Pythagoras and the

Early Pythagoreans

L EO N ID Z H M U D
Translated from Russian by Kevin Windle and Rosh Ireland

1
 3
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
# Oxford University Press 2012
First published in Russian 1994 # Leonid Zhmud
In German 1997 # Akademie Verlag
The moral rights of the author have been asserted
First Edition published in 2012
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Crown copyright material is reproduced under Class Licence
Number C01P0000148 with the permission of OPSI
and the Queen’s Printer for Scotland
British Library Cataloguing in Publication Data
Data available
Library of Congress Cataloguing in Publication Data
Library of Congress Control Number : 2012930974

ISBN 978–0–19–928931–8

Printed in Great Britain

on acid-free paper by
MPG Books Group, Bodmin and King’s Lynn
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
 Contents

Abbreviations xi

Introduction: The Pythagorean Question:

Problems, Methods, and Sources 1
1. The Early Tradition on Pythagoras and its Development 25
1.1 Features of the Pre-Platonic tradition 25
1.2 Evidence 30
1.3 Pythagoras before and after Plato 50
2. Biography: Sources, Facts, and Legends 61
2.1 Main sources 61
2.2 Life on Samos 78
2.3 Pythagoras’ journey 83
2.4 Magna Graecia 92
3. Who Were the Pythagoreans? 105
3.1 The Pythagoreans after Pythagoras 105
3.2 The catalogue of Aristoxenus 109
3.3 The prosopography and chronology of the Pythagoreans 119
4. The Pythagorean Communities 135
4.1 What kind of community did Pythagoras found? 135
4.2 School, thiasos, hetairia 141
4.3 A Pythagorean ‘sect’? 148
5. Mathematici and Acusmatici. The Pythagorean ‘Symbols’ 169
5.1 Two traditions 169
5.2 In search of the acusmatici 175
5.3 Formation of a legend 183
5.4 The ‘symbols’ 192
6. Shamanism and Metempsychosis 207
6.1 Shamanism in ancient Greece? 207
6.2 Metempsychosis: Pythagoreanism and Orphism 221
x Contents
7. Mathematics 239
7.1 Greek mathematics and the Orient 239
7.2 Deductive proof 251
7.3 Pythagoras’ mathematics in fourth-century testimonies 255
7.4 Pythagoras as a mathematician 270
7.5 Hippasus and Pythagorean mathematics of the first
half of the fifth century 274
8. Harmonics and Acoustics 285
8.1 Pythagoras and the science of music 285
8.2 The theory of harmonic intervals 293
8.3 Acoustics: theory and experiments 303
9. Astronomy 315
9.1 Greek astronomy and the Orient 315
9.2 Pythagorean astronomy before Philolaus? 322
9.3 ‘The harmony of the spheres’ 337
10. Medicine and Life Sciences 347
10.1 Medicine and Pythagoreanism 347
10.2 Physiology and anatomy 365
10.3 Embryology 374
10.4 Botany 380
11. Pythagorean Philosophies 387
11.1 Theories of the soul 387
11.2 All is number? 394
12. Pythagorean Number Doctrine in the Academy and
Lyceum 415
12.1 Pythagoreanism and the Academy 415
12.2 Aristotle and the —ıŁÆªæØØ 433
12.3 Number doctrine at the Lyceum? 452

Bibliography 457
Index of Names 473
Index of Passages 480
 Introduction

The Pythagorean Question: Problems,

Methods, and Sources

The modern history of the study of Pythagoreanism, which began

with August Böckh’s book on Philolaus,1 now reaches back almost
two centuries. In that time hundreds if not thousands of books and
articles have been written about Pythagoras and the Pythagoreans,2
yet the body of facts on which all scholars would agree is far from
large, and widely differing and often mutually exclusive interpreta-
tions are legion. The Pythagorean question remains one of the most
intricate in the history of early Greek science, philosophy and religion
and has every chance of being consigned to the category of insoluble
problems.
This is not because every generation takes a new view of the
personality and teaching of Pythagoras: this is common to all Greek
thinkers who, like Socrates, Plato and Aristotle, retain their intellec-
tual appeal in the modern world. It is not even because of the
existence of almost insurmountable differences of opinion within
each generation of scholars, all divided into their own disciplines
(philology, the histories of science, of medicine, philosophy, religion,
etc.) and national schools of thought, each resting primarily on its
own tradition. The problem as I see it lies in the fact that within the
scholarly community no consensus has yet been reached on the most

1
A. Böckh, Philolaos des Pythagoreers Lehren nebst den Bruchstücken seines
Werkes (Berlin, 1819).
2
See the general and specialized bibliographies: L. Paquet et al., Les Présocratiques:
Bibliographie analytique (1879–1980), 2 vols. (Paris, 1988–9); L. E. Navia, Pythagoras:
An Annotated Bibliography (New York, 1990); id., The Presocratic Philosophers: An
Annotated Bibliography (New York, 1993); B. Šijakovič, Bibliographia Praesocratica
(Paris, 2001).
2 Pythagoras and the Early Pythagoreans
fundamental facts and the separation on the basis of these of soluble
problems from fundamentally insoluble. Although no ‘definitive’
interpretation of Plato’s philosophy is possible, there is general agree-
ment that Plato was a pupil of Socrates, the teacher of Aristotle, and
the author of philosophical dialogues. But as to whether Pythagoras
was taught by Egyptian priests or by Pherecydes of Syros, whether he
studied philosophy and science, whether there were any texts that he
himself wrote, whether his students included mathematicians and
philosophers – all this remains the subject of debate.
To hope for a solution to the Pythagorean question in these
circumstances would be an unforgivable illusion. Having spent
many years in the study of Pythagoreanism, I have no such illusions.
If I am returning to this problem it is only because I remain convinced
that, like any other complex scientific problem, it can be broken down
into a number of smaller, particular ones, which may prove amenable
to solution. There are many facts on which agreement may be
reached; there is also an undoubted hierarchy of interpretations,
ranging from those admittedly impossible or unverifiable to those
which are more probable and internally consistent. The fact that the
situation is not hopeless, given – naturally – a willingness to accept
the facts and take account of the achievements and errors of our
predecessors, is indicated by the resolution of one particular question
of great importance, that of the authenticity of the fragments of the
Pythagorean Philolaus (c.470–c.400).3
It is hardly a matter of chance that Böckh was the one who posed
this question, given that he understood classical philology as Alter-
tumswissenschaft, whose task was to investigate the ancient world
in its entirety. Pythagoreanism sets before us precisely the kind of
problem in which politics and religion, philosophy and science are
closely intertwined. Before all else it was necessary to demarcate
the range of most reliable sources, and Böckh’s choice proved to
be completely correct: the fragments of Philolaus are ‘ein lichter
Punkt’ in the ‘labyrinthischen Gewirre der Überlieferungen über die
Pythagoreische Weisheit und Pythagoreische Gesellschaft, welche
großentheils durch späte und urtheilslose Schriftsteller . . . zu uns her-
übergekommen sind’.4 If these fragments are spurious it is extremely
difficult to assert that a scientific and philosophical school existed

