundlegung seiner Seinslehre (B27), Rheinisches Museum
120 (1977), 97137.
ber Aufbau und Bedeutung des Parmenideischen Gedichtes, SitzungKranz, W. U
sberichte der Konigl.-Preussischen Akademie der Wissenschaften 47 (1916).
. Vorsokratisches, Hermes 69 (1934) 114119, 226228.
. Vorsokratische Denker, 3rd ed. (Berlin: Weidmann, 1959).
Krug, W. T. System der theoretischen Philosophie, Vol. I: Denklehre oder Logik (Konigsberg: Goebbels und Unzer, 1806).
Kuhner, R. Ausfurliche Grammatik der griechischen Sprache, 2nd ed., Vol. II(1) (Hannover: Hansche Buchhandlung, 1870).
Lafrance, Y. La theorie platonicienne de la Doxa (Paris/Montreal: BellarminLes
Belles Lettres, 1981).
. Le sujet du Poe`me de Parmenide: Letre ou lunivers? Elenchos 20(2)
(1999), 265308.
Lami, A. I Presocratici (Milan: Rizzoli, 1991).
Lepore, E. Strutture della colonizzazione focea in Occident, La parola del passato
25 (1970) 1954.
Lesher, J. H. The Signicance of kata` panta<s>te in Parmenides Fr. 1.3, Ancient
Philosophy 14 [1994] 120.
Leszl, W. Approccio epistemologico allontologia parmenidea, La parola del passato
43 (1988), 281311.
. Parmenide e lEleatismo. Dispensa per il corso di Storia della Filosoa Antica,
Universita` degli Studi di Pisa, Dipartimento di Filosoa (May 1994).
Levi, A. Sulla dottrina di Parmenide e sulla teoria della doxa, Athenaeum 5 (1927),
269287.
Lloyd, G. E. R. Polarity and Analogy (Cambridge, England: Cambridge University
Press, 1966).
Loenen, J. H. M. M. Parmenides, Melissus, Gorgias: A Reinterpretation of Eleatic Philosophy (Assen: Van Gorcum, 1959).
Loew, E. Ein Beitrag zum heraklitisch-parmenideischen Enkenntnisproblem, Archiv fur Geschichte der Philosophie 31 [1917] 6390, 125152.
. Das Lehrgedicht des Parmenides: Gliederung und Gedankengang, Rheinisches Museum 78 (1929), 148165.
. Das Verhaltnis von Logik und Leben bei Parmenides, Wiener Studien 53
(1935) 137.
Long, A. A. The Principles of Parmenides Cosmogony, Phronesis 8 [1963] 90107.
. Review of Taran, Journal of the Hellenic Studies 86 (1966).
Mabbott, J. D. in Knowledge, Experience and Realism, ed. G. Ryle, The Aristotelian
Society, Vol. Supp. 9 (1929).
Manchester, P. B. Parmenides and the Need of Eternity, The Monist 62 (1979)
81106.