3
This and all other ancient dates are BC unless otherwise indicated.
4
Böckh, Philolaos, 3.
 The Pythagorean Question: Problems, Methods, and Sources 3
within the framework of early Pythagoreanism, that is, from Pytha-
goras to Philolaus. The question posed but not resolved by Böckh
was discussed for almost a century and a half, during which time
some scholars leaned towards accepting the authenticity of all the
fragments of Philolaus and others doggedly rejected them. A solution
was found by Walter Burkert in his epoch-making book. Burkert was
the first to divide the fragments of Philolaus into two unequal parts
and demonstrate convincingly that the smaller part (B 1–7, 13, 16–17)
was authentic, while the others bore the stamp of later ideas and
terminology that was alien to the Presocratics.5
Burkert, who did much to revolutionize the study of Pythagorean-
ism, demonstrated yet again that success in this field, even partial
success, can be achieved only by examining the sources thoroughly
and sifting out those which can provide a basis for our reconstruc-
tions. Most of the information from the classical period about Pytha-
goras and early Pythagoreanism has been preserved in the work of
later writers; separating it from the accretions of the Hellenistic and
Imperial eras is extraordinarily difficult. The results of nineteenth-
century research in this field were summed up by Eduard Zeller, who
noted that the further we get from the time of Pythagoras, the greater
the quantity of sources, while the reliability of these declines.6 Zeller’s
attempt to rely on sources from the fifth and fourth centuries and that
part of the later tradition which concurs with them on important
details appears fully justified to this day. Burkert made important
corrections to Zeller’s approach, showing that it is a necessary but not
a sufficient condition, since the early sources are also problematic and
contradictory. In part this is due to their fragmentary nature, but the
main difficulty is that they not only relate to various aspects of ancient
Pythagoreanism in its almost 200-year history – from the rise of the
Pythagorean community in Croton in c.530 to the disappearance of
the school after 350 – but they also give varying interpretations of
those aspects. Burkert distinguished two basic lines of interpretation,
the Platonic and the Aristotelian, giving clear preference to the latter.7
One of the aims of the present study is to show that the Aristotelian
interpretation of Pythagoreanism, which most scholars are inclined to

5
Burkert, 218 ff. One of his predecessors was R. Mondolfo, ‘Sui frammenti di
Filolao’, RFIC 15 (1937), 225–45. See also Zeller i. 371 n. 3.
6
Zeller i. 364. For a review of sources, see Zeller and Mondolfo i. 313–85.
7
Burkert, 53 ff., esp. 79 f.
4 Pythagoras and the Early Pythagoreans
accept, does not really deserve credit, whereas the very existence of
the Platonic interpretation, as defined by Burkert, is open to question.
Burkert’s conclusion on the authenticity of some of Philolaus’
fragments was accepted surprisingly quickly by the great majority of
students of Presocratic philosophy. However, as usually happens in
science, the solution of one particular problem immediately gives rise
to new ones. Burkert ‘rescued’ some of the fragments of Philolaus, but
not Philolaus as a philosopher and scientist himself. Burkert regarded
Philolaus as a figure representing the transition from the religious-
mythological lore and number symbolism of the time of Pythagoras
to Pythagorean science represented by the generation of Archytas.
Carl Huffman, who relied heavily on Burkert’s analysis, tried to move
further forward and reconstruct as fully as possible the philosophical
and scientific teaching of Philolaus, which survived in the authentic
fragments and testimonia.8 In the attempt to ‘rescue’ Philolaus, as a
philosopher and scientist, Huffman considered it possible to sacrifice
both Pythagoras and practically all Pythagoreans before Philolaus.
None of them, in his view, were scientists or philosophers, or if they
were, they failed to set down their ideas in writing. Naturally, Huff-
man came up against a serious problem in making Philolaus a bearer
of the Pythagorean tradition. He resolved it by presenting a Philolaus
who, while never completely ceasing to be a Pythagorean, becomes
increasingly ‘Presocratic’.9
The same approach may be seen in Huffman’s fundamental
new work on Archytas, based on a detailed analysis of all available
sources.10 Huffman is the first scholar to present such a full picture
of this outstanding Pythagorean scientist, philosopher and politician,
who exerted considerable influence on Plato. However, the Pythago-
reanism of Archytas explains hardly anything in his science
and philosophy; moreover, he himself needs to be explained, since it
emerges that he did less to continue Pythagoras’ line than to break with
it. Why did the contemporaries of Philolaus and Archytas consider
them Pythagoreans, and what did it mean to be a Pythagorean in the

8
C. A. Huffman, Philolaus of Croton: Pythagorean and Presocratic (Cambridge,
1993). According to Huffman’s analysis, testimonia A 7a, 9–10, 16–24, 27–9 are
authentic.
9
Cf. L. Zhmud, ‘Some Notes on Philolaus and the Pythagoreans’, Hyperboreus 4
(1998), 243–70.
10
C. A. Huffman, Archytas of Tarentum: Pythagorean, Philosopher and Mathema-
tician King (Cambridge, 2005).
 The Pythagorean Question: Problems, Methods, and Sources 5
fifth century? Is Pythagoreanism possible without Pythagoras, like
Orphism without Orpheus, of whose personality we have absolutely
no need; or is Pythagoreanism possible in spite of Pythagoras, like
some intellectual and spiritual movements which have evolved in
directions opposite to the designs of their founders? I do not believe
that much can be achieved by taking this path. While acknowledging
that Pythagoras is the most complex component of the Pythagorean
question, we should aim not to cast him aside, but to try and find the
links connecting him with the Pythagoreans of the sixth to fourth
centuries, and through them with the larger phenomenon of ancient
Greek Pythagoreanism. To divide the larger problem into a number of
smaller ones – Pythagoras, the Pythagorean school, the Pythagoreans,
Pythagoreanism – seems to me if not a guarantee of success at least a
step in the right direction.
What are the connections between these related but not fully
overlapping concepts? The Pythagorean school includes only those
Pythagoreans who left their mark in philosophy, science, and medi-
cine. ‘The Pythagoreans’ is a more general term, which also embraces
those who were members of the Pythagorean political societies (hetair-
iai) and/or representatives of the Pythagorean way of life. Both these
groups ceased to exist in the middle of the fourth century, and with
them went ancient Pythagoreanism. But Pythagoreanism as a whole,
as the totality of what was conveyed in antiquity (and often later) by
the name of Pythagoras, lived on after that, and with time assumed
new forms. Among its filiations were ‘Pythagorizing’ philosophers, for
example, Diodorus of Aspendus (second half of the fourth century),
who had nothing to do with the politics, philosophy, or science of
the Pythagoreans but merely led an ascetic way of life which had
become popular. Their reflection in Middle Comedy, the so-called
Pythagorists, often appeared on the Athenian stage after the mid-
fourth century as indigent preachers of metempsychosis and vegetari-
anism (DK 58 E).
Even clearer evidence of the end of ancient Pythagoreanism is
provided by the pseudo-Pythagorean writings which appeared at
the turn of the third century, signed with the names of Pythagoras
and historical or invented Pythagoreans.11 These texts, whose authors

11
For discussion of the date and place of the creation of the pseudo-Pythagorean
literature, see W. Burkert, ‘Hellenistische Pseudopythagorica’, Philologus 105 (1961),
16–43, 226–46; id., ‘Zur geistesgeschichtlichen Einordnung einiger Pseudopythagorica’,
6 Pythagoras and the Early Pythagoreans
are still unknown, were fabricated throughout the Hellenistic period
and the early Roman Empire without any discernible link with the
original writings of the Pythagoreans of the fifth and fourth centu-
ries.12 The ‘Pythagoreanism’ of pseudo-Pythagorean writing comes
down to a small number of very general notions which had a well-
known connection with Pythagoras and his school: arithmology,
cosmic and musical harmony, etc. As a rule the authors of these
works relied on Academic and Peripatetic interpretations of Pytha-
goreanism, or directly on the theories of Plato and Aristotle.
Published under the name of Pythagoras and his disciples, these
writings were evidently intended to demonstrate precisely whom
Greek philosophy had to thank for all that was best in it. The
abundance of this material stands in contrast to its almost complete
uselessness for any historical reconstruction of the teachings of
Pythagoras and the Pythagoreans.
The growing body of pseudo-Pythagorean literature, the revival of
dogmatic Platonism, and the undying fame of Pythagoras laid the
ground in the first century for the birth of Neo-Pythagoreanism.
Along with the pseudonymous tracts appeared the writings of those
who saw themselves as followers of the Platonized Pythagoras, but
wrote in their own names. With few exceptions, all known neo-
Pythagoreans of whom written works or fragments are preserved
were Platonists: Eudorus of Alexandria (second half of the first
century), Moderatus of Gades and Apollonius of Tyana (second half
of the first century AD), Nicomachus of Gerasa and Numenius
of Apameia (second century AD), and others.13 The final synthesis of
Neoplatonism and neo-Pythagoreanism was achieved by such

in K. von Fritz (ed.), Pseudepigrapha I (Geneva, 1971), 25–55; H. Thesleff, An Introduc-

tion to the Pythagorean Writings of the Hellenistic Period (Åbo, 1961); id., ‘On the
problem of the Doric pseudo-Pythagorica’, Pseudepigrapha I, 57–102; A. Städele, Die
Briefe des Pythagoras und der Pythagoreer (Meisenheim am Glan, 1980); B. Centrone, ‘La
letteratura pseudopitagorica: origine, diffusione e finalità’, in G. Cerri (ed.), La letteratura
pseudepigrafa nella cultura greca e romana (Naples, 2000), 429–52.
12
H. Dörrie, ‘Der nachklassische Pythagoreismus’, RE 24 (1963), 271.
13
See J. Dillon, The Middle Platonists, 2nd edn. (London, 1996). ‘To propound
Pythagorean views, or adopt Pythagorean practices, was not an alternative to being
Platonist: depending on one’s position in a long-running debate, Pythagoreanism was
Platonism properly interpreted or Platonism with optional extras’, G. Clark, ‘Philo-
sophic Lives and the Philosophic Life: Porphyry and Iamblichus’, in T. Hägg and
P. Rousseau (eds.), Greek Biography and Panegyric in Late Antiquity (Berkeley,
2000), 36.
 The Pythagorean Question: Problems, Methods, and Sources 7
important figures for the Pythagorean tradition as Porphyry of Tyre
(c. AD 235–c.305), the pupil of Plotinus, and especially Iamblichus
of Chalcis in Syria (c. AD 245–c.325), the pupil of Anatolius and later
Porphyry. Porphyry’s Life of Pythagoras and Iamblichus’ On the
Pythagorean Life, with Nicomachus’ popular introductions to mathe-
matics, became canonical texts which defined the picture of Pytha-
goras and his school right down to the nineteenth century.
Thanks to the collective efforts of many generations of admirers
and interpreters, Pythagoreanism was the only strain of Presocratic
thought to survive, albeit in much-modified form, until the end of
antiquity, and Pythagoras vied with Socrates and Plato (far outstrip-
ping their predecessors) in influence on thinkers of later eras.14 Here,
however, it would be more fitting to speak of ‘Pythagoras’, the author
of the pseudepigrapha, the hero of aretalogical biographies and fan-
tastic novels, than of the real historical figure of interest to us. These
two hypostases of Pythagoras bear roughly the same relationship to
each other as Alexander the Great and the protagonist of A Novel
about Alexander, except that Pythagoras did not have any Ptolemy or
even any Callisthenes of his own. This being so, any interpretation of
him as a person, of his teaching and his work, is far more problematic
than is the case with Philolaus, Archytas, and many other Pythagore-
ans. Although the central figure in pseudo-Pythagorean literature is
not Pythagoras but Archytas, and for every authentic fragment of
Philolaus and Archytas there are several (sometimes several dozen)
that are spurious, it remains possible to set against these late forgeries
some quite tangible evidence. In the case of Pythagoras’ teaching, the
basis of our reconstructions is much less sure. For this reason, even if
we confine ourselves to establishing the vast but elusive influence of
Pythagoras on Archaic and Classical culture, historians inevitably
draw on something which lies outside the limits of the sources:
their own general views on the rise of Greek philosophy and science,
and on the influence exerted on these by Greek religion and Oriental
knowledge. By superimposing on this historical background what is
known of Pythagorean religion, philosophy, and science in the sixth
to fourth centuries, they then try to determine the contribution of

14
On Pythagoreanism in the Middle Ages and the modern age, see Kahn, 157 ff.;
Ch. Riedweg, Pythagoras: His Life, Teaching, and Influence (Ithaca and London,
2005), 169 f.; Ch. Celenza, Piety and Pythagoras in Renaissance Florence: The Symbo-
lum Nesianum (Leiden, 2001).
8 Pythagoras and the Early Pythagoreans
Pythagoras himself. Given the varied nature of the initial premises, it
is clear that the results will rarely be conclusive.
If we turn to the sources, the difficulties are not limited to the usual
ones faced in studying the Presocratics, like interpreting philosophi-
cal and scientific views which are to be reconstructed on the basis of a
small number of fragments and indirect evidence. From Pythagoras
himself not so much as one line has reached us. Apparently he really
did not write anything.15 All that remains is the deep mark he left in
the ancient tradition, a mark which is very difficult to interpret. In the
literature of the fifth and fourth centuries Pythagoras already emerges
as an outstanding thinker and mathematician, a religious and ethical
reformer, a wise teacher, an influential politician, a demigod to his
disciples and a charlatan to some of his contemporaries, and the
founder of a scientific school that was at the same time a religious
brotherhood. These contradictory views, both ancient and modern,
may largely be attributable to his unique personality. In him, it seems,
are found almost all those departures from the ‘normal’ Presocratic as
an author of a philosophical work, which characterized other thinkers
of his era – Thales, Empedocles, Archytas.
Thales, an older contemporary of Pythagoras, also wrote nothing,
but here we are dealing with a problem of a quite different order. The
philosophy of Thales remains his alone, provided, of course, that we
overlook the fact that it was the first in Greece. No arguments
developed about him in the schools of Plato and Aristotle; we know
of no ‘Thalesians’ or ‘neo-Thalesians’, whereas the Pythagoreans and
neo-Pythagoreans are well known. Pythagoras had more followers
and disciples than any of the Presocratics even in his lifetime (and
many more after his death). By analysing various aspects of ancient
Pythagoreanism, we can establish the areas in which the influence of
Pythagoras was most enduring. At the same time, the wide variety of
individuals and forms which we see in Pythagoreanism inevitably
raises the question: do all elements of ancient Pythagoreanism owe
their existence to its founder? Some general considerations and his-
torical parallels suggest they do not, and there are still fewer grounds

15
Although the established view that Pythagoras wrote nothing developed only in
the age of Hellenism, it does not follow from this that he actually did set anything
down. Cf. Ch. Riedweg, ‘ “Pythagoras hinterliess keine einzige Schrift” – ein Irrtum?’,
Mus. Helv. 54 (1997), 65–92.
 The Pythagorean Question: Problems, Methods, and Sources 9
to argue for direct continuity with regard to Pythagoreanism after the
mid-fourth century.
As often happens, the decline of the Pythagorean school after
350 coincides with a veritable boom in philosophical and historical
literature about Pythagoras and the Pythagoreans, first in the Acad-
emy and the Lyceum, and later outside these. Even the Stoic Zeno
wrote his —ıŁÆªæØŒ (D.L. VII, 4). In the last third of the fourth
century at least four biographies of Pythagoras were written, and
with each century that passed their number increased, while
pseudo-Pythagorean writings multiplied at an even greater rate.16
The neo-Pythagorean biographers Nicomachus and Apollonius
selected from this large body of literature the material that best suited
their tastes and views, and combined it with Platonism and the
popular religious notions of their time. This trend was continued by
their Neoplatonic successors Porphyry and Iamblichus, who created
the image of Pythagoras the ‘divine sage’ at a time when the influence
of Christianity was already rapidly increasing. The biography found
in the collection of Diogenes Laertius is more sober, but like all the
other biographies of Pythagoras which have come down to us it is the
product of literary invention and of use only on those rare occasions
when it relies directly or indirectly on the scant trustworthy evidence
to be found in the fourth-century writers – Aristoxenus, Dicaearchus,
Neanthes of Cyzicus, Timaeus of Tauromenium, and others.
The critical investigations into the late tradition begun by Zeller
were carried further by Erwin Rohde, Armand Delatte, Isidore Lévy,
André-Jean Festugière, Kurt von Fritz, Walter Burkert, and others.17
Their research showed that material on ancient Pythagoreanism
dating back to fourth-century authors survived in texts from the
Hellenistic and Roman periods only in the form of occasional
brief passages. Unlike the search for secondary sources, attempts to