206

Bibliography

Mansfeld, J. Die Offenbarung des Parmenides und die menschliche Welt (Assen: Van
Gorcum, 1964).
. review of Taran, Mnemosyne 20 (1967).
. Aristote et la structure du De sensibus de Theophraste, Phronesis 41 (1996),
158188.
. Parmenide et Heraclite avaient-ils une theorie de la perception? Phronesis
44 (1999).
Manuzio, A. Theocritus (Venice: In Aedibus Aldi, 1496).
Marcovich, M. Eraclito. Frammenti (Florence: La Nuova Italia, 1978).
Marsoner, A. La struttura del Proemio di Parmenide, Annali dellIstituto Italiano
per gli Studi Storici 5 (19761978).
Martineau, M. Le coeur de lAletheia, Revue de philosophie ancienne 4 (1986)
3386.
Martinelli, F. Fra Omero e Pindaro: Parmenide poeta, Forme del sapere nei Presocratici, ed. Capizzi, A., and Casertano, G. (Athens/Rome: Edizione dellAteneo,
1987) 169186.
Meijer, P. A. Das methodologische im 5. Fragment des Parmenides, Classica at
medievalia 30 [1969] 102108.
. Parmenides Beyond the Gates: The Divine Revelation on Being, Thinking and the
Doxa (Amsterdam: J. C. Gieben, 1997).
Merlan, P. Neues Licht auf Parmenides, Archiv fur Geschichte der Philosophie 48
(1966) 267276.
Misch, G. Der Weg in die Philosophie (Berlin/Leipzig: 1926).
de Moerbeke, G. trans. Proclus, Commentaire sur le Parmenide de Platon, critical
edition of Steel, C., Vol. II (Louvain: Presses Universitaires de Louvain, 1985).
Mondolfo, R. Discusioni su un testo parmenideo (fr. 8.56), Rivista critica di storia
della losoa 19 (1964), 310315.
Montaner, A. La struttura del Proemio di Parmenide, Annali dellIstituto Italiano
per gli Studi Filosoci V (19761978).
Montero Moliner, F. Parmenides (Madrid: Gredos, 1960).
Mookerjee, S. The Buddhist Philosophy of Universal Flux (Calcutta: University of Calcutta, 1935).
Moravcsik, J. M. E. Being and Meaning in the Sophist, Acta philosophica fennica 14
(1962).
Morel, J. P. Les Phoceens en Occident: certitudes et hypothe`ses, La parola del passato 21 (1966) 379420.
. Sondages sur lacropole de Velia (Contribution aux premiers temps de la
cite), La parola del passato 25 (1970) 131145.
Morot-Sir, E. La pensee negative (Paris: Aubier, 1947).
Morrow, G. R., and Dillon, J. M., eds. Proclus Commentary on Platos Parmenides
(Princeton, NJ: Princeton University Press, 1987).
Mosse, C. La Gre`ce archaque dHome`re a` Eschyle (Paris: Seuil, 1997).
Mourelatos, A. P. D. Phrazo and Its Derivatives in Parmenides, Classical Philology
60 (1965), 261262.
. The Route of Parmenides (Yale, CT: Yale University Press, 1970).
. Heraclitus, Parmenides, and the Naive Metaphysics of Things, Phronesis,
Supp. I, Exegesis and Argument, Studies in Greek Philosophy presented to G.
Vlastos, ed. Lee, E. N., Mourelatos, A. P. D., and Rorty, R. M. (Assen: 1973)
1648.

Bibliography

207

. Some Alternatives in Interpreting Parmenides, The Monist 62 (1979) 317.