16
In all, six biographies of Pythagoras have come down to us. They range from a
few pages in the Suda to a whole treatise by Iamblichus. In this Pythagoras outstripped
even Plato and Aristotle.
17
Rohde, 102 ff.; J. Mewaldt, De Aristoxeni Pythagoricis sententiis et Vita Pytha-
gorica (diss. Berlin, 1904); W. Bertermann, De Iamblichi vitae Pythagoricae fontibus
(diss. Königsberg, 1913); Delatte, Lit.; id., Pol.; id., Vie; H. Jäger, Die Quellen des
Porphyrios in seiner Pythagoras-Biographie (diss.; Zurich, 1919); Lévy; A.-J. Festu-
gière, ‘Sur la “Vita Pythagorica” de Jamblique’ (1937), in his Études de philosophie
grecque (Paris, 1971), 437–62; von Fritz, Pol.; id., ‘Pythagoras’, RE 47 (1963), 171–203;
Burkert.
10 Pythagoras and the Early Pythagoreans
reconstruct authentic Pythagorean texts from the fifth and fourth
centuries brought no result. The idea that the Pythagorean Memoirs
transmitted by Alexander Polyhistor are a fourth-century source was
rebutted by Willy Theiler, and later by Festugière.18 Theiler showed
that the greater part of the Pythagorean texts examined by Delatte
(including the famous  Iæe ºª) were late forgeries.19 The inter-
pretation of Pythagoras’ speeches, found in Iamblichus, as a fifth-
century source has also been rejected,20 as has Corssen’s theory that
Androcydes’ book On Pythagorean Symbols was written by a fourth-
century doctor.21 Perhaps because of the absence of palpable success
in this area of Quellenforschung, in recent decades very few scholars
have ventured far into it. Most such attempts have proved to be only
one more rehearsal of previously rejected ideas,22 or an unwarranted
revision of well-established opinions.23 A tendency now widespread
in classical philology to suppose that many late authors who were
previously regarded as compilers were in fact not compilers24 has also

18
See M. Wellmann, ‘Eine pythagoreische Urkunde des 4. Jh. v.Chr.’, Hermes 54
(1919), 225–45; Delatte, Vie, 197 ff.; W. Wiersma, ‘Das Referat des Alexandros Poly-
histor über die Pythagoreische Philosophie’, Mnemosyne 10 (1941), 97–112. Cf.
W. Theiler, Review, Gnomon 2 (1926), 147–56; Lévy, 74 f.; A.-J. Festugière, ‘Les
“Mémoires pythagoriques” cités par Alexandre Polyhistor’, REG 58 (1945), 1–65.
Zeller, iii. 2, 103 ff., 108, dated the Memoirs in the late 2nd – early 1st cents.
19
See also K. von Fritz, ‘Pythagoreer’, RE 47 (1963), 239 ff.
20
See A. Rostagni, ‘Pitagora e i Pitagorici in Timeo’ (1914), in his Scritti minori,
ii/1 (Turin, 1956), 3–50; id., ‘Un nuovo capitolo nella storia della retorica e della
sofistica’ (1922), in Scritti minori, i (Turin, 1955), 3–59; Delatte, Lit., 85 f; id., Pol., 39 f.;
cf. Burkert, 104 n. 37.
21
See P. Corssen, ‘Die Schrift des Arztes Androkydes —æd ıŁÆªæØŒH ı -
ºø ’, RhM 67 (1912), 240–62. Cf. Burkert, 167 n. 9. On Androcydes see below,
171, 192.
22
De Vogel, 70 ff., argued that the speeches of Pythagoras reported by Iamblichus
are an early source. Cf. Burkert, 104 n. 37; id., Review, Gymnasium 74 (1967), 458–60;
M. Zucconi, ‘La tradizione dei discorsi di Pitagora in Giamblico’, Vita Pythagorica,
37–57, RFIC 98 (1970), 491–501. The suggestion of J. C. Thom, The Pythagorean
Golden Verses (Leiden, 1995), that the pseudo-Pythagorean Golden Verses date from
the 4th century, is even less convincing.
23
J. Philip, ‘The Biographical Tradition – Pythagoras’, TAPA 90 (1959), 185 and
P. Gorman, ‘The “Apollonios” of the Neoplatonic Biographies of Pythagoras’, Mne-
mosyne 38 (1985), 130–44, rejected the idea that Nicomachus and Apollonius
wrote biographies of Pythagoras.
24
This tendency has shown itself e.g. in many recent works on Iamblichus:
D. O’Meara, Pythagoras Revived: Mathematics and Philosophy in Late Antiquity
(Oxford, 1989); G. Staab, Pythagoras in der Spätantike: Studien zu De Vita Pythago-
rica des Iamblichos von Chalkis (Munich, 2002); M. Lurje, ‘Die Vita Pythagorica als
Manifest der neuplatonischen Paideia’, in M. von Albrecht et al. (eds.), Jamblich.
 The Pythagorean Question: Problems, Methods, and Sources 11
done nothing to foster the development of Quellenforschung, which is
increasingly associated with something hopelessly outmoded.25
Nevertheless, whatever the extent of originality of the late antique
writers and of our faith in nineteenth-century philology, it seems
obvious that we have not yet exhausted our chances of finding in
the ancient tradition evidence of the classical period on the Pythago-
reans, even if our searches are not always successful.26
An uncritical adherence to the late tradition – still quite often met
with – poses artificial barriers along the path of research. Is it possible
to reconstruct the philosophy and science of the early Pythagoreans if
their community was dominated by the absolute authority of the
Master and all doubts were swept aside with ‘Ipse dixit’, if the teach-
ing before the time of Philolaus was oral and secret, and all achieve-
ments were attributed to Pythagoras?27 But even if we cast aside these
and similar late legends, it is possible to come to a dead end by placing
excessive faith in the authorities of the classical era. By restricting our
research into early Pythagoreanism to the evidence of writers before
300, we are taking only the first step in our search for reliable sources.
The legendary tradition about Pythagoras, which reaches back to his
lifetime, evolved in accordance with the laws of this genre of folklore,
constantly incorporating similar elements (miracles, prophecies and
the like), and sometimes losing any tangible connection with the
person who gave birth to it. In any case, there are no grounds
whatever for giving the legendary tradition privileged status because
of its supposed ‘archaic’ character: virtually every source of a legend
about Pythagoras has introduced something new. In a similar way
a legendary tradition developed about the Pythagoreans, which is
reflected in a book by the Milesian Sophist Anaximander the
Younger, the Interpretation of Pythagorean Symbols, as well as in
Aristotle, and later in a long series of new interpretations of ‘Pytha-
gorean symbols’ (58 C 1–6).