Mullach, F. G. A. Aristotelis, De Melisso, Xenophane et Gorgia disputationes, cum Eleaticorum Philosophorum fragmentis (Berlin: 1845).
. Fragmenta philosophorum graecorum (Paris: Didot, 1860).
Muller, M. Nouvelles lecons sur la science du langage (Paris: 1868).
berlieferung der Schriften des Sextus Empiricus, RheiMutschmann, H. Die U
nisches Museum 69 (1909).
Napoli, M. La ricerca archeologica di Velia, La parola del passato 21 (1966) 191226.
. Civilita` della Magna Grecia (Rome: Eurodes, 1978).
Nehamas, A. On Parmenides Three Ways of Inquiry, Deucalion 3334 (1981) 197
211.
Nutton, V. The Medical School of Velia, La parola del passato 25 (1970) 211225.
OBrien, D. Temps et intemporalite chez Parmenide, Les etudes philosophiques 3
(1980) 257272.
. Etudes sur Parmenide, Vol. I, ed. Aubenque, P. (Paris: Vrin, 1987).
` propos de la syntaxe de petho et de pisteuo, Revue des etudes grecques
Oguse, A. A
78 (1965).
Owen, G. E. L. Eleatic Questions, Classical Quarterly 10 (1960), 84102.
Palmer, J. H. Platos Reception of Parmenides (Oxford: Oxford University Press, 1999).
Pasquinelli, A. I Presocratici (Turin: Einaudi, 1958).
Patin, A. Parmenides im Kampfe gegen Heraklit, Jahrbuch fur klassische Philologie,
Supplementband 25 (1899), 489660.
Pellikaan-Engel, M. E. Hesiod and Parmenides: A New View on Their Cosmologies and
on Parmenides Poem (Amsterdam: Adolf M. Hakkert, 1974).
Peyron, A. Empedoclis et Parmenidis fragmenta ex codice Taurinensis Bibliothecae restituta et illustrata (Leipzig: I. A. G. Weigel, 1810).
Pfeiffer, R. Gottheit und Individuum, Ausgewahlte Schriften (Munich: Beck, 1960).
Phillips, E. D. Parmenides on Thought and Being, Philosophical Review 64 (1955)
547559.
Pieri, A. Parmenide e la lingua della tradizione epica greca, Studi Italiani di lologia classica 49 (1977).
Popper, K. The World of Parmenides (London/New York: Routledge, 1998).
Pugliese Carratelli, G. Pholarkhos, La parola del passato 18 (1963).
. Parmendes phusikos, La parola del passato 20 (1965).
. Ancora su pholarkhos, La parola del passato 25 (1970).
. Nascita di Velia, La parola del passato 25 (1970).
Ralfs, G. Der Satz von Widerspruch bei Parmenides, Lebensformen des Geistes, ed.
Glockner, H. Kant-Studien, Erganzungsheft 86, 1964.
Ranulf, S. Der eleatische Satz vom Widerspruch (Copenhagen: Gyldendal, 1924).
Raven, J. E. Pythagoreans and Eleatics (Cambridge, England: Cambridge University
Press, 1948).
Reale, G. in Zeller, E., and Mondolfo, R., La losoa dei Greci, Part I, Vol. 3, (Florence:
La Nuova Italia, 1967).
. Melissus: Testimonianze e frammenti (Florence: La Nuova Italia, 1970).
, trans. Parmenides, Poema sulla natura (Milan: Rusconi, 1991).
Redard, G. Recherches sur khre, khresthai (Pars: H. Champion, 1953).
Regvald, R. Parmenide: Le trajet de la non-concidence, La revue philosophique 176
(1986), 1329.
Reich, K. Parmenides und die Pithagoreer, Hermes 82 (1954) 287294.

208

Bibliography

Reinhardt, K. Parmenides und die Geschichte der griechischen Philosophie (Frankfurt:

Vittorio Klostermann, 1916).
Riaux, F. M. Essai sur Parmenide dElee (Paris: Librairie de Joubert, 1840).
Riezler, K. Parmenides (Frankfurt am Main: Klosterman, 1934).
de Rijk, L. M. Did Parmenides Reject the Sensible World? Graceful Reason, ed.
Gerson, L. P. (Toronto: Pontical Institute of Mediaeval Studies, 1983), 2953.
Rinieri, J. J. Le Poe`me de Parmenide (Paris: Presses Universitaires de France, 1955).
Robin, L. La pensee grecque et les origines de lesprit scientique, 3rd ed. (Paris: La
Renaissance du Livre, 1963).
Robinson, T. M. Parmenides on the Ascertainment of Real, Canadian Journal of
Philosophy 4 (1975) 56.
. Parmenides on the Real in Its Totality, The Monist 62 (1979) 5460.
Ruggiu, L. Parmenide (Venice: Marsilio, 1975).
ber Parmenides, Verhandl. d. 52 Vers. deutsch. Philol. u. Schulm. zu
Rustow, A. U
Marburg (1913).
Ryle, G., ed. Knowledge, Experience and Realism The Aristotelian Society, Vol. Supp.
9 (1929).
Sartre, J. P. Letre et el Neant (Paris: Gallimard, 1943).
Schwabl, H. Sein und Doxa bei Parmenides, Wiener Studien 66 (1953), reprinted
in Um die Begriffswelt der Vorsokratiker, ed. Gadamer, H. G. (Darmstadt: Wissenschaftliche Buchgesellschaft, 1968).
. Hesiod und Parmenides, Rheinisches Museum 106 (1963) 134142.
Schwyzer, E. Griechische Grammatik, Vol. II (Munich: C. H. Beck, 1950).
Seligman, P. Being and Not-being: An introduction to Platos Sophist (The Hague: Martinus Nijhoff, 1978).
Sider, D. Textual Notes on Parmenides Poem, Hermes 113 (1985) 362366.
Sigwart, C. Logik, 5th ed. (Tubingen: Mohr, 1924).
Snell, B. Aischylos und das Handeln in Drama, Philologus, Supp. 20, I (1928).
Solmsen, F. The Tradition About Zeno of Elea Re-examined, Phronesis 16 (1971)
116141.
Solovine, M. Democrite (Paris: Alcan, 1928).
Somigliana, A. I versi 4 ss. del fr. B 6 di Parmenide e la supposta polemica antieraclitea, Rivista di losoa neoscolastica 67 (1975).
Somville, P. Parmenide dElee: Son temps et le notre (Paris: Vrin, 1976).
Stahl, J. M. Kritisch-historischer Syntax des griechischen Verbums der klassischen Zeit
(Heidelberg: C. Winter, 1907).
Stein, H. Die Fragmente des Parmenides Per Phuseos, Symbola philologrum Bonnensium in honorem F. Ritschelii (Leipzig: Teubner, 18641867).
Stevens, A. Posterite de letre: Simplicius interpre`te de Paemenide (Brussels: Ousia,
1990).
Taran, L. El signicado de noen en Parmenides, Anales de lologa clasica 7 (1959)
122139.
. Parmenides (Princeton, NJ: Princeton University Press, 1965).
. Perpetual Duration and Temporal Eternity in Parmenides and Plato, The
Monist 62 (1979) 4353.
ber die Negation, den Widerspruch und den Gegensatz, dissertation, (Berlin,
Thiede, J. U
1883).
Trabattoni, F., trans. Parmenides, I frammenti. (Milan: Marcos y Marcos, 1985).

Bibliography

209

Trendelenburg, F. A. Logische Untersuchungen, Vol. I, 3rd ed. (Leipzig: S. Hirzel,

1870).
Tugwell, S. The Way of Truth, Classical Quarterly 14 (1964), 3641.
Tzavaras, G. To` Poema tou Parmende (Athens: Domos, 1980).
berweg, F. Grundriss der Geschichte der Philosophie, Vol. I, 12th ed. (Basel: Schwabe,
U
1953).
Untersteiner, M. Parmenide. Testimonianze e frammenti (Florence: La Nuova Italia,
1958).
. Problemi di lologia losoca, ed. Sichirollo, L., and Venturi Ferriolo, M. (Milan: Cislapino-Goliardica, 1980).
. Les Sophistes Vol. I, (Paris: Vrin, 1993).
Vallet, V., and Villard, F. Les Phoceens en Mediterranee Occidentale a` lepoque
archaque et la fondation de Hyele`, La parola del passato 21 (1966) 166190.
Verdenius, W. J. Parmenides: Some Comments on His Poem (Groningen: Hakkert,
1942).
. Parmenides B 2.3, Mnemosyne 15 (1962) 237.
. Der Logosbegriff bei Heraklit und Parmenides, II, Phronesis 12 (1967) 99
117.
Vernant, J. P. Hestia-Herme`s, Mythe et pensee chez les Grecs, Vol. 1 (Paris: Maspero,
1971).
. Raisons du mythe, Mythe et societe en Gre`ce ancienne (Paris: Maspero,
1974), 195250.
Vitali, R. Parmenide di Elea: Per` phuseos (Faenza: Lega, 1977).
Vlastos, G. Parmenides Theory of Knowledge, Transactions and Proceedings of the
Amer. Philol. Assoc. 77 (1946) 6677.
. Equality and Justice in Early Greek Cosmologies, Classical Philology 42
(1947).
de Vogel, C. Platon a-t-il ou na-t-il pas introduit le mouvement dans son monde
intelligible? Actes du XIe Congre`s International de Philosophie XII, Brussels (1953)
6167.
Vuia, O. Remontee aux sources de la pensee occidentale: Heraclite, Parmenide, Anaxagore
(Paris: Centre Roumain de Recherches, 1961).
Wackernagel, J. Vermischte Beitrage zur griechischen Sprachkunde (Basel: 1897).
Wahl, J. Vers la n de lontologie (Paris: Societe de lEnseignement Superieur, 1956).
Wiersma, W. Notes on Greek Philosophy (Parm. 1.17, 2.4, 8.61), Mnemosyne 20
(1967).
Wiesner, J. Die Negation der Entstehung des Seienden: Studien zu Parmenides B
8.521, Archiv fur Geschichte der Philosophie 52 (1970) 125146.
berlegungen zu Parmenides, fr. VIII, 34, Etudes sur Parmenide, Vol. II,
. U
ed. Aubenque, P. (Paris: Vrin, 1987), 170191.
. Parmenides: Der Beginn der Aletheia (Berlin/New York: Walter de Gruyter,
1996).
Wilamowitz, U. v. Lesefruchte, Hermes 34 [1899] 203230.
Wittgenstein, L. Tractatus (Frankfurt/Main: Suhrkamp, 1989).
Woodbury, L. Parmenides on Names, Harvard Studies in Classical Philology 63
(1958), 145160.
. Apollodorus, Xenophanes and the foundation of Massilia, The Phoenix 15
(1961) 134155.