Pythagoras: Legende – Lehre – Lebensgestaltung (Darmstadt, 2002), 221–52. It is

revealing that neither O’Meara nor Staab (Pythagoras, 222) analyse Iamblichus’
sources.
25
J. Mansfeld, Studies in the Historiography of Greek Philosophy (Assen, 1990), 345.
26
Burkert, who had asserted earlier that the story of the mathematici and acusmatici,
found in Iamblichus, derives from Aristotle (192 ff.), was forced to admit that this cannot
be proved: id., ‘Pythagoreische Retraktationen’, in W. Burkert et al. (eds.), Fragment-
sammlungen philosophischer Texte der Antike (Göttingen, 1998), 314 f.
27
D.L. VIII, 15, 42, 46; Porph. VP 57–8; Iamb. VP 158, 198, 199, 226–7, 246–7.
12 Pythagoras and the Early Pythagoreans
Like no one else among the Presocratics, Pythagoras very soon
became the subject of constant disputes. Beginning with Xenophanes
and Heraclitus, almost all of the early tradition is polemical. That is
why it has far more to say about Pythagoras than about any other
thinker of his day. The first book about the Greek philosopher was
Democritus’ Pythagoras (A 33,1). When in the Academy the earliest
genre of philosophical historiography arose – a monograph devoted
to an individual thinker or school of thought – among the very first
specimens were —ıŁÆªæØÆ by Xenocrates (fr. 2) and On the Pytha-
goreans by Heraclides of Pontus (fr. 22, 40–1). In addition, Pytha-
goras became the hero of Heraclides’ fantastic dialogues Abaris (fr.
73–5, cf. fr. 90) and On the Woman who Stopped Breathing (fr. 87–9),
which drew on the legends and elaborated on them. Speusippus wrote
a book which was to become highly influential, On Pythagorean
Numbers (fr. 28). It is commonly accepted now that the Platonists
presented Pythagoras and the Pythagoreans as the precursors of Plato
and ascribed Academic notions to them. Whether this is the case will
be examined later (}12.1). At any rate, the Platonists were sympa-
thetic towards Pythagoras and the Pythagoreans. In contrast to
them Isocrates, the rival of the Academy, was rather ironic about
Pythagoras’ çغ ç Æ (Bus. 28), and his student, the historian Theo-
pompus, expressed great hostility to Pythagoras’ philosophy (FGrHist
115 F 72).
Discussing the theories of the Presocratics in De caelo, Aristotle
makes an interesting psychological remark: ‘It is what we are all
inclined to do, to direct our enquiry not to the matter itself, but to
the views of our opponents’ (294b5). Besides his works on individual
Pythagoreans or against them (On Archytas’ Philosophy, Against
Alcmaeon), Aristotle wrote two special monographs: On the Pytha-
goreans (fr. 191–6), containing a collection of mostly legendary
material, and Against the Pythagoreans (fr. 198–205), in which he
criticized their philosophical and scientific theories.28 He took issue
with the Pythagoreans in his Physics, De caelo, Metaphysics, and other

28
In the catalogue of Aristotle’s writings (D.L. V, 22–7), dating back to the 3rd
cent., both works are listed (nos. 97, 101), each one being the size of a book. In about
the 2nd cent. the two books were combined into a single book, which later writers
cited using different titles. See P. Moraux, Les Listes anciennes des ouvrages d’Aristote
(Louvain, 1951), 107; J. Philip, ‘Aristotle’s Monograph on the Pythagoreans’, TAPA 94
(1963), 185 ff.; Burkert, 29; O. Gigon (ed.), Aristotelis opera, iii. Librorum deperdi-
torum fragmenta (Berlin, 1987), 408 f.
 The Pythagorean Question: Problems, Methods, and Sources 13
works, but quite often his true opponents were not so much the actual
representatives of that school as his colleagues at the Academy. This
factor left a deep imprint on his interpretation of Pythagoreanism and
doomed to failure any attempt to view Aristotle uncritically as the
most reliable source on Pythagorean views.29 The famous ‘number
doctrine’,30 passed on by Aristotle, still considered to be the quintes-
sence of Pythagorean philosophy, cannot be found in the early
Pythagoreans or in Philolaus, and it has proved impossible to trace
it back to Pythagoras. Aristotle’s interpretation of Pythagorean num-
ber philosophy can be understood only in the context of Plato’s
unwritten doctrine and the theories of the Academics that were
built on it. And while Charles Kahn, the author of a recent mono-
graph on Pythagoras and the Pythagoreans, believes that ‘Aristotle
was the last author to draw a clear distinction between the two
schools’,31 there are good grounds to wonder whether Aristotle really
made a strict distinction between Pythagoreanism and Platonism,
and whether he did not ascribe to the Pythagoreans ideas which
were alien to them.
In the next generation, Theophrastus at least once (Met.
11a27–b10) linked ‘the Pythagoreans’ with the principles of Plato’s
unwritten doctrine (in late Hellenism, this became a defining char-
acteristic of a pseudo- and neo-Pythagorean tradition). Fortunately
this tendency does not affect the individual Pythagoreans who figure
in his doxographical compendium the Opinions of the Physicists.
Pythagoras and anonymous Pythagoreans, the bearers of the number
doctrine, are not included there. The writings about Pythagoras and
his followers by the Peripatetic Aristoxenus, who in his youth studied
under the last of the Pythagoreans, paint an idealized portrait of
philosophers, scientists, and politicians living in harmony with their
ethical principles. This picture differs from the one found in his
biographies of Socrates and Plato, which abound in the most varied
and scandalous accusations. In spite of the unconcealed personal
sympathies and antipathies of Aristoxenus, the teachings which he

29
One such attempt was by Philip, 5 f. Cherniss’s criticism of Aristotle is generally
excessive, but mostly accurate with regard to the Pythagorean material (Cherniss,
Criticism); cf. below, }}11.2, 12.2.
30
Hereafter I will apply the name ‘number doctrine’ (number philosophy) to that
form of the philosophy of number according to which the world arose out of numbers
and consists of numbers, i.e., number is an ontological principle.
31
Kahn, 65. Cf. Zhmud, ‘Some Notes’, 259 ff.
14 Pythagoras and the Early Pythagoreans
passed on in his Pythagorean Precepts are suspiciously like those
of the Academy. Dicaearchus, his less biased colleague from the
Lyceum, also makes Pythagoras, Plato, and Socrates the heroes of
his philosophical biographies. In the works of the Peripatetic Eude-
mus on the history of mathematics and astronomy we find a picture
which is rather different from Dicaearchus’, although it does not
contradict it.
Thus, if we confine ourselves to sources from before 300, we find in
them the same basic hypostases of Pythagoras as are discussed in
contemporary scholarship: a religious teacher, a politician, a philoso-
pher, and a scientist. Although the proportions of these different
aspects vary in the works of modern scholars, and depend on their
personal tastes, as a rough approximation the differing views may be
reduced to two main trends (within which, however, there are sub-
stantial variations). The first accepts in essence the ancient tradition
of the philosophical and scientific work of Pythagoras and his im-
mediate students.32 The second trend, much more critical of Pytha-
goras, emerges in the early twentieth century,33 and is most marked in
Burkert, whose book gave it broad currency.34 In that view, the early
(pre-Platonic) tradition contains no evidence of the scientific or
philosophical work of Pythagoras and his closest followers, and the
evidence which appeared later is merely a projection of the work of
the later Pythagoreans – Philolaus, Archytas, and their students – into
the past. Thus Pythagoras appears principally as a religious teacher
(a ‘shaman’ to Burkert, a ‘guru’ to Riedweg), preaching a doctrine
close to Orphism, on the transmigration of souls, and founding
a secret sect in which his followers led a life ruled by stringent and
absurd taboos.

32
Guthrie, i. 146–359; K. von Fritz, ‘Pythagoras’, RE 47 (1963), 171–203; id.,
‘Pythagoreer’; de Vogel; van der Waerden; Kahn.
33
See e.g. H. Vogt, ‘Die Geometrie des Pythagoras’, Bibl. Mathematica 9 (1909),
15–54; K. Reinhardt, Parmenides und die Geschichte der griechischen Philosophie
(Bonn, 1916), 231 f.; E. Sachs, Die fünf platonischen Körper (Berlin, 1917); Frank;
Lévy, 6; W. Rathmann, Quaestiones Pythagoreae Orphicae Empedocleae (diss.; Halle,
1933), 23 ff.; W. A. Heidel, ‘The Pythagoreans and Greek Mathematics’, AJP 61
(1940), 1–33; O. Gigon, Der Ursprung der griechischen Philosophie (Basel, 1945),
142 ff.
34
See e.g. Knorr, 5 ff.; J. Barnes, The Presocratic Philosophers, 2nd edn. (New York,
1982), 78 ff.; Huffman, Philolaus; P. Kingsley, Ancient Philosophy, Mystery, and
Magic: Empedocles and Pythagorean Tradition (Oxford, 1995); M. Giangiulio, Pita-
gora: le opere e le testimonianze, i–ii (Milan, 2000); Riedweg, Pythagoras.
 The Pythagorean Question: Problems, Methods, and Sources 15
It is hardly possible to unite harmoniously all versions of the
classical sources, because the reality was no less contradictory than
the tradition. Should we therefore sacrifice one of the aspects of early
Pythagoreanism – the scientific, the philosophical, the political, or the
religious – as the price of greater inner unity in our reconstruction?
‘A minimalism that eliminates every aspect of tradition which seems
in any respect questionable cannot help giving a false picture.’35
Burkert in his book twice demonstrated the justice of this assertion:
first, by rehabilitating a somewhat questionable tradition about Phi-
lolaus, and then by trying to eliminate all evidence of scientific and
philosophical work by Pythagoras, as well as his closeness to the
Ionian ƒ æ Æ.
It is quite natural that science should usually be the first casualty of
a selective approach, especially when we consider how eagerly many
twentieth-century scholars tried to cast aside the heritage of ‘positi-
vism’ (variously understood) and the ‘modernization’ of Archaic
Greece.36 Unfortunately, what took the place of ‘modernization’ was
frequently not an unbiased approach, but artificial archaizing of the
Presocratics: contrary to Aristotle’s judgement, they were brought
together with the theologians and wonder-workers of the Archaic
period;37 the sources of their thinking were sought in Oriental
mythology and theogony, or even in the religion of hunting tribes
(shamanism) and the like. Paradoxical though it may seem, this
archaizing tendency is to a large extent linked with the fact that
attempts were made to judge the Presocratics by the standards of
the age of positivism and, if the material did not match the accepted
image of a rationalist and scientist, the scientific component of the
Archaic culture was rejected outright. In the meantime, what in the
sixth and fifth centuries was coming to be science was not yet science
in full measure, and the image of a scientist as we know it was only
beginning to take shape from the variety of human material. Not only
do Thales the politician, Xenophanes the poet, Pythagoras the

35
Burkert, 10.
36
While Rohde (103 f.) regarded Pythagoras not as a philosopher but as a religious
reformer and a scientist, in Frank’s view (67) ‘old Pythagoras’ could in a certain sense
be termed a philosopher, but he had very little to do with science. Philip (24 ff., 200 ff.)
likewise allows that the early Pythagoreans were philosophers, while rejecting their
claim to science.
37
On Aristotle’s distinctions between ‘physicists’ and ‘theologians’, see L. Zhmud,
The Origin of the History of Science in Classical Antiquity (Berlin, 2006), 130 f.
16 Pythagoras and the Early Pythagoreans
religious teacher, and Archytas the army commander not resemble
university professors, they do not resemble specialized Hellenistic
scientists either. However, the variety of human individuality, formed
in a particular cultural situation, and the uniqueness of the very
period of the ‘Greek miracle’,38 should not discourage those who
see that the methods by which Thales, Pythagoras, and Archytas
solved scientific problems are the same as those of mature science.
The real alternative to modernization and archaization lies in
acknowledging that each sphere of activity of Pythagoras and the
Pythagoreans – politics, religion, philosophy, and science – has its
own internal logic and independent history, and each must therefore
receive its own specific explanation. Each explanation must be com-
patible with all the others, but not reducible to a single unified all-
embracing construction, such as that offered by F.M. Cornford and
J. Burnet – Pythagorean science as ‘purification of the soul’.39 The
Platonic origin of this notion prevented it from taking hold as a
convincing explanation for Pythagoreanism. Another construction,
which arose in late Antiquity, proved to be far more enduring. To
resolve the contradictions in the tradition surrounding Pythagoras
and his disciples, two different trends, or degrees of initiation, which
supposedly existed among the early Pythagoreans, were invented: the
scientific mathematici, and the religious acusmatici, who engaged in
political life. The dispute between them, described by Iamblichus
(Comm. Math. 76.16 ff.), in some ways recalls the dispute between
modern scholars about science and religion in the Pythagorean
school. While the acusmatici refused to recognize the mathematici
as fellow Pythagoreans, the mathematici did not deny the Pythagore-
anism of the acusmatici, but claimed to be following Pythagoras even
more closely.
Many modern studies add to this construction the thesis of a
gradual rationalization of Pythagoreanism, which would explain the
development of myth and number symbolism into philosophy and

38
For the first attempt at a systematic study of the ‘Greek miracle’, applying
sociological and socio-psychological methods, see Zaicev. Unfortunately, responses
to this book have been few, partly because the ‘Greek miracle’ itself is losing its
erstwhile appeal.
39
F. M. Cornford, ‘Mysticism and Science in the Pythagorean Tradition’, CQ 16
(1922), 137–50; 17 (1923), 1–12; Burnet, 97 f.; cf. the critique by Burkert, 211 f.
 The Pythagorean Question: Problems, Methods, and Sources 17
science. Leaving aside the fact that this development is itself extreme-
ly questionable, the thesis of rationalization as a form of linear
evolutionary progression ‘from myth to reason’40 runs counter both
to the steady increase in mythical elements in the tradition about
Pythagoras and the Pythagoreans, and to the real history of the
Pythagorean society. Rationalization does nothing to explain Pytha-
goras himself, or such figures in his entourage as the athlete and army
commander Milon, the mathematician Hippasus, the doctor Demo-
cedes, and the natural philosopher Alcmaeon. Since the political
dominance of the Pythagoreans, first in Croton, then in other cities,
is an indisputable fact of the history of Magna Graecia in the years
c.510–450, the Pythagorean society should be regarded above all
within the context of that history.41 Those who suppose that it was
originally a sect of superstitious ritualists42 need to explain how this
sect came to lead the southern Italian aristocracy and hold on to
power for more than half a century, making Croton famous for its
unprecedented number of victors at the Olympic Games and other
sporting competitions.43
It is revealing that the rationalization of Pythagoreanism is dated in
widely varying periods. To Erich Frank, Aristotle’s ‘so-called Pytha-
goreans’, whom he identified with Archytas and his pupils, were
responsible for all the achievements attributed to the early school.44
Burkert, who accepted Frank’s idea that early Pythagorean science
and philosophy were a retrospective Academic projection, took him
to task for exaggeration and dated the beginning of the Pythagoreans’
philosophical and scientific work to the mid-fifth century. The well-
known historian of science B. L. van der Waerden was prepared to go