210

Bibliography

Wyatt, W. F. The Root of Parmenides, Harvard Studies in Classical Philology 94

(1992), 113120.
Zeller, E. Die Philosophie der Griechen in ihrer geschichtlichen Entwicklung (Leipzig:
G. R. Reisland, 1892).
Zucchi, H. Estudios de losofa antigua y moderna (Tucuman: Universidad Nacional
de Tucuman, Instituto de Filosoa, 1956).

List of Ancient Authors Cited

Aeschylus 12, 128

Aetius 159, 160
Alexander of Aphrodisias 147
Ameinias 9, 159
Anaxagoras 11, 94
Anaximander 20, 30, 159
Anaximenes 20, 159
Antiphon 8, 12
Antisthenes ix, 151, 177, 178, 179
Apollodorus 5, 6, 7, 8
Aristophanes 122
Aristotle ix, 9, 10, 11, 19, 66, 71, 105,
117, 118, 122, 130, 143, 144, 159,
162, 167, 176, 182, 183
Asclepius 175
Athenaeus 8
Bacchylides 129
Bias 181
Cicero 160
Clement 32, 161, 183
Crates 122
Critias 122, 127
Democritus 94
Demosthenes 122, 156
Diogenes Of Apollonia 94
Diogenes Laertius 4, 5, 6, 7, 8, 9, 10, 11,
130, 147, 182
Dionysius Of Thrace 121
Empedocles 12, 15, 41, 128, 129, 183
Epiphanius 182
Euclid 32
Euripides 26, 35
Gorgias 12, 46, 77, 87, 104, 132, 152, 179
Hecateus 181
Heraclitus 7, 8, 11, 16, 30, 33, 50, 68,
123, 132, 139, 144, 160, 178, 181

Herodotus 3, 4, 34, 35, 121

Hesiod 14, 15, 21, 24, 28, 29, 33, 54, 61,
101, 104, 127, 128, 136, 137, 140,
151
Hesychius 29
Hippocrates 35, 131
Homer 11, 13, 14, 15, 24, 25, 28, 29, 32,
33, 37, 38, 54, 57, 61, 62, 79, 85, 101,
121, 123, 128, 134, 136, 156, 179
Isocrates 156, 182
Macrobius 8
Melissus 10, 11, 46, 94, 168, 170, 173,
177, 178, 182, 183
Mimnermos 29
Pindar 12, 25, 157
Plato ix, x, 5, 6, 7, 8, 9, 10, 11, 12, 13, 20,
21, 22, 24, 25, 29, 30, 31, 33, 35, 37,
44, 50, 70, 74, 80, 82, 89, 109, 110,
117, 118, 121, 123, 127, 131, 133,
134, 135, 140, 142, 144, 154, 158,
159, 160, 161, 163, 172, 173, 174,
175, 176, 177, 178, 181, 182, 183, 184
Plutarch ix, 9, 13, 29, 31, 161, 183
Proclus x, 5, 10, 14, 32, 37, 46, 84, 88,
94, 152, 161, 178, 182, 188
Protagoras 44, 122
Pseudo-Alexander 117
Pseudo-Galen 182
Pythagoras 4, 9, 27, 181
Sextus Empiricus ix, x, 13, 16, 21, 154,
161, 185
Simplicius ix, x, 12, 13, 31, 35, 37, 46,
63, 64, 70, 84, 85, 88, 90, 91, 94, 98,
99, 110, 111, 112, 113, 114, 115, 117,
118, 122, 125, 126, 131, 142, 161,
165, 166, 168, 175, 183, 187, 188
Socrates 6, 7, 25, 123, 135
Solon 8, 29

212
Sophocles 26, 35
Sophron 122
Sotion 9
Stobaeus 161
Strabo 11, 182
Thales 20
Theophrastus 14, 19, 162, 183

List of Ancient Authors Cited

Thucydides 35, 156
Timon 8
Xenophanes 9, 10, 11, 14, 121, 129, 130,
160, 181
Xenophon 35, 94, 122
Zeno Of Elea ix, 6, 181, 182

List of Modern Authors Cited

Aalto, P. 40
Albertelli, P. 94, 168
Allen, R. E. 8
Arrighetti, G. 15, 24
dAsola, F. 113, 114
Aubenque, P. 71, 129
Austin, S. 7156
Baldwin, B. 12
Ballew, L. 127, 146
Barnes, J. 65, 69, 176, 178
Basson, A. H. 65, 66
Baumann, A. 76
Beaufret, J. and Rinieri, J. J. 56, 133
Becker, O. 28, 46, 80, 110, 111, 112, 129
Benveniste, E. 21, 53, 59, 60
Beraud, J. 3, 4
Bergson, H. 75, 77
Bernabe, A. 33, 94
Berti, E. 149
Bessarion, I. 113
Bicknell, P. J. 7, 16, 110
Boardman, J. 3
Bohme, R. 14
Bollack, J. 48, 92, 160
Bormann, K. 12, 25, 26
Brague, R. 34, 35
Brandis, C. A. 91, 114
Brocker, W. 81
Brucker, I. 147
Brugmann, K. 51, 53
Brumbaugh, R. S. 111
Burkert, W. 27, 29
Burnet, J. 48, 109
Buroni, G. 69
Buxton, R. G. A. 104
Calogero, G. 4, 126
Casertano, G. 24, 48
Cassin, B. 24, 31, 34, 39, 41, 52, 62, 63,
85, 92, 162, 173
Cassin, B., and Narcy, M. 162

de Cecco, D. 149
Cerri, G. 10, 27, 41
Chalmers, W. R. 106
Chantraine, P. 41, 61
Ciaceri, E. 4
Collobert, C. 24, 38
Colombo, A. 64
Conche, M. 24, 31, 39, 40, 41, 85, 99,
127
Constantineau, P. 40, 56
Corbato, C. 10
Cordero, N. L. xi, 10, 11, 12, 15, 26, 27,
28, 30, 31, 61, 81, 83, 90, 91, 92, 112,
113, 114, 134, 148, 173, 181, 182,
183, 185
Cornford, F. M. 12, 46, 56, 77, 91, 100,
106, 133
Cosgrove, M. R. 24
Couloubaritsis, L. 38, 39, 42, 138, 139,
148, 173
Coxon, A. H. 24, 25, 27, 31
Curd, P. 152
Curtius, G. 61
Das, A. C. 76
Dehon, P. J. 32, 34, 35
Deichgraber, K. 29
Denniston, J. D. 102, 123, 124, 142
Destree, P. 128, 157
Detienne, M. 22, 24, 29, 32, 104
Diels, H. 6, 10, 13, 16, 34, 41, 43, 46, 50,
81, 84, 85, 86, 87, 91, 94, 99, 101,
102, 108, 109, 110, 111, 112, 113,
114, 115, 116, 118, 119, 120, 121,
122, 126, 142, 144, 156, 160, 182,
185
Diels, H., and Kranz, W. 16, 50, 91, 94,
160
Dillon, J. 84
Dixsaut, M. 135, 175
Dorion, L. A. 122
Dumont, J. P. 27