40
R. Buxton (ed.), From Myth to Reason? Studies in the Development of Greek
Thought (Oxford, 2001).
41
See the classical studies of Pythagorean politics: Delatte, Pol.; von Fritz, Pol.; Minar;
Dunbabin, and the following more recent works: M. Giangiulio, Ricerche su Crotone
arcaica (Pisa, 1989); M. Bugno, Da Sibari a Thurii: La fine di un impero (Naples, 1999).
42
W. Burkert, ‘Craft Versus Sect: The Problem of Orphics and Pythagoreans’, in
B. F. Meyer (ed.), Jewish and Christian Self-Definition, iii (London, 1983), 1–22.
43
On the Pythagorean athletes, see Giangiulio, Ricerche 102 f., 291 f.; C. Mann,
Athlet und Polis im archaischen und frühklassischen Griechenland (Göttingen, 2001),
164 ff.
44
Relying on Aristotle’s phrase ƒ ŒÆº  Ø —ıŁÆªæØØ, Frank, 69, argued that
these Pythagoreans were not true followers of Pythagoras. Cf. Cherniss, Criticism, 348;
Guthrie, i. 155; J. Philip, ‘Aristotle’s sources for Pythagorean doctrine’, Phoenix 17
(1963), 252 f.; Burkert, 29 ff.; Huffman, Philolaus, 31 f.
18 Pythagoras and the Early Pythagoreans
even further in the rehabilitation of the early school. He recognized its
great successes in mathematics, astronomy, and physics, but refused
to attribute them to Pythagoras himself, whom he depicted as merely
an apt pupil of the Egyptians and Babylonians, who related to his
followers the essence of the knowledge he had received.45 Finally
Riedweg, who generally shared Burkert’s approach, acknowledges
such an important element of the tradition as the coining by Pythagoras
of the word çغ ç,46 and even Pythagorean science: though ‘a
speculative theory of numbers with certain mythical traits’, it is never-
theless a science, at least in the sense of Lévi-Strauss’s pensée sauvage.47
Unlike pensée sauvage and other twentieth-century anthropologi-
cal constructs, such as ‘Greek shamanism’ or ‘mythical thinking’,48
the science and philosophy of the sixth century, represented by the
likes of Thales, Anaximander, Anaximenes, and Xenophanes, are
absolutely real. Our task is to determine to what extent Pythagoras
and his earlier followers were mediators between the sixth-century
Ionians, on the one hand, and the science and philosophy of the
second half of the fifth century (Pythagorean or not), on the other. On
the whole we have to admit that Pythagoras’ contribution to philo-
sophy and science may be reconstructed with varying degrees of
accuracy. His students did not set down his views, unlike those
of Socrates, for example. The Pythagorean sources make no mention
of Pythagoras at all. The early tradition presents him as a ‘wise man’,
but how his wisdom showed itself we learn principally from the
evidence of the post-Platonic time. As a result, we have no direct,
reliable access to his philosophical teachings. In order to achieve even
a relative degree of reliability, it is necessary to compare the ideas of
the early Pythagoreans, from whom we have some fragments and
testimonia, with those fifth- and fourth-century sources on Pytha-
goras which have withstood a preliminary historico-philological scru-
tiny. The ideas, which do not fundamentally contradict attested early
Pythagorean views, define the limits of the possible. In order to

45
Van der Waerden, 14 f.
46
Ch. Riedweg, ‘Zum Ursprung des Wortes “Philosophie” ’, in A. Bierl et al. (eds.),
Antike Literatur in neuerer Deutung (Munich, 2004), 147–81. Cf. W. Burkert, ‘Platon
oder Pythagoras? Zum Ursprung des Wortes “Philosophie” ’, Hermes 88 (1960),
159–77.
47
Riedweg, Pythagoras, 90. Cf. L. Zhmud, Review, AncPhil 23 (2003), 416–20.
48
L. Zhmud, ‘On the Notion of “Mythical Thinking” ’, Hyperboreus 1 (1994/5),
155–69.
 The Pythagorean Question: Problems, Methods, and Sources 19
establish what within those limits is most reliable, we need to apply
additional criteria, for example, the fact that Pythagoras’ philosophy
should be post-Milesian and pre-Eleatic.
The path proposed here is far from straightforward. The teachings
of the early Pythagoreans which have come down to us – of Alc-
maeon, Hippasus, Menestor, Hippon, and others – are too highly
individual to be seen as a reflection of Pythagoras’ own system. In
fact, there is no certainty at all that such a system actually existed.
Furthermore, the search for traces of influence of Pythagoras’ philo-
sophy on his younger contemporaries Parmenides and Heraclitus,
and later on Anaxagoras, Empedocles, and Zeno has so far yielded
few decisive results. Yet it is difficult to see any serious alternative to
this approach. Any reconstruction of the philosophical views of
Pythagoras must be founded primarily on the views of his students
and followers, and use the teachings of the philosophers of the sixth
century and early fifth century for purposes of verification.
The most important premise for such an approach is continuity in
the evolution of Pythagorean theories. Since continuity is most
plainly visible in the exact sciences, by virtue of their cumulative
development, the reconstruction of Pythagoras’ achievements here,
or at least of the range of problems he dealt with, will be most reliable.
Here it is possible to establish the individual links in the chain of
scientific discoveries linking Pythagoras with Ionian geometry and
astronomy (Thales, Anaximander) on the one hand, and on the other
with Pythagorean mathematics (Hippasus, Theodorus, Archytas). We
may note that such a reconstruction is not only possible, but essential.
While the Eleatics, Heraclitus and Empedocles may be described – if
with some reservations – as thinkers who owe little or nothing to the
influence of Pythagoras the philosopher, or, in the case of Empedo-
cles, owe only part of their religious teaching to him, the geometry,
astronomy, and especially arithmetic and harmonics of the mid-fifth
century are left hanging in the air if we exclude Pythagoras and the
early Pythagoreans from the ranks of those who contributed to their
development. The question of their mathematical work is thus in-
separable from another question of equal importance: who created
the Greek geometry and astronomy which was taken up and devel-
oped by Oenopides and Hippocrates of Chios? A comprehensive
picture of the development of mathēmata will never be achieved,
20 Pythagoras and the Early Pythagoreans
but even a partial reconstruction is unquestionably preferable to
simply renouncing this line of investigation.49
Strange as it may seem, the thesis of the continuity of Pythagore-
anism from the sixth to the fifth and fourth centuries applies least in
the area of religion.50 The tradition contains no evidence of any
religious teaching or practice by the Pythagoreans known to us. We
simply know next to nothing about what Milon, Brontinus, Demo-
cedes, Alcmaeon, Hippasus, Iccus, Menestor, Hippon, Theodorus,
Philolaus, Lysis, Eurytus, Echecrates, Xenophilus, Archytas, Hicetas,
Ecphantus, and other historically attested Pythagoreans believed in,
what they worshipped, or how. In particular, we do not know whether
any of them shared even Pythagoras’ best-known and reliably attested
religious doctrine, metempsychosis, or practised the vegetarianism
which was associated with it.51 The dearth of information here is
largely due to the fragmentary and selective nature of our sources. By
itself, however, this cannot account for the fact that in the area of
religion we are compelled to rely almost exclusively on sources out-
side the Pythagorean school which deal with either Pythagoras or
anonymous Pythagoreans, ƒ —ıŁÆªæØØ. It must be admitted that,
with regard to religious matters, neither the writings of the Pythagor-
eans known to us by name nor their way of life offered anything of
interest to the ancient doxographical and biographical traditions.
The dual nature of the figure of Pythagoras himself was noted
by his contemporaries (see Heraclitus’ ƒ æ Æ and ŒÆŒå Æ, B
129) and attested by Aristotle: ‘Pythagoras, the son of Mnesarchus,
first dedicated himself to the study of mathematical sciences, espe-
cially numbers, but later could not refrain from the wonder-working
of Pherecydes’ (fr. 191). This combination of the rational and the