214
Ebeling, H. 54
Ebert, T. 162
Eggers Lan, C. 33
Estienne, H. 13, 117
Falcon Martnez, C., FernandezGaliano, E., and Lopez Melero, R.
26
Falus, R. 34, 47, 57, 60, 69
Finkelberg, A. 45, 139
Finley, M. I. 3
Floyd, E. D. 14
Frankel, H. 49, 50, 51, 52, 54, 78, 106
von Fritz, K. 85, 86, 137
Fronterotta, F. 122, 125, 126, 142, 171
Fulleborn, G. G. 109, 114, 151
Furley, D. 134, 135, 166
Furth, M. 65
Gallop, D. 56, 66, 153
Garca Calvo, A. 33, 116
Germani, G. 133, 148, 157, 160
Giannantoni, G. 91, 118, 149
Gigante, M. 9
Gigon, O. 54
Gomez-Lobo, A. 8, 22, 24, 25, 28, 35, 39,
41, 49, 85, 106, 118
Gomperz, H. 7, 41, 81
Goulet, R. 9
Guazzoni Foa`, V. 69, 145
Gunther, H. C. 27, 42
Guthrie, W. K. C. 33, 49, 56, 68, 77, 106,
136
Hadot, I. 12
Hadot, P. 11
Hegel, G. F. 74
Heidegger, M. 67, 71, 86, 108, 11, 142
Hobbes, T. 75
Hoffding, H. 75
Hoffmann, E. 86
Holscher, U. 53, 56, 66, 81, 100, 106, 129
Huxley, G. L. 3, 4
Imbraguglia, C. 79
Jacoby, F. 6
Jaeger, W. 14, 15, 106
Jantzen, J. 41, 67, 76, 133

List of Modern Authors Cited

Kahl-Furthmann, G. 78
Kahn, C. H. 27, 33, 40, 48, 49, 52, 53, 61,
62, 65, 71, 76, 91, 139
Karsten, S. 37, 91, 99, 109, 114, 115, 123,
132
Kent Sprague, R. 16, 105, 107, 109, 110
Kern, O. 28
Ketchum, R. 81
Kirk, G. S. 49, 123
Kirk, G. S., Raven, J. E., and Schoeld,
M. 56, 94
Klibansky, R., and Labowsky, L. 84
Klowski, J. 71, 81, 99
Kranz, W. 16, 25, 34, 50, 81, 94, 109, 160
Krug, W. T. 75
Kuhner, R. 52
Lafrance, Y. 31, 48
Lami, A. 33
Lepore, E. 4
Lesher, J. H. 27
Leszl, W. 86, 125, 135, 146, 154
Levi, A. 95
Liddell, H. G., Scott, R., and Jones,
H. S. 34, 88, 126
Lloyd, G. E. R. 71, 149
Loenen, J. H. M. M. 46, 80, 144
Loew, E. 100, 130, 144, 145
Long, A. A. 101, 111, 152
Mabbott, J. D. 75
Manchester, P. B. 66
Mansfeld, J. 40, 45, 46, 47, 48, 57, 66, 86,
93, 100, 127, 128, 129, 130
Manuzio, A. 113, 114
Marcovich, M. 33
Marsoner, A. 23, 24, 26
Martineau, M. 31
Martinelli, F. 35
Meijer, P. A. 46, 87, 91, 92, 106, 118, 123
Merlan, P. 25
Misch, G. 23
Mondolfo, R. 47, 80, 172
Montaner, A. 15
Montero Moliner, F. 45
Mookerjee, S. 75
Moravcsik, J. 119
Morot-Sir, E. 69, 72
Mosse, C. 3