49
Cf. ‘In the absence of earlier documentation, the history of Pythagoreanism
before Philolaus, like the history of Greek mathematics before Hippocrates of Chios,
must remain an area for informed speculation’ (Kahn, p. ix).
50
Continuity of the Pythagoreans’ oligarchic policies in Magna Graecia was
interrupted by the anti-Pythagorean uprising in the mid-5th century, when, according
to Polybius (II,39,1; from Timaeus), ‘the best men in every polis’ perished. Later the
Pythagoreans managed not only to adapt to democracy, but to produce a political
leader as outstanding as Archytas.
51
Huffman’s point of departure is the premise that metempsychosis was ‘tacitly’
shared by all Pythagorean philosophers: C. A. Huffman, ‘The Pythagorean Concep-
tion of the Soul from Pythagoras to Philolaus’, in D. Frede and B. Reis (eds.), Body
and Soul in Ancient Philosophy (Berlin, 2009), 21–44. There is, however, no evidence
for this.
 The Pythagorean Question: Problems, Methods, and Sources 21
religious is not unique among the Presocratics: the natural philoso-
pher Empedocles pretended to be a wonder-worker and was a pro-
ponent of metempsychosis. The modern age affords numerous
examples of successful combinations of scientific thought and an
interest in astrology, alchemy, hermetism, magic, Cabbalism, and
other occult and mystic trends.52 Pythagoras claimed to possess
supernatural powers and was the kind of personality who attracted
legends, even legends which had originally referred to other, less
famous wonder-workers. But, unlike Pythagoras, none of the ancient
Pythagoreans known to us is linked – in the reliable part of the
tradition – with anything remotely supernatural or miraculous. This
is one of our most serious problems. The difference between the
Pythagoreans and Pythagoras is striking, and gives rise to some
obvious questions: were they really his disciples and followers, and
why do we not find among them a single religious figure with even a
distant resemblance to Pythagoras? If these people were Pythagoreans
and no others can be found, this should tell us much about Pytha-
goras himself and the society he founded. In this case, Pythagoras, by
combining too much variety within himself, may turn out to be an
exception among the Pythagoreans, who adopted only that part of his
legacy which corresponded to their own inclinations and interests.
Later this happened with Aristotle and the Peripatetics.
Not even the names of the individuals who followed Pythagoras’
religious teaching and the prescriptions and taboos associated with it
are known to us. The superstitious ritualists who avoided walking
along main roads, bathing in public baths, talking in the dark, step-
ping over yokes, and using knives to stoke fires, always turn out to be
anonymous figures from the legendary, not the historical tradition –
unlike the Pythagorean politicians, athletes, doctors, philosophers,
and mathematicians. Whatever the case, that anonymity, which is
also a feature of Orphism, does not give cause to doubt that Pytha-
goras had followers who valued his religious teaching above all. The
aim here is not to reduce the religious aspect of Pythagoreanism to
a minimum, but rather to establish its real dimensions, with a his-
torically viable way of life and a belief system for its followers.

52
C. Webster, From Paracelsus to Newton: Magic and the Making of Modern
Science (Cambridge, 1982); B. Vickers (ed.), Occult and Scientific Mentalities in the
Renaissance (Cambridge, 1986); J.-F. Bergier (ed.), Zwischen Wahn, Glaube und
Wissenschaft: Magie, Astrologie, Alchemie und Wissenschaftsgeschichte (Zurich, 1988).
22 Pythagoras and the Early Pythagoreans
In appealing to the legendary tradition, we must recognize clearly that
we are dealing with religious folklore, not with the realities of the
Pythagorean way of life which Plato mentioned with such respect
(Res. 600a–b).
The intention here, then, is to build up an individual portrait
of Pythagoras against the background of a collective portrait of the
Pythagoreans, allowing the two portraits to complement and correct
each other. The collective portrait itself should consist of individual
portraits of particular historical figures, and not be a collage of
assorted features allegedly common to the Pythagoreans ‘as a whole’.
This approach has its problems, not least because scholarship has not
yet established who is to be deemed a Pythagorean or by what criteria.
In spite of some disagreements though, the individual Pythagoreans
known to us by name constitute a fully tangible group, as distinct from
the anonymous Pythagoreans seen as standardized bearers of a gen-
eralized ‘school doctrine’. The collective portrait of the anonymous
Pythagoreans, such as is found, for example, in Aristotle’s writings,
or in modern works on the history of philosophy,53 is inevitably
anachronistic. Unlike the Academy, Garden, and Stoa, which were
institutionalized philosophical schools with a range of well-defined
doctrines, varied though these were at different times, the Pythagorean
school arose not as a philosophical school, but as a political society, a
hetairia. The teachings of its founder were not set down in writing, and
the school itself, widely scattered in many cities, evolved in the course
of almost two centuries. It is no surprise that in reliable sources we can
find nothing resembling a Pythagorean orthodoxy. All Pythagoreans
were different, although all shared common features with other Pytha-
goreans. Orthodoxy appears only in the late pseudo-Pythagorean
literature, but this is founded not on the authentic Pythagorean tradi-
tion, but on Platonism and/or Aristotelism.
In the modern world, as in antiquity, the stories which are passed
down of certain social, ethnic or religious minorities often differ from
those told about the individuals who make up those minorities. While
the former are by no means bound to be unreliable, or the latter
truthful, the two should be carefully distinguished. If we were to
collect all the information on individual Pythagoreans and compare
it with what the fifth- to fourth-century writers tell us about the

53
See e.g. Guthrie i., 146 f.
 The Pythagorean Question: Problems, Methods, and Sources 23
—ıŁÆªæØØ as a kind of collective entity, these portraits would be
substantially different. Sometimes they differ even within the work of
a single author: the teachings of individual Pythagoreans, conveyed
by Aristotle, do not remotely resemble the number doctrine which he
attributes to the Pythagoreans as a group. Of course, —ıŁÆªæØØ is
often no more than a façon de parler, concealing real and identifiable
people, like Archytas, for example, who is present behind the Pytha-
goreans in Plato’s Republic (530a–531c), or Philolaus, whose astro-
nomical theory Aristotle ascribes to some anonymous Pythagoreans
(Cael. 293a18 ff.). But frequently it is impossible to identify the
collective Pythagoreans with any of the historical figures or groups
of Pythagoreans that we know. If the doctrines or actions attributed to
the collective Pythagoreans are not confirmed at the level of indivi-
duals, and especially if they run directly counter to that part of the
tradition, such evidence needs to be regarded with a high degree of
scepticism.
Notwithstanding the indisputable service rendered by Aristotle in
creating the history of philosophy,54 it must be admitted that the
history of Pythagorean philosophy proved to be beyond his capabil-
ities. If we follow in his footsteps and deduce the philosophy of the
Pythagoreans from their work in mathematics,55 we risk overlooking
clear evidence that the source of the philosophical views of the early
Pythagoreans lay not so much in mathematics as in natural sciences
and medicine, which were closely interconnected. The first Pythagor-
ean whose philosophy shows traces of the influence of mathematics is
Philolaus. Before him, the influence of Pythagorean mathematics was
visible in the teachings of the Eleatics and Heraclitus,56 although this
fact is the subject of constant dispute, like almost everything else that
may serve to confirm Pythagoras’ role as a mediator between Ionian
and Italian science and philosophy.
Here we cannot fail to perceive one of the paradoxes of Pythagore-
an studies: those who deny that Pythagoras was one of the mediators
between the thought of the Greek East and West are apt to see him

54
L. Zhmud, ‘Doxographische Tradition’, in H. Flashar and G. Rechenauer (eds.),
Grundriss der Geschichte der Philosophie. Philosophie der Antike, i. Vorsokratiker
(Basel, 2012), 150–74.
55
‘The Pythagoreans, as they are called, devoted themselves to mathematics; they
were the first to advance this study, and having been brought up in it they thought its
principles were the principles of all things’ (Arist. Met. 985b23 f.).
56
On Parmenides see below, 251 ff.; on Heraclitus, see below, 35 n. 29.
24 Pythagoras and the Early Pythagoreans
rather as a mediator between Orient and Occident as a whole, as a
cultural hero who united Egyptian and Babylonian mathematics with
Indian metempsychosis and Scythian shamanistic rites.57 The ancient
image of him as the interlocutor of Zarathustra and pupil of the
Chaldeans, Brahmans and Druids seems to be self-reproducing and
therefore ineradicable. Remaining a riddle in his own right, Pytha-
goras has served for two and half millennia as the key to everything
that those who write about him would like to resolve and explain.
If this study can do anything to supplant this image with the real
historical (and therefore contradictory) figure, it may be deemed to
have succeeded.

57
W. Burkert, Greek Religion (Cambridge, Mass., 1985), 445.