List of Modern Authors Cited

Mourelatos, A. P. D. 14, 23, 33, 40, 41,
50, 65, 66, 68, 71, 88, 93, 94, 104,
109, 111, 124, 134, 137, 144
Mullach, F. G. A. 43, 91, 115
Muller, M. 61
Mutschmann, H. 27
Napoli, M. 4
Nehamas, A. 121, 148
Nutton, V. 9
OBrien, D. 63, 148
Owen, G. E. L. 47, 48, 49
Palmer, J. H. 13
Pasquinelli, A. 144
Patin, A. 115, 129
Pellikaan-Engel, M. E. 29
Peyron, A. 43
Pfeiffer, H. 127
Phillips, E. D. 86
Pieri, A. 29, 37
Popper, K. 9, 48
Pugliese Carratelli, G. 4, 9
Ralfs, G. 70
Ranulf, S. 100, 119
Raven, J. E. 9
Reale, G. 27, 172, 173, 175, 177
Redard, G. 54
Regvald, R. 60
Reich, K. 118
Reinhardt, K. 16, 25, 47, 118, 126, 133,
139, 142, 159, 176
Riaux, F. 24, 46, 47, 91, 99, 109, 115
Riezler, K. 100
de Rijk, L. M. 33, 140
Robin, L. 47, 144, 145, 146
Robinson, T. M. 49, 67
Ruggiu, L. 22, 54, 81, 119, 170
Rustow, A. 158
Ryle, G. 74
Sartre, J. P. 74
Scaliger, J. J. 13, 14, 114, 116
Schuhl, P. M. xi, 148
Schwabl, H. 29, 33, 80, 151
Schwyzer, E. 52, 61
Seligman, P. 143
Sider, D. 126

215
Sigwart, C. 69, 75
Snell, B. 128
Solmsen, F. 182
Solovine, M. 94
Somigliana, A. 131
Somville, P. 30
Stahl, J. M. 34
Steel, C. 84
Stein, H. 43, 91, 115, 139
Stevens, A. 85, 88, 166
Taran, L. 13, 22, 23, 24, 25, 33, 34, 39,
41, 42, 45, 46, 51, 63, 70, 83, 85, 87,
91, 93, 99, 105, 106, 107, 108, 110,
111, 112, 123, 124, 134, 139, 144,
160, 170, 171, 176
Tardieu, M. 12
Thiede, J. 75
Trabattoni, F. 27
Trendelenburg, F. A. 75
Tugwell, S. 48
Tzavaras, G. 27
berweg, F. 106
U
Untersteiner, M. 4, 6, 8, 24, 34, 41, 43,
44, 48, 66, 67, 69, 93, 94, 99, 104,
105, 106, 118, 129, 142, 147, 149, 175
Vallet, G., and Villard, F. 4
Verdenius, W. J. 34, 49, 50, 66, 76, 80,
87, 93, 129, 134, 136, 137, 144
Vernant, J. P. 39, 104
Vitali, R. 37, 38, 98, 105, 107, 115, 116, 166
Vlastos, G. 27, 50, 128, 129
de Vogel, C. 174
Vuia, O. 86
Wackernagel, J. 54
Wahl, J. 145
Wiersma, W. 33, 49
Wiesner, J. 41, 42, 45, 48, 66, 71, 79, 83,
85, 86, 88, 89, 92, 95, 99, 102, 106,
118, 127, 132, 141, 170
Wilamowitz, U. V. 34
Wittgenstein, L. 75
Woodbury, L. 10, 48, 88
Zeller, E. 16, 172
Zucchi, H. 64

This page has been intentionally left blank